201 9

Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against 18 Stress and Antibiotics

Anwar Alam, Ashutosh Kumar, Prajna Tripathi, Nasreen Z. Ehtesham, and Seyed E. Hasnain

Abstract Biofilm, as a heterogenous congregation of microbial cells enclosed within a pellicle, has largely gained attention due to their historical importance in environ- ment as sludges, flocs, slimes, etc. Biofilms in medical research have been an active area of research in periodontics, in wounds, and in surgical implants. With the availability of whole genome sequences, it is now evident that the mechanisms that control biofilm formation have largely remained conserved during the course of evolution, pointing to the fact that biofilm formation is an

Anwar Alam, Ashutosh Kumar and Prajna Tripathi have contributed equally to this chapter.

A. Alam JH Institute of Molecular Medicine, Jamia Hamdard, New Delhi, Delhi, India Inflammation Biology and Cell Signaling Laboratory, ICMR-National Institute of Pathology, Safdarjung Hospital Campus, New Delhi, Delhi, India A. Kumar Department of Microbiology, Tripura University (A Central University), Agartala, Tripura, India P. Tripathi National Institute of Immunology, New Delhi, Delhi, India JH Institute of Molecular Medicine, Jamia Hamdard, New Delhi, Delhi, India N. Z. Ehtesham Inflammation Biology and Cell Signaling Laboratory, ICMR-National Institute of Pathology, Safdarjung Hospital Campus, New Delhi, Delhi, India e-mail: [email protected] S. E. Hasnain (*) JH Institute of Molecular Medicine, Jamia Hamdard, New Delhi, Delhi, India Dr Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Hyderabad, India e-mail: [email protected]

# Springer Nature Singapore Pte Ltd. 2019 315 S. E. Hasnain et al. (eds.), Mycobacterium tuberculosis: Molecular Infection Biology, Pathogenesis, Diagnostics and New Interventions, https://doi.org/10.1007/978-981-32-9413-4_18 316 A. Alam et al.

integral part in the lifecycle of any unicellular organism. The ability to easily inter- switch between planktonic to a sessile life cycle is an important armor for these unicellular organisms to overcome stress. The matrix not only acts a physical barrier that protects the bacteria but also provides an ecological niche for close interaction and communication among these unicellular entities. This coordinated community-like behavior synchronizes metabolic upregulation or downregulation, both in time and space, and allows these microorganisms to achieve physiological proficiency in terms of ability to tolerate stress that might not be possible as a single cell. Research on biofilms from the perspective to explore the mechanisms of drug tolerance is now considered an apt model as compared to the use of planktonic microorganisms. The increasing use of medical implants further necessitates the need to accelerate research in anti-biofilm strategies for these medical devices. This chapter presents an overview of the mechanisms of biofilm formation and the various interventions for prevention of biofilms.

Keywords Biofilm · Persister · Drug tolerance · Mycobacterium tuberculosis

Abbreviations

ABC ATP-binding cassette transporters CaCl2 Calcium chloride cAMP Cyclic adenosine monophosphate DNA Deoxyribonucleic acid FDA Food and Drug Administration GMP Guanosine monophosphate kHz Kilohertz MgSO4 Magnesium sulfate NaCl Sodium chloride RNA Ribonucleic acid

18.1 Introduction

Microbial biofilm refers to assemblage of monospecies or poly-species microbial cells that is irreversibly attached with surface or with each other and is embedded within an extracellular matrix of polymeric substances. Cells within the microbial biofilms exhibit an altered phenotype in terms of growth rate or gene transcription and are therefore distinct as compared to the colonies of bacteria growing on agar plate macromolecular polymer matrix. Antony van Leeuwenhoek is credited with the first account of biofilm when he reported (1683–1708) the presence of aggregated microbes in the scurf of his teeth and tongue. Bacterial adhesion to surfaces is a common phenomenon that has routinely been observed in the environment. The term film has been used in marine microbiology 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 317 to distinguish sessile microorganisms from the planktonic forms. The “bottle effect” observed for marine microorganisms pointed that bacterial growth and activity increased significantly due to attachment of these organisms on the surface. Initial investigation about the composition of biofilm using “ruthenium red” dye followed by fixation with osmium tetroxide showed that biofilms consist of polysaccharides. The medical rele- vance of biofilm emanated from the studies carried out on sputum samples of patients suffering from lung infection due to cystic fibrosis (CF) caused by Pseudomonas aeruginosa.Thefirst image of medical biofilms, published in 1977, was obtained from lung biopsies and showed aggregated bacteria surrounded by “glycocalyx.” The term glycocalyx was soon replaced with “biofilms” to emphasize the in vivo sessile growth of colonized microorganisms that were considered to cause antimicrobial resis- tance as compared to planktonic microorganisms (Costerton et al. 1987). The focus of research that was initially directed toward testing medicaments on planktonic forms of microorganisms soon diverged as it was observed that organisms in the biofilms showed altered physiology and gene expression. Nearly 80% of the hospital-borne infections is attributed to biofilm-forming microorganism and hence studies on biofilm-forming microorganism have taken the front seat of all major research on drugs (Kumar et al. 2017). The major thrust in biofilm research has been due to the advent of instrumentation such as scanning electron microscopy (SEM), confocal laser scanning microscopy, Robbins device biofilm sampler, etc.; novel techniques such as that have complemented standard microbiological culture techniques for biofilm characterization. Bacteria ubiquitously form biofilm and the mechanism of adhesion is highly conserved across different species of bacteria during the course of evolution. Biofilm formation is an inherent and distinct phenomena among bacteria surviving in the environment, where adhesion and protection is of higher importance and hence a sessile life cycle is favored. The evolution of nonpathogenic environmental bacteria to patho- genic strains has led to shedding of the biofilm machinery and enhancement of motility machinery required for pathogenicity-invasion of host cells and escape from immune cells. This is evident from the observation that subculturing of planktonic laboratory strains of bacterium leads to shunting off of the genetic machinery involved in expres- sion of cellular components required for adhesion, thereby rendering these cells inca- pable to form biofilms. In the presence of selective pressure such as surfactants and antibiotics, genetic machinery essential for formation of protective surface glycocalyx is triggered. Cells with surface-bound glycocalyx continue to survive while planktonic cells lacking protective surface structures fail to survive. The interaction of bacteria with the environment, within and outside the community, and the outcome of the response toward drugs are predominantly influenced by the phenotype as well. In drug discovery strategies, biofilm model could provide more relevant data that may address core questions as compared to the use of planktonic laboratory strains.

18.2 Biofilm Formation by Bacteria

Bacterial species are known to inhabit a variety of ecosystems and persist even under extremely adverse conditions. The survival mechanisms of some pathogenic microbes can be clearly credited to their ability to form biofilms. In addition to 318 A. Alam et al. acting as a physical barrier against shear forces and antimicrobial agents generated by the host system, biofilms also provide a protective niche for the survival of bacteria (Jefferson 2004). Bacteria use their ability to form biofilms as a defense strategy against host-mediated assaults like perturbations in pH, generation of reactive oxygen intermediates, nutrient deprivation, attack by phagocytic cells, etc. Biofilms are considered as the product of evolution against adverse conditions. The fact that certain stress regulators and chaperones significantly contribute to the phenomenon of biofilm formation indicates that a major motivation behind biofilm formation is defense. Stress response mediators such as σB, RpoS, GroEL, DnaK, and DnaJ have been implicated in microbial biofilm formation. In addition to stress, there are other factors as well that trigger the formation of such a sessile microbial community. Pathogenic and commensal bacteria also use biofilm development as a means to colonize and persist in environments that seem propitious within the host and thus flourish well with time. Bacterial species such as Staphylococcus, Strepto- coccus, Pseudomonas, Vibrio cholerae, etc., display elaborate biofilm development under conditions that offer readily utilizable carbon sources in abundance suggesting that this may also act as a trigger factor. Biofilms formed at high shear locations, such as blood or saliva stream, are highly viscous and display remarkable tensile strength, thus conferring tremendous survival advantage to the constituent microbial cells. Moreover, microbes also reap other benefits from this interactive community such as division of metabolic labor. Close proximity of microorganisms within the biofilm also fosters horizontal exchange of genetic information as well as imparts reproductive advantage to these organisms and may eventually act as a driving force for biofilm formation. Taken together, the community approach of growth endows microorganisms with many benefits and there may be different forces that motivate them to enter into this state.

18.3 Process of Biofilm Formation

The formation of this close-knit community of bacteria called biofilm is a complex process involving multiple steps (Fig. 18.1). The series of events that occur during this process, which eventually leads to adaptation under diverse environmental and nutritional conditions, include (i) initial adherence to surface (ii) microcolony forma- tion (aggregation and encapsulation), (iii) formation of a three-dimensional matrix and maturation, and (iv) detachment and dispersal. Initiation of each of the above stages may be contingent upon regulation of the expression of different sets of genes. Biofilm development is initiated by a reversible attachment of motile bacterial cells to a biotic or abiotic surface. This initial adherence is mediated by physical forces like hydrophobic interactions, van der Waals forces, etc., and thus is a noncommittal stage where microbial cells can detach themselves from the surface. Various environmental and genetic factors contribute to the process of adhesion. Bacteria cells may choose to come off if the surface doesn’t provide requisite anchorage or nutrients to them or if other variables such as bacterial orientation, local temperature, pressure, etc., do not seem favorable. However, once the cell 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 319

Fig. 18.1 Process of biofilm formation. Planktonic forms of bacteria reversibly attach to the substratum and release quorum sensing molecules that allow them to irreversibly aggregate. The sessile bacteria endogenously secrete matrix around the cells, which provide a niche for growth and cell division. In presence of stress, such as drugs, few bacteria cells undergo dormancy and become non-replicating persister cells. Within the mature biofilm the cells communicate and show community-like behavior. As soon as stress is withdrawn, the dormant bacteria become active and are released as planktonic forms to start the next cycle of biofilm formation surface structures like pili, fimbriae, flagella, etc., come into play, the bacterial cells become immobilized and proceed to a state of irreversible attachment. Evidence shows that microbial adhesion extensively depends on the surface properties of substratum. Additionally, it is reported that the bacterial appendages promote elec- trostatic interactions and chemical reactions in order to consolidate the bond between surface and bacterial cells, e.g., surface pili type IV, have been implicated in adherence and movement of P. aeruginosa colonies through viscous cell surfaces (Klausen et al. 2003). Attachment of bacteria to biotic surfaces is largely governed by the interaction of host surface proteins and extracellular carbohydrate moieties with that of the invading bacterial cell. E. coli is reported to adhere to host cells using its pili type I adhesions like FimH that bind to mannosylated moieties on the surface of host (Wright et al. 2007). Similarly, Enterococcus species have been shown to adhere to host cells and aggregate further with the help of adhesion molecules like Sag (Mohamed and Huang 2007). Irreversible adhesion is followed by aggregation of bacterial cells leading to the formation of a microcolony. This results in multiplication of bacteria and coloniza- tion of the site. The formation of such a community demands for the development of methods to communicate and exchange information between the constituent micro- bial cells. Crosstalk between bacterial cells induces the production of exopolysaccharides, signaling molecules, and peptides that enhance structural integ- rity of the biofilm and coordinate bacterial growth via quorum sensing. Quorum sensing can be described as a mechanism where a variety of autoinducers stimulate genetic expression of regulators of cell density. 320 A. Alam et al.

The stimulation of cell signaling molecules and selective regulation of certain genes responsible for extracellular polysaccharide (EPS) secretion triggers the for- mation of a three-dimensional extracellular matrix. EPSs have been found to play a pivotal role in determining the phenotype of a bacterial biofilm. Many bacteria use cyclic-di-GMP as the second messenger for production of various EPS components of the matrix. Cellulose has been shown to be an important component of the extracellular matrix formed by E. coli and Salmonella typhimurium. The complex three-dimensional architecture of a biofilm has channels that aid in the uptake of nutrients and release of waste material. The Staphylococcus species produce poly- saccharide intercellular adhesin (PIA) and poly-N-acetyl glucosamine (PNAG) polymers as the major components of their biofilm matrices, whereas, in the case of P. aeruginosa, polysaccharides such as Psl, Pel, and alginate are the main constituents of extracellular matrix (Orgad et al. 2011). Thus, great diversity exists in the phenotype and nature of different bacterial biofilms. Similarly, dramatic differences in the matrix architecture may result from slightest of environmental perturbations. Other constituents of the EPSs may include proteins, amyloids, DNA, lipids, etc. The concluding stage of biofilm development is characterized by detachment of cells leading to dispersal of bacteria. Bacteria switch from sessile to motile lifestyle either by passive sloughing due to fluid shear or by specific dispersal due to nutrient limitations, environmental cues, and other signals. These signals stimulate the expression of EPS-degrading enzymes and motility structures while downregulating the genes mediating bacterial adherence (McDougald et al. 2012). Detached cells may become quickly adapted to planktonic life or may disseminate further to colonize newer sites.

18.4 Factors Involved in Regulation of Biofilm Formation

The initiation of the complex process of biofilm development relies largely on the dynamics of bacterial attachment to a substratum. Bacterial cell surface properties such as hydrophobicity, presence of motility appendages, and surface molecules play an important role in initial attachment to a surface. Certain features of a surface, such as roughness, may favor biofilm formation. Rough surfaces present higher surface area and arrest shear forces. The rate and extent of bacterial adherence may also be affected by the physicochemical properties of a surface. Interaction with hydrophobic nonpolar surfaces results in stronger attachment as they appear to enable bacteria to defy repulsive forces. Fluctuations in the nutrient pool or ionic concentrations in the biotic or abiotic habitat of bacteria may alter the properties of biofilm. In some strains of Lactobacil- lus, nutrient limitation (especially low carbon to nitrogen ratio) may trigger biofilm formation. Contrary to this, in P. aeruginosa nutrient deprivation induces detach- ment of the biofilm. Similarly, availability of peptides and amino acids in host tissues and body fluids has been reported to be crucial for several bacteria to form biofilm. 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 321

Biofilm formation in Listeria monocytogenes proceeds only in the presence of carbohydrates like mannose and trehalose and appropriate levels of phosphate in the local environment. The presence of salt such as NaCl, MgSO4, and CaCl2 in appropriate concentrations has also been reported to be crucial for biofilm formation (Marsden et al. 2017). Environmental factors also play a significant role throughout the process. The concentration of second messengers such as cyclic-di-GMP and cAMP, crucial for surface attachment of bacteria, is controlled by the levels of carbon and oxygen in the local environment (McDonough and Rodriguez 2012). Perturbations in the environ- mental conditions result in dynamic changes in biofilm structure; e.g., under oxygen- rich conditions P. aeruginosa displays a mushroom-shaped morphology with water channels between macro-colonies consisting of rodlike cells, whereas, under anoxic conditions, it has a three-dimensional mesh-like morphology with channels between macro-colonies of elongated filamentous cells. These morphological changes possi- bly alter the diffusion of nutrients and aid the bacteria to adapt metabolically. Insufficient oxygen may abrogate the process of biofilm formation in E. coli. Temperature is reported to be an important factor for the regulation of EPS produc- tion in Clostridium perfringens and hence it dramatically affects the biomass, density, and thickness of the biofilm. Similarly, L. monocytogenes loses its ability to form biofilms at higher temperature. In addition to temperature, variations in pH have also been shown to influence biofilm formation in the clinical isolates of Burkholderia pseudomallei. While alkaline environment is required by Vibrio cholerae cells to multiply and form biofilm, in the case of Lactobacillus rhamnosus, biofilm formation is modulated by the presence of bile and mucins, low pH, and high osmolarity (Lebeer et al. 2007). Overall, such changes in the morphology of bacterial biofilms may be seen as a fitting response to the environmental cues and a practice for better adaptation to the surrounding environment. The intrinsic biofilm formation ability is also contingent upon genetic factors. Studies on the regulatory mechanism driving bacterial biofilm formation have identified several genes such as the flagella regulator sinR in Bacillus subtilis and transcription activator prfA in L. monocytogenes as the major orchestrators of the process (Newman et al. 2013). Becker et al. (2001) reported that Staphylococcus aureus residing in biofilms showed enhanced expression of enzymes such as phos- phoglycerate mutase and alcohol dehydrogenase that are involved in glycolysis and fermentation. In addition, the transcript levels of two other genes sarA and sigB were found to be distinctly upregulated in S. aureus biofilms. Bacteria deficient in sigB were found to have lost the ability to form biofilm. Gene expression studies in E. coli biofilms revealed enhanced expression of ompC and slp, the genes encoding outer membrane protein and lipoprotein, respectively, which have recently been associated with the initial steps of E. coli biofilm formation. The microarray analysis of biofilms also exhibited differential expression of genes under hypoxic and nutrition-deprived conditions. RpoS, the σS subunit of RNA polymerase, is consid- ered to play a pivotal role in biofilm formation. Majority of the differentially regulated genes identified till date are involved in bacterial motility, chemotaxis, attachment, membrane bioenergetics, glycolysis, and phage-related functions. 322 A. Alam et al.

18.5 Interaction of Microbial Community Within Biofilms

Within the biofilm, bacterial cells can coexist with members of same species of bacteria or with other species of microorganisms. The metabolic potential of each species of bacteria govern the relative composition of bacteria within the biofilms. The substrate concentration can influence the population of the bacteria; the fast- growing population of bacteria can dominate over the slow-growing bacteria. The interaction between cells may be competitive in terms of nutrient, space, metabolites, etc.; however in certain conditions cells may act cooperatively to facilitate survival of bacteria from same or other species. The distribution of bacterial population is also dependent on oxygen tension; the core of the biofilm is dominated by cells that are anaerobic while the outer layer of biofilms support aerobic bacteria. This spatial distribution of cells within biofilms is evident in the case of biofilms inhabitated by aerobic Burkholderia cepacia and facultative aerobic Klebsiella oxytoca.Atan initial microcolony formation stage when the biofilm matrix is minimum, B. cepacia dominates the population due to its higher growth rate. As the biofilm matrix accumulates it hinders diffusion of oxygen and creates a gradient of oxygen tension. Under low oxygen tension, K. oxytoca gains advantage and grows at a faster rate as compared to B. cepacia. Bacteria produce toxins and metabolites that confer them with competitive edge over interÀ/intraspecies of bacteria and allow them to gain foothold within the matrix. P. aeruginosa species can produce growth inhibi- tory toxins that allow it to invade the biofilms caused by B. cepacia and eventually displace the competitor cells. When both species of bacteria produce inhibitory toxins, then the net effect on competitive survival of cells depends on timing of release of toxins and the relative toxicity response of bacteria from each species. In a mixed biofilm, toxins produced by Pseudoalteromonas tunicata suppress the growth of sensitive species like Cytophaga fucicola but not of insensitive species such as Roseobacter gallaeciensis. The protective efficacy of biofilms is enhanced due to the synergistic presence of multiple species of bacteria within biofilms (Tolker-Nielsen 2015). A study showed that marine bacteria, when co-cultured with other species of bacteria, achieved nearly twofold higher levels of survival within biofilms as com- pared to monospecies (Burmølle et al. 2006). Staphylococcus aureus and Candida albicans are often found to coexist in biofilms and exhibit higher virulence and tolerance to antimicrobial drugs such as vancomycin. C. albicans is usually absent on teeth of health individuals and does not cause symptomatic disease. However, when C. albicans partners with dental bacteria such as Streptococcus gordonii or S. oralis, it enhances bacterial aggregation, colonization, and biofilm formation. These cross-kingdom interactions between fungi and bacteria can be either coopera- tive or competitive and mediate synergistic or antagonistic associations that are further modulated by host immune responses and local environmental factors. 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 323

18.6 Antibiotic Resistance of Bacterial Biofilms

Majority of the pathogens and environmental organisms are found to be present in biofilm and as single entity in nature. Majority of the microbial and chronic infections are associated with biofilms that increase chances of nosocomial infections. Bacterial biofilm resists phagocytosis and immune response on pathogens. It is known that the aquatic biofilms are reservoirs of resistant bacteria that present very close to each other increase chances of genetic exchanges such as antibiotic resistance genes (Balcázar et al. 2015). There are bacterial biofilms present on the root and other parts of plants that may protect from plant pathogens (Bogino et al. 2013). It has been reported that during maturation of biofilms there are increases in the expression of genes related to energy metabolism and transporter genes whereas decrease in the expression of genes related to cell structure, central intermediary metabolism, energy metabolism, nucleotide biosynthesis, and metabo- lism that indicates that biofilm cells are metabolically active but retard its cell division. There are abnormal conditions present inside biofilms where depletion of oxygen concentration, slow growth, increased doubling times due to nutrient limita- tion, gradients in pH, change in the metabolic activity of microorganisms, presence of compounds that may inactivate antibacterial compounds, and slow diffusion of antibiotics make microorganisms resistant to antibiotics and drugs. The antibiotic resistance in biofilms may be divided into innate biofilm factors such as oxidative stress, microenvironments within biofilms, biofilm matrix as a diffusion barrier, etc., or induced resistance factors such as changes in the protein expression at transcrip- tional level, stress protein expression, increased horizontal gene transmission resulting in drug resistance, etc. Reduced antibiotic susceptibility due to biofilm formation at the implanted devices causes persistent infections. It has been reported that P. aeruginosa isolates from cystic fibrosis (CF) patients are sensitive to antibiotics in planktonic form but difficult to eradicate due to the presence in biofilms (Hengzhuang et al. 2012). In the case of tuberculosis, M. tuberculosis face unfavor- able conditions inside hosts that are similar to biofilms that increase the chances of drug tolerance by pathogen.

18.7 Tolerance to Antibiotics and Efflux Pumps

More than 80% of nosocomial infections are related to biofilms that may be present at the medical devices. In the biofilms, bacteria encounter several types of stresses; because of that there are upregulation of toxin-antitoxin systems, different stress proteins, proteins involved in growth regulation, and downregulation of metabolic proteins that causes tolerance to antimicrobial drugs. There are bacteria of different physiological states present in the biofilm due to differential gene expression resulting in differences in the susceptibility of antimicrobial compounds. There are 324 A. Alam et al. different mechanisms that may be activated during drug tolerance such as modifica- tion in drug target site, inactivation of drugs, upregulation of efflux pumps, etc. It is known that expression of efflux pumps found in both bacteria and eukaryotic organisms changes in response to different physiological and environmental conditions (Sun et al. 2014), and these pumps may efflux antibiotics, heavy metals, pollutants, chemicals involved in quorum sensing, compounds produced by plants, bacterial metabolites, etc. (Blanco et al. 2016). In the biofilm, there is activation of efflux pumps and stress response toxin-antitoxin systems that increase tolerance to drugs, so these may be effective targets for anti-biofilm strategies. It has been reported that there are different efflux pumps in P. aeruginosa that are involved in resistance to different antibiotics such as ciprofloxacin, gentamicin, and tobramycin. MacABCsm, an ABC-type tripartite efflux pump, is involved in drug resistance and biofilm formation in Stenotrophomonas maltophilia. The multidrug resistance efflux pumps of Salmonella enterica serovar Typhimurium are involved in biofilm forma- tion and drug resistance. There is increase in the cases of drug resistance in the tuberculosis infection. M. tuberculosis survives inside macrophages that induces efflux pumps resulting in increases in drug tolerance and treatment duration (Adams et al. 2011). Inhibition of these efflux pumps may be effective strategies to decrease treatment duration.

18.8 Impact of Biofilm Formation and Phenotypic Behavior

The genes that regulate biofilm formation have largely remained conserved along the course of evolution, pointing to the fact that biofilm formation is an essential component in the life cycle of the unicellular organism. Biofilm formation offers an ecological niche for the unicellular microorganisms to come in contact with each other and offers the unicellular organisms with a great chance to adopt a lifestyle where the functions of many cells cooperatively act to drive across a process that might be impossible to carry as a single cell. The close proximity of unicellular microorganisms within the matrix allows cells to transfer genes through conjugation. It has been established that the conjugation rates in biofilms are as high as 1000-fold as compared to planktonic forms (Stalder and Top 2016). Biofilms are thus consid- ered “hot spots” for horizontal gene transfer and allow bacteria to evolve and adapt to changing environment and stress such as antibiotics, pollutants, and heavy metals. Within the matrix of biofilms, cells attain dormancy and undergo several phenotypic changes including a subpopulation of non-replicating cells called persisters. Persister cells are endowed with higher drug tolerance and can survive nearly 1000-fold higher minimum inhibitory concentration of drugs. The vast diversity of mechanism for biofilm formation among different species of bacteria has posed a serious challenge in terms of devising strategies to combat biofilm. A recent study has implicated the role of chaperonic proteins in biofilm formation. Chaperons aid in protein folding and shape the three-dimensional structure of many proteins simulta- neously. Peptidyl-prolyl isomerases (PPIase) are a class of cyclophilins that are ubiquitously expressed in prokaryotes and eukaryotes and are involved in 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 325 interconversion of cis and trans isomers of peptide bonds with proline residue. PPIase possess chaperonic activity and regulate several key functions in bacteria including iron regulation, signaling, and immune defense. In Mycobacterium tuber- culosis, PPIase is present in two forms – PpiA and PpiB. M. tuberculosis PpiA share structural and phylogenetic similarity with cyclophilins present in eukaryotes. M. tuberculosis PpiB is essential for the survival of the pathogen and is involved in biofilm formation (Pandey et al. 2016, 2017). M. smegmatis overexpressing M. tuberculosis PpiB showed enhanced biofilm formation and exhibits drug toler- ance to anti-TB drugs. As a result, higher dosage of anti-TB drugs are required for treatment of patients suffering from tuberculosis (TB), thereby causing toxicity in patients. The involvement of M. tuberculosis PpiB in biofilm formation as well as in several key pathways of the bacterial metabolism makes it a “choke point” enzyme for inhibitors. It has been shown that cyclophilin inhibitors such as cyclosporine A and several FDA-approved drugs including acarbose can bind with PpiB. Cyclosporine A, acarbose, and gallium nanoparticle can inhibit the activity of M. tuberculosis PpiB, thereby suppressing formation of biofilm. The reduction in biofilm formation results in higher penetrance of drugs through biofilm matrix. Tackling biofilm-related infection is a cumbersome task due to the fact that several heterogenous species of bacteria exist within biofilms. The use of inhibitors can selectively eliminate a group of bacteria from the heterogenous population; however a subpopulation of bacteria that are resilient to the activity of these inhibitors can transfer gene to other bacteria. As a result, several inhibitors that have been tested against biofilm-forming microorganism either failed or their efficacy gradually reduced with time. The bottleneck in devising strategy for targeting biofilms has been the lack of a target that is essential across a wide spectrum of microorganism. PpiB can prove to be a novel target for biofilm-forming organisms due to its presence in all biofilm-forming bacteria. Besides, PpiB homologues in several biofilm- forming bacteria possess similar amino residues in the domain that binds with cyclosporine A, acarbose, and gallium nanoparticle (Kumar et al. 2019). Targeting PpiB could prove to be the masterstroke against a wide genre of biofilm-forming bacteria involved in multiple pathogenic symptoms.

18.9 Pathology Associated with Biofilm-Forming Bacteria

Biofilms are a major reason behind the recalcitrant nature of pathogenic microbes toward antimicrobial agents and host immune responses, as biofilms act as a physical barrier against drugs and provide a protective niche for mycobacterial survival. Conventional biofilms are extremely resistant to host inflicted insults and can endure antimicrobial agents as much as 10–1000 times of that required to eradicate plank- tonic microbes. Biofilms have also been established as a crucial factor in the pathogenesis of a large number of bacteria. The nature of biofilms also makes it formidable for the host immune cells to attack or opsonize bacteria harbored within them (Piatek et al. 2013). S. aureus evades opsonization by phagocytic cells in blood by producing coagulase that activates prothrombin and promotes production of 326 A. Alam et al. insoluble fibrin which eventually leads to strengthening of the biofilm (Cheng et al. 2010). In cystic fibrosis patients, alginate and rhamnolipid production is enhanced by P. aeruginosa as a strategy to escape macrophages and stimulate necrosis of polymorphonuclear leukocytes, respectively (Van Gennip et al. 2009). The oral cavity provides an excellent niche for survival of nearly 800 species of microorganisms that form a complex-multispecies biofilm or plaque. Anaerobic bacteria such as Porphyromonas gingivalis colonize the biofilm at a later stage, erode the dental surface with acidic toxins, and cause periodontitis and peri- implantitis. P. gingivalis differentially express nearly 20% of its genome during the various stages of biofilm formation and virulence. The versatility of P. gingivalis in attaching to oral soft tissue, dental implant, and other oral bacteria is due to the presence of fimbriae on the cell surface. Loss of Fim A fimbriae results in inhibition of adherence to epithelial cells and dental fibroblasts. Recent studies showed that arginine plays a key role in fimbriae-mediated biofilm formation. The activity of streptococcal Arc A, which catalyzes hydrolysis of arginine to citrulline, suppresses fimAexpression in P. gingivalis and prevents biofilm formation. Unlike other bacteria that utilize siderophores for sequestration of iron, P. gingivalis possess proteases such as gingipains for acquiring iron from host heme. Several species of dental microbiota exhibit mutualistic coexistence with P. gingivalis. Isobutyric acid produced by P. gingivalis aid in growth of Treponema denticola, while succinate produced by T. denticola enhances growth of P. gingivalis. P. gingivalis infections lead to chronic inflammatory disease of oral cavity characterized by loss of teeth and increased risk of cardiovascular diseases. M. tuberculosis is endowed with persistence and drug tolerance, typical of biofilm-associated infections caused by pathogenic bacteria. It has been reported that M. tuberculosis develops biofilms within host cells that allow it to evade the immune surveillance. M. tuberculosis biofilms are associated with several other non-tubercular bacteria and opportunistic pathogens. M. tuberculosis biofilms are mainly composed of methoxy-free mycolic acids and its biosynthesis is modulated by fatty acid synthase complex II-GroEL-mediated pathway. Disruption of polyke- tide synthases, pks16 and pks1/15, is associated with reduced biofilm formation in M. tuberculosis and suggests the role of surface lipids in initial aggregation of mycobacteria. As such the high lipid content of the bacterial cell wall limits the diffusion of drugs; biofilm-encapsulated mycobacterial cells show enhanced drug refractory physiology. M. tuberculosis exhibit enhanced tolerance to up to 1000 times higher minimum inhibitory concentration of drugs due to the presence of persister cells within the biofilms. S. epidermidis biofilm provides resistance to phagocytosis by interfering with the activation of complement protein C3b (Kristian et al. 2008). Biofilm infections have also been implicated in various other diseases such as endocarditis, chronic otitis media, obstructive pulmonary diseases, and chronic (diabetes) wound infections. Biofilms also dwell in patients with internal medical devices for treatment (Table 18.1). Similarly, lung infections by P. aeruginosa can be significantly attenuated by targeting the quorum sensing pathway (O’Loughlin et al. 2013). Staphylococcal biofilm formation is demonstrated to be suppressed in vivo by using RNAIII-inhibiting peptide. 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 327

Table 18.1 List of clinically relevant biofilm-associated diseases Organ/ body part Drugs/ affected Organism Disease/symptom interventions Eye Staphylococcus epidermidis, Endophthalmitis, Tobramycin, (contact Staphylococcus aureus, keratitis, scleral buckle polymyxin, lens) Staphylococcus infection, lacrimal bacitracin, saprophyticus, Serratia infection, periorbital levofloxacin marcescens, Fusarium infections solani Wound Bacillus, Clostridium, PASH syndrome, delayed Infliximab, (burns) Staphylococcus epidermidis, healing in burns clindamycin, Staphylococcus aureus metronidazole, Amikacin Dental Porphyromonas gingivalis, Dental plaque, dental Sodium Lactobacillus casei, caries, pulpitis, apical hypochlorite, Streptococcus sobrinus periodontitis, gingivitis acetaminophen, corticosteroid Prosthetic Enterococci, diphtheroids, Endocarditis Ampicillin, heart valve candida spp. gentamicin, penicillin G Breast Staphylococcus epidermidis, Chronic infection of Chloramex, implant Corynebacterium, breast Fucidin, Mycobacterium fortuitum Terramycin-coated implant Urinary Neisseria gonorrhoeae, Urinary tract infections Fluoroquinolone, catheter Staphylococcus aureus, cephalosporin, Enterococcus faecalis aminoglycoside Orthopedic Pseudomonas, streptococci, Chronic osteomyelitis, Titanium nitride metal Proteus mirabilis peri-prosthetic joint coating, implants infections gentamicin-coated polyurethane sleeves Lung Burkholderia cepacia, Cystic fibrosis, Ivacaftor, Pseudomonas aeruginosa, ventilator-associated bronchodilators, Aspergillus fumigatus pneumonia ureidopenicillin, carbapenem Ear Haemophilus influenza Influenza, chronic otitis Ofloxacin ear drop, media amoxicillin Vagina Corynebacterium spp., Intrauterine device- Metronidazole, Micrococcus spp., Candida associated infections, gentamycin albicans, Lactobacillus Gardnerella vaginitis plantarum, group B streptococci 328 A. Alam et al.

18.10 Anti-biofilm Strategies

Due to the adverse impact of biofilms on human health, various methods such as mechanical disruption, disinfectants, chlorination, and ultraviolet have been used traditionally. Recent advances in mechanism of biofilms have provided new insights to control biofilm-based infections. Table 18.2 shows the list of various anti-biofilm strategies that are either used alone or in combination. Extracts from plants such as Rhodiola crenulata, Malus pumila, Dolichos lablab, etc., show potent anti-biofilm activity. Chemical derivatives of active ingredients of plant extracts such as icariin and resveratrol can be synthesized and exhibit significant anti-biofilm activity. Green tea possesses high level of anti-oxidants and can suppress biofilm caused by Streptococcus mutans. Honey possesses antimicrobial property due to the presence of antimicrobial peptide, defensins, and is effective in inhibiting Enterococcus spp. biofilm formation. Similarly, several essential oils such as cumin oil, cinnamon oil, and oregano oil are effective against biofilm caused by Klebsiella pneumonia and mixed-species biofilms of enteropathogenic E. coli and S. aureus, respectively (Ong et al. 2018). The emergence of antibiotic resistance has highlighted the use of bacteriophages in controlling bacterial infections. Phages are highly specific toward bacterial targets, persist as long as the host bacterium is present, and do not show side effects on normal microbiota. Bacteriophages can potentially release enzymes such as polysaccharide depolymerase that liquefy the biofilm matrix and allow the phage to penetrate the biofilm. Genetically engineered phages such as E. coli-specific phage (T7), cloned with dispersin (dspB) gene, have shown higher efficacy in biofilm degradation. The major technological challenges in using phages for anti- biofilm strategy are i) reducing the release of toxins by phage, ii) overcoming bacterial resistance to phages, and iii) neutralizing the virulence genes of phages from incorporating into the host bacterial genome. The quorum sensing (QS) molecules such as acetyl homoserine lactone (AHL), autoinducer 2 (AI-2), etc., aid in aggregation of bacteria. Studies have shown that inhibitors that quench the activity of quorum sensing molecules can disrupt the aggregation of cells, considered as an essential step in biofilm life cycle. Naturally occurring quorum sensing inhibitors such as furanone, ajoene, naringin, curcumin, epigallocatechin gallate, etc., have shown anti-biofilm properties against several gram-negative and gram-positive bacterial species. Quorum quenching enzymes such as AHL lactonases, AHL acylases, and oxidoreductases are effective in disruption of biofilm without affecting the microorganism. Therefore treatment with antibiotics in combi- nation with QS molecules or quorum quenching enzymes can be effective in targeting bacteria within biofilms. QS inhibitors such as furanone and penicillic acid increased the efficacy of antibiotic tobramycin against P. aeruginosa biofilms. Blockage of cell to cell communication by using peptides such as RNA III-inhibiting peptide (RIP) that inhibits phosphorylation of TRAP (target of RNA III) proteins can prevent initial stages of biofilm formation. The formation of biofilms on medical devices and implants poses a unique challenge for engineering devices that are inert 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 329

Table 18.2 List of anti-biofilm strategies Strategy Mode of action Efficacy against Plant extracts Act as antioxidants Streptococcus mutans, Green tea, Lactobacillus plantarum, Cuminum S. epidermidis cyminum Cinnamon oil Oregano oil Thyme oil Bacteriophage Infects and lyse bacteria cells P. aeruginosa, P. fluorescens Engineered with dispersion B gene Phage PhilBB-PF 7A Phage in combination with antibiotics Quorum sensing Blocks intercellular communication P. aeruginosa, B. cepacia, inhibitors carried out through quorum sensing S. aureus, Y. enterocolitica AHL cyclase molecules AHL lactonases AIP inhibitors Oxidoreductases Surface Prevents attachment and aggregation S. epidermidis, biofilms on heart modification of cells on surface devices Coating with furanone Zirconium oxide Nanocrystalline silicon carbide Polyethylene glycol Nanoparticle Generation of reactive oxygen Candida spp., E. coli, Silver species, destabilizing cell membrane Enterococcus spp., Salmonella Gallium nitride and enzymes spp. Titanium oxide Silica nanoparticle Enzymes Breakdown of components of biofilm Streptococcus pyogenes, DNase 1 matrix Acinetobacter baumannii, fl Lysostaphin Haemophilus in uenza Lyase Photodynamic Activation of photosensitizing drugs C. albicans, Enterococcus Therapy due to light faecalis Methylene blue Toluidine blue (continued) 330 A. Alam et al.

Table 18.2 (continued) Strategy Mode of action Efficacy against Bacteriocin Formation of pores on cell membrane Listeria monocytogenes, Nisin A of bacteria S. aureus Lacticin Q Ultrasonic Mechanical disruption E. coli, P. aeruginosa treatment >20 kHz frequency of sound waves Anti-biofilm Disruption of initial stages of biofilm Wound colonizing P. aeruginosa, agents matrix peridontal colonizing Molsidomine Aggregatibacter Diethylamine actinomycetemcomitans, nonoate Salmonella spp., Porphyromonas diethylammonium gingivalis Povidone iodine Xylitol Dispersin B to biofilm formation (Koo et al. 2017). Since surface attachment is the first step of biofilm formation, coating the surface with inert substances or chemical modification of surface is an innovative strategy for preventing biofilm formation on medical devices. Surface adsorption of furanone and quaternary ammonium salts prevents Staphylococcus epidermidis biofilm formation on catheters. Nanoparticles such as silver possess antimicrobial properties. Surface medication with silver coating prevents biofilm formation and studies have shown that silver nanoparticles can reduce 97% of biofilm formation caused by Candida spp. Similarly, titanium dioxide nanomaterials are effective against fungal biofilms and are widely used in dental implants (Roy et al. 2018). Gallium shares similar charge and shape with ferric ion and binds with iron siderophores. Iron acts as cofactor of key enzymes involved in bacterial metabolism and disruption of iron metabolism can significantly inhibit growth of bacteria. Gallium competes with ferric ion and inactivates Fe siderophores, thereby leading to suppression of growth and metabolism of bacteria. Recent studies show that gallium nanoparticles not only inhibit biofilm formation but also improve the efficacy of anti-tuberculosis drugs against mycobacterial biofilms (Kumar et al. 2019). Several constituents of biofilms such as DNA, proteins, and polysaccharides can be targeted by using enzymes. A mixture of deoxyribonucleotides, glycosidases, and proteases has shown significant anti- biofilm potential. Treatment with DNase1 (5 μg/ml) either alone or in combination with antibiotics (rifampicin, ampicillin, azithromycin, etc.) reduced biofilm forma- tion of S. aureus, S. pyogenes, Acinetobacter baumannii, Haemophilus influenza, etc. Lysostaphin (LS) in combination with nafcillin and enzymes such as amylase, lyase, and lactonase are effective against methicillin-resistant S. aureus (MRSA). 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 331

Photodynamic therapy is the process of activating drugs, called photosensitizer, using light of particular wavelength. During the exposure of light, photosensitive drugs produce reactive oxygen that destroys adjacent cells. Oral biofilms that consist of multispecies microorganism are treated with helium laser light in the presence of toluidine blue.

18.11 Conclusion

Combating microbial biofilm infections and getting rid of it completely is a daunting task for clinicians. Tackling biofilms is difficult as there are several proteins that are involved in biofilm formation and the involvement of these proteins also varies across biofilm-forming organism. Prolonged treatment with combinations of antibiotics that possess different killing mechanisms can be used to eliminate the infection. However, the efficacy of this method may still be questionable. The mutation rate of bacterial genome is high and offers it a unique advantage to adapt to stress. As a result of mutation the protein targets change their conformation or the active site amino acids get altered, thereby reducing the interaction of inhibitors to target genes or proteins. Apart from this genotypic adaptation, bacteria can tempo- rarily adapt phenotypically by altering the secretion of extracellular molecules that in turn form an impregnable coating. The extracellular coating or matrix can deposit over biotic or abiotic surfaces. In other cases, removal of medical devices and abscesses is necessary for a successful outcome. Treatment only with antibiotics is not a viable option in the case of biofilm-associated infection, partly because the mutation in bacterial genes renders the protein targets inaccessible to drugs and the gradient of drugs across the matrix decreases toward the core of the biofilm. A successful medicament against biofilm would involve a combination of several anti-biofilm strategies that target different steps of biofilm formation. In the light of the fact that dissemination of persister cells from the mature biofilm is preceded by dissolution or loosening of the matrix, the auto-dissolution of matrix is governed by mechanisms that are inherently controlled by the persister cells. The signal for stress removal is communicated from the cells on the outer surface of the matrix to the persister cells within the core of the matrix. It may inevitably lead to change in gene expression of persister cells that in turn result in generation of “factors” that eventually dissolve the matrix. RNA sequencing of persister cells may shed light about “wake up” genes that regulate the expression of these “factors” that can potentially dissolve the matrix. The usefulness of these “factors” as a futuristic tool against biofilm-associated infections is an open question. Other future strategies for biofilm control may include targeting the essential regulatory pathways such as nucleotide signaling and damaging the amyloid structures which play a key role in biofilm formation and adaptation (Romero et al. 2010). Alternatively, chaperonic PpiB proteins, considered as a “choke point protein” involved in biofilm formation across many biofilm-forming pathogens and which regulate other proteins in the cell, could pave the way for a successful target against biofilm-borne infections. 332 A. Alam et al.

References

Adams KN, Takaki K, Connolly LE, Wiedenhoft H, Winglee K, Humbert O, Edelstein PH, Cosma CL, Ramakrishnan L (2011) Drug tolerance in replicating mycobacteria mediated by a macrophage-induced efflux mechanism. Cell 145(1):39–53 Balcázar JL, Subirats J, Borrego CM (2015) The role of biofilms as environmental reservoirs of antibiotic resistance. Front Microbiol 6:1216 Becker P, Hufnagle W, Peters G, Herrmann M (2001) Detection of differential gene expression in biofilm-forming versus planktonic populations of Staphylococcus aureus using micro- representational-difference analysis. Appl Environ Microbiol 67(7):2958–2965 Blanco P, Hernando-Amado S, Reales-Calderon JA, Corona F, Lira F, Alcalde-Rico M, Bernardini A, Sanchez MB, Martinez JL (2016) Bacterial multidrug efflux pumps: much more than antibiotic resistance determinants. Microorganisms 4(1):pii. E14 Bogino PC, Oliva Mde L, Sorroche FG, Giordano W (2013) The role of bacterial biofilms and surface components in plant-bacterial associations. Int J Mol Sci 214(8):15838–15859 Burmølle M, Webb JS, Rao D, Hansen LH, Sørensen SJ, Kjelleberg S (2006) Enhanced biofilm formation and increased resistance to antimicrobial agents and bacterial invasion are caused by synergistic interactions in multispecies biofilms. Appl Environ Microbiol 72:3916–3923 Cheng AG, McAdow M, Kim HK, Bae T, Missiakas DM, Schneewind O (2010) Contribution of coagulases towards Staphylococcus aureus disease and protective immunity. PLoS Pathog 6(8): e1001036 Costerton JW, Cheng KJ, Geesey GG, Ladd TI, Nickel JC, Dasgupta M, Marrie TJ (1987) Bacterial biofilms in nature and disease. Annu Rev Microbiol 41:435–464 Hengzhuang W, Wu H, Ciofu O, Song Z, Høiby N (2012) In vivo pharmacokinetics/pharmacody- namics of colistin and imipenem in Pseudomonas aeruginosa biofilm infection. Antimicrob Agents Chemother 56(5):2683–2690 Jefferson KK (2004) What drives bacteria to produce a biofilm? FEMS Microbiol Lett 236 (2):163–173 Klausen M, Heydorn A, Ragas P, Lambertsen L, Aaes-Jorgensen A, Molin S, Tolker-Nielsen T (2003) Biofilm formation by Pseudomonas aeruginosa wild type, flagella and type IV pili mutants. Mol Microbiol 48(6):1511–1524 Koo H, Allan RN, Howlin RP, Stoodley P, Hall-Stoodley L (2017) Targeting microbial biofilms: current and prospective therapeutic strategies. Nat Rev Microbiol 15(12):740–755 Kristian SA, Birkenstock TA, Sauder U, Mack D, Götz F, Landmann R (2008) Biofilm formation induces C3a release and protects Staphylococcus epidermidis from IgG and complement deposition and from neutrophil-dependent killing. J Infect Dis 197(7):1028–1035 Kumar A, Alam A, Rani M, Ehtesham NZ, Hasnain SE (2017) Biofilms: survival and defense strategy for pathogens. Int J Med Microbiol 307(8):481–489 Kumar A, Alam A, Grover S, Pandey S, Tripathi D, Kumari M, Rani M, Singh A, Akhter Y, Ehtesham NZ, Hasnain SE (2019) Peptidyl-prolyl isomerase-B is involved in Mycobacterium tuberculosis biofilm formation and a generic target for drug repurposing-based intervention. NPJ Biofilms Microbiomes 5:3 Lebeer S, De Keersmaecker SC, Verhoeven TL, Fadda AA, Marchal K, Vanderleyden J (2007) Functional analysis of luxS in the probiotic strain Lactobacillus rhamnosus GG reveals a central metabolic role important for growth and biofilm formation. J Bacteriol 189(3):860–871 Marsden AE, Grudzinski K, Ondrey JM, DeLoney-Marino CR, Visick KL (2017) Impact of salt and nutrient content on biofilm formation by Vibrio fischeri. PLoS One 12(1):e0169521 McDonough KA, Rodriguez A (2012) The myriad roles of cyclic AMP in microbial pathogens: from signal to sword. Nat Rev Microbiol 10(1):27–38 McDougald D, Rice SA, Barraud N, Steinberg PD, Kjelleberg S (2012) Should we stay or should we go: mechanisms and ecological consequences for biofilm dispersal. Nat Rev Microbiol 10 (1):39–50 18 Biofilms: A Phenotypic Mechanism of Bacteria Conferring Tolerance Against... 333

Mohamed JA, Huang DB (2007) Biofilm formation by enterococci. J Med Microbiol 56 (12):1581–1588 Newman JA, Rodrigues C, Lewis RJ (2013) Molecular basis of the activity of SinR, the master regulator of biofilm formation in Bacillus subtilis. J Biol Chem 288(15):10766–10778 O’Loughlin CT, Miller LC, Siryaporn A, Drescher K, Semmelhack MF, Bassler BL (2013) A quorum-sensing inhibitor blocks Pseudomonas aeruginosa virulence and biofilm formation. Proc Natl Acad Sci 110(44):17981–17986 Ong KS, Mawang CI, Daniel-Jambun D, Lim YY, Lee SM (2018) Current anti-biofilm strategies and potential of antioxidants in biofilm control. Expert Rev Anti-Infect Ther 16(11):855–864 Orgad O, Oren Y, Walker SL, Herzberg M (2011) The role of alginate in Pseudomonas aeruginosa EPS adherence, viscoelastic properties and cell attachment. Biofouling 27(7):787–798 Pandey S, Sharma A, Tripathi D, Kumar A, Khubaib M, Bhuwan M, Chaudhuri TK, Hasnain SE, Ehtesham NZ (2016) Mycobacterium tuberculosis peptidyl-prolyl isomerases also exhibit chaperone like activity in-vitro and in-vivo. PLoS One 11(3):e0150288 Pandey S, Tripathi D, Khubaib M, Kumar A, Sheikh JA, Sumanlatha G, Ehtesham NZ, Hasnain SE (2017) Mycobacterium tuberculosis peptidyl-prolyl isomerases are immunogenic, alter cytokine profile and aid in intracellular survival. Front Cell Infect Microbiol 7:38 Piatek R, Zalewska-Piatek B, Dzierzbicka K, Makowiec S, Pilipczuk J, Szemiako K, Cyranka- Czaja A, Wojciechowski M (2013) Pilicides inhibit the FGL chaperone/usher assisted biogene- sis of the Dr fimbrial polyadhesin from uropathogenic Escherichia coli. BMC Microbiol 13:131 Romero D, Aguilar C, Losick R, Kolter R (2010) Amyloid fibers provide structural integrity to Bacillus subtilis biofilms. Proc Natl Acad Sci 107(5):2230–2234 Roy R, Tiwari M, Donelli G, Tiwari V (2018) Strategies for combating bacterial biofilms: a focus on anti-biofilm agents and their mechanisms of action. Virulence 9(1):522–554 Stalder T, Top E (2016) Plasmid transfer in biofilms: a perspective on limitations and oppurtunities. NPJ Biofilms Microbiomes 2:16022 Sun J, Deng Z, Yan A (2014) Bacterial multidrug efflux pumps: mechanisms, physiology and pharmacological exploitations. Biochem Biophys Res Commun 453(2):254–267 Tolker-Nielsen T (2015) Biofilm development. Microbiol Spectr 3(2):MB-0001-2014 Van Gennip M, Christensen LD, Alhede M, Phipps R, Jensen PØ, Christophersen L, Pamp SJ, Moser C, Mikkelsen PJ, Koh AY, Tolker-Nielsen T, Pier GB, Høiby N, Givskov M, Bjarnsholt T (2009) Inactivation of the rhlA gene in Pseudomonas aeruginosa prevents rhamnolipid production, disabling the protection against polymorphonuclear leukocytes. APMIS 117 (7):537–546 Wright KJ, Seed PC, Hultgren SJ (2007) Development of intracellular bacterial communities of uropathogenic Escherichia coli depends on type 1 pili. Cell Microbiol 9(9):2230–2241 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 15

Ashutosh Kumar, Anwar Alam, Pranami Bharadwaj, Sharmistha Tapadar, Mamta Rani, and Seyed E. Hasnain

Abstract Mycobacterium tuberculosis (M.tb), by virtue of its ability to evolve, has devel- oped mechanisms that enable it to modulate its growth through regulation of replication, transcription, translation, generation of heterogeneous population of persister cells, etc. for survival in different stressful environment during its infection cycle. Toxin-antitoxin (TA) systems are ubiquitous in prokaryotic genomes that enable them to survive in various unfavourable conditions. A toxin protein may inhibit the growth, whereas an antitoxin may neutralize the effect of toxin in different ways. TA systems are involved in stress adaptation, antimicrobial tolerance or resistance, modification in the physiological state of organisms, biofilms formation, growth regulation for survival, plasmid maintenance, anti-phage activities, virulence, and programmed cell death.

Ashutosh Kumar and Anwar Alam have equally contributed to this chapter.

A. Kumar · P. Bharadwaj · S. Tapadar Department of Microbiology, Tripura University (A Central University), Agartala, Tripura, India A. Alam JH Institute of Molecular Medicine, Jamia Hamdard, New Delhi, Delhi, India Inflammation Biology and Cell Signaling Laboratory, ICMR-National Institute of Pathology, Safdarjung Hospital Campus, New Delhi, Delhi, India M. Rani Plant Microbe Interactions Laboratory, National Institute of Plant Genome Research, New Delhi, Delhi, India S. E. Hasnain (*) JH Institute of Molecular Medicine, Jamia Hamdard, New Delhi, Delhi, India Dr Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Hyderabad, India e-mail: [email protected]

# Springer Nature Singapore Pte Ltd. 2019 257 S. E. Hasnain et al. (eds.), Mycobacterium tuberculosis: Molecular Infection Biology, Pathogenesis, Diagnostics and New Interventions, https://doi.org/10.1007/978-981-32-9413-4_15 258 A. Kumar et al.

Environmental microorganisms express a wider repertoire of TA systems as compared to intracellular human pathogens due to a higher probability to encoun- ter different environmental stresses within their ecosystem. However, the pres- ence of high level of TA systems in M.tb is due to the fact that M.tb has to endure several types of stresses including acidic, hypoxic, oxidative, and immune sur- veillance within the host for its survival. TA systems are also present in patho- genic bacteria infecting plants. Based on the mechanism of action, different types of TA systems are classified within the microorganisms. Recently, genes related to type II TA systems have been proposed to be useful in genotyping of tubercu- losis caused by different strains of M.tb.

Keywords Toxin-antitoxin systems · Mycobacterium tuberculosis · Stress survival · Growth regulation · Drug tolerance

Abbreviations

ATP Adenosine triphosphate DATIN Dormancy-associated translation inhibitor DNA Deoxyribonucleic acid DR Direct repeat E. coli Escherichia coli IS Insertion sequence M.tb Mycobacterium tuberculosis mRNA Messenger RNA MTBC M.tb complex PCD Programmed cell death PSK Post-segregational killing SNPs Single-nucleotide polymorphisms sRNA Small regulatory RNAs TA Toxin-antitoxin TAC Toxin-antitoxin-chaperone Vap Virulence-associated protein VNTR Variable number tandem repeats

15.1 Introduction

Toxin-antitoxin (TA) systems in bacteria were first recognized as plasmid-borne loci which help in plasmid maintenance through elimination of daughter cells lacking TA encoding plasmid (Guglielmini and Melderen 2011). A set of linked genes, together encoding a protein ‘poison’ and a corresponding ‘antidote’, forms the TA system (Gerdes 2000). The TA systems present on plasmids make sure that only the 15 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 259 daughter cells inheriting the plasmid survive after cell division. In daughter cells devoid of plasmid, unstable antitoxin is degraded while the stable toxic protein kills the new cell, and this phenomenon is known as ‘post-segregational killing’ (PSK). TA systems are present in multiple copies in prokaryotes (Yamaguchi et al. 2011). Various microbial genome analyses have comprehensively highlighted the diversity in the distribution of TA systems. Previous studies have shown that the genomes of Nitrosomonas europaea, Sinorhizobium meliloti,andMycobacterium bovis contain more than 50 presumptive TA systems, whereas Rickettsia prowazekii, Bacillus subtilis, Campylobacter jejuni, etc. contain no or very few TA systems (Pandey and Gerdes 2005; Sevin and Barloy- Hubler 2007). However, there is little consensus to prove correlation between the number of TA systems and the growth rate of the members within a phylum. Additionally, diversity in the distribution of TA systems among different isolates of the same species is also observed. Detailed study of phylogenetic patterns of TA loci in several prokaryotic genomes suggests presence of multiple TA loci in free-living prokaryotes and few or no TA loci in obligate intracellular prokaryotes (Pandey and Gerdes 2005). TA loci are beneficial to organisms that confront stressful environment. TA systems, also referred to as junk, are considered to be constituents of plasmids and have been retained within the cells due to their addictive nature (Kroll et al. 2010). Some toxins act as general repressors of gene expression, while others are more specificin regulating gene expression (Engelberg-Kulka et al. 2006; Pimentel et al. 2005). Some TA systems that act as bacteriostatic toxins play a key role in growth regula- tion and may restrict growth rather than kill the host cell (Diago-Navarro et al. 2010). ‘Persisters’ are slow-growing population of cells, which can survive stress and later grow into actively dividing cells when the environment is favourable (Kussell et al. 2005). It has been demonstrated that an imbalance between the level of toxin and its antitoxin due to overexpression or mutations in either of them results in high persistence (Fridman et al. 2014; Korch and Hill 2006). Interruption of transcription and translation machinery of host cells due to bacteriophage may also trigger activation of TA system that in turn limits phage replication, termed as antiphage mechanism (Hazan and Engelberg-Kulka 2004). M.tb faces different stresses in its pathogenesis and possesses several proteins such as two component systems, sigma factors, TA systems, acid response, halophilic proteins, etc. for its survival (Kumar et al. 2018). As compared to other mycobacteria, M.tb shows presence of a significant number of TA systems in its genome, which during the state of persistence are induced by active toxins, which may largely contribute towards its pathogenesis (Ramage et al. 2009). Transcriptomic analyses of antibiotic-induced M.tb persisters showed that about 10 TA systems were signifi- cantly upregulated, pointing to the importance of TA system in M.tb persistence (Keren et al. 2011). Interestingly, M.tb possesses abundant number of TA loci, while M. leprae has none, possibly due to the fact that M. leprae has evolved from M.tb through reductive evolution (Ramage et al. 2009; Cole et al. 2001). The presence of TA system provides an evolutionary edge to M.tb in terms of aiding its survival in both extra- and intracellular conditions as compared to M. leprae which can survive 260 A. Kumar et al. only as obligate intracellular pathogen. Similarly, obligate intracellular microorganisms such as Rickettsia and Buchnera spp. have either very few or no TA loci, although several exceptions are also in existence (Pandey and Gerdes 2005; Leplae et al. 2011). Analyses of bacterial gene sequences by Shao et al. (Shao et al. 2011) conclusively point to the presence of TA systems in symbiotic bacteria and overrules previous studies by Pandey et al. that reported absence of TA systems in symbiotic bacteria (Pandey and Gerdes 2005; Shao et al. 2011).

15.2 Classes of Toxin-Antitoxin Systems

The TA loci are classified into different groups such as vapBC, parDE, relBE, ccd, phd/doc, mazEF, and higBA due of differences in mechanism of action (Gerdes et al. 2005). In M.tb, majority of TA systems belong to the class of VapBC (virulence- associated protein) (Gerdes and Maisonneuve 2012). VapC induces dormancy by suppressing translation and induction of vapB transcription which later leads to revival of cells (Winther and Gerdes 2009). It has been shown recently by deletion and overexpression studies that some members of the VapBC TA systems of M.tb are involved in bacteriostasis, morphological changes, growth arrest, and mycobac- terial pathogenesis (Agarwal et al. 2018). M.tb VapBC30 system has been shown to be involved in growth regulation through ribonuclease activity (Deep et al. 2018). M. tb toxin VapC30 inhibits the growth of Escherichia coli (E. coli) when expressed without its cognate antitoxin VapB30. There is no effect of M.tb VapC30 on E. coli when co-expressed with M.tb VapB30. M.tb VapC30 degrades RNA molecules that are magnesium and manganese ion dependent (Lee et al. 2015). M.tb MazF toxin members, when expressed in E. coli or Mycobacterium smegmatis, affect their growth (Gupta 2009; Zhu et al. 2006). The overexpression of MazF3, MazF6, and MazF9 of M.tb in Mycobacterium bovis BCG induces bacteriostasis (Tiwari et al. 2015). M.tb MazF toxins are also involved in the drug tolerance, virulence, and stress adaptation (Tiwari et al. 2015). The RelBE system is among the most characterized TA systems that bind with the A site of the ribosome and affect protein synthesis by cleaving the mRNAs prefera- bly between the second and third nucleotides of the termination codon (Pedersen 2003). In contrast, RelE binds to initial coding region and cleaves the first 100 codons of mRNA and inhibits growth (Hurley et al. 2011). Variety of stress conditions such as nitrosative stress, oxidative stress, and antibiotic stress affect the transcript profiles of RelE toxins of M.tb. The overexpression of toxin RelE of M.tb affected growth of the E. coli and M.tb. The three RelE toxins of M.tb are involved in individual antibiotic specific tolerance (Singh et al. 2010). The YefM antitoxin is highly unstable as it is prone to degradation by Lon, an ATP-dependent serine protease. YefM is co-expressed with the YoeB toxin and the resultant complex so formed consists of dimer of YefM and single molecule of YoeB (Kamada and Hanaoka 2005). Rv3357–Rv3358 of M.tb codes for YefM/YoeB system. 15 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 261

The HigBA family, HigB toxin (Rv1955) and HigA antitoxin (Rv1956), are part of the operon comprising of Rv1954A and Rv1957. Rv1957 is found as a SecB-like chaperone required for antitoxin stabilization. HigB inhibits protein synthesis by cleaving mRNAs that are being translated in E. coli (Smollett et al. 2009; Fivian- Hughes and Davis 2010; Christensen-Dalsgaard et al. 2010; Bordes et al. 2011). The tripartite toxin-antitoxin-chaperone system (TAC) complex is induced during heat shock, hypoxia, nutrient starvation, and persistence. Within the TAC complex, the chaperone directly binds to HigA antitoxin and prevents it aggregation or degradation, thereby aiding in HigA folding and successive interaction with HigB toxin (Bordes et al. 2011). M.tbH37Rv also possesses two ParDE systems, and ParDE2 operon has been investigated recently and was found that toxin MParE2 interacts with GyrB subunit and inhibits bacterial growth by inhibiting DNA gyrase, thereby blocking DNA replication (Gupta et al. 2016).

15.3 Types of Toxin-Antitoxin Systems

TA system is characterized by neutralization of toxin by the antitoxin (Fig. 15.1). In case of a type I TA system, translation of messenger RNA encoding the toxin is inhibited by binding of a non-coding RNA antitoxin to the mRNA. The protein toxin in case of type II TA system is inhibited post-translationally through binding of another protein antitoxin. In case of type III TA systems, a small RNA binds directly to the toxin protein. Type IV-VI TA systems are also reported but are relatively less common. Toxin-antitoxin genes, transferred predominantly through horizontal gene transfer, are mostly associated with pathogenic bacteria and with plasmids confer- ring antibiotic resistance or virulence (Mine et al. 2009; Van Melderen and Saavedra De Bast 2009).

15.3.1 Type I TA System

In this type of TA system, antitoxins consist of small regulatory RNAs (sRNA) comprising of 50–200 nucleotides. A non-coding RNA antitoxin complementarily binds to the toxin-encoding mRNA resulting in mRNA degradation or inhibition of toxin translation (Brielle et al. 2016; Brantl and Jahn 2015). Type I toxin and antitoxins are transcribed from their own promoter, while in other types of TA systems, they are part of operon with other genes. The translation of the type I toxin mRNA is inhibited by base pairing with the antitoxin sRNA that prevents interaction with ribosome resulting in inhibition of translation of toxin mRNA (Brantl 2012; Fozo et al. 2008). Most of the type I toxins are small hydrophobic proteins that create pores in the inner membrane leading to breakdown of membrane potential to stop ATP synthesis, thereby blocking energy-demanding activities such as protein synthesis (Wen et al. 2014; Lee and Lee 2016). symR/symE module of E. coli is considered as an example of type I TA system (Kawano et al. 2007). 262 A. Kumar et al.

Fig. 15.1 Types of toxin-antitoxin systems: Toxin proteins inhibit the growth of cells and there are different mechanisms through which antitoxins neutralize the effect of toxins: (a) Suppression in translation of toxins by complementary RNA complex formation between toxin and antitoxin mRNA. (b) Non-functional toxin-antitoxin protein complex neutralizes the effect of toxins on growth. (c) Non-functional complex formation due to interaction of antitoxin mRNA with toxin proteins. (d) Interaction of antitoxin with target proteins stops toxin activities. (e) Antitoxin protein degrades mRNA of toxin resulting in suppression of toxin protein synthesis. (f) Antitoxins facilitate degradation of toxins by proteinases resulting in rescue of cell growth 15 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 263

15.3.2 Type II TA System

Type II TA systems are most extensively characterized in both prokaryotes and archaea. In type II TA system, the functional activity of toxin proteins is inhibited due to interaction between stable toxin and labile antitoxin. Toxin and antitoxin proteins are expressed simultaneously as both are organized into operons. The activity of different toxin protein regulates various mechanisms within the cell. For example, CcdB protein of E. coli (strain K12) inhibit the function of DNA gyrase by inactivating DNA topoisomerase II, whereas MazF cleaves cellular mRNAs for the inhibition of protein synthesis (Bernard and Couturier 1992; Zhang et al. 2003). The TA complex acts as a repressor for TA operon system as it binds to the palindromic sequence of the promoter region. Due to antitoxin degradation, the concentration of TA complex reduces thereby leading to production of more toxins and antitoxin.

15.3.3 Type III TA System

Type III TA system is characterized by direct interaction between a toxic protein and RNA antitoxin. The RNA gene involved totally neutralizes the effects of the toxic protein. ToxI-ToxN TA system from a plant pathogen named Erwinia carotovora subspecies atrosepticum (Pectobacterium atrosepticum) is a perfect example of type III TA system. The function of toxin protein (ToxN) is directly suppressed by interaction with antitoxin RNA, forming aToxN-RNA (an RNA antitoxin) complex (Fineran et al. 2009; Blower et al. 2012).

15.3.4 Type IV TA System

In type IV TA system, there is no direct interaction between toxin and antitoxin proteins. Antitoxin protein interacts with the target of toxin protein, thereby suppressing the activity of toxin on its target. The functional aspect of type IV TA system is exemplified in case of toxin CbeA of E. coli (1303) that prevents polymerization of the cytoskeletal proteins (MreB and FtsZ) and inhibits cell division. The antitoxin CbeA (YeeU) protein inhibits by binding directly with the target, namely, MreB and FtsZ toxin, instead of forming a toxin-antitoxin complex (Masuda et al. 2012).

15.3.5 Type V TA System

In the type V TA system, antitoxin protein retards synthesis of toxin protein by degrading the mRNA transcribed to code toxin protein. This is exemplified by GhoT 264 A. Kumar et al. of E. coli (strain K12), a protein that induces persistence and ghost cell formation with damaged membrane. The antitoxin GhoS is an endoribonuclease and precisely cleaves the mRNA encoding for membrane-lytic peptide toxin GhoT (Wang et al. 2013).

15.3.6 Type VI TA System

In the type VI TA system, antitoxin protein facilitates the proteolytic degradation of toxin protein. There is no degradation of toxin protein by proteases in the absence of antitoxin, and if toxin-antitoxin proteins are together, toxin degradation occurs. In SocAB TA system in Caulobacter crescentus, SocB protein inhibits DNA elonga- tion by intercalating with DnaN, whereas antitoxin SocA facilitates degradation of SocB in the presence of protease ClpXP (Aakre et al. 2013).

15.4 Biological Roles of Toxin-Antitoxin Systems

TA systems regulate bacterial survival in different types of unfavourable conditions. TA systems are involved in several biological functions such as growth regulation, physiological changes of the cells, programmed cell death, etc. (Fig. 15.2). A more detailed importance of the TA systems in various physiological conditions is described below:

Fig. 15.2 Roles and applications of toxin-antitoxin systems: Toxin-antitoxin systems play impor- tant roles such as plasmid maintenance, drug resistance/tolerance, biofilm formation, growth regulation, antiphage activity, genotyping, programmed cell death, etc. 15 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 265

15.4.1 Stress Survival

TA systems act on regulatory machinery that control mechanisms critical for sur- vival of bacteria (Engelberg-Kulka et al. 2006; Yamaguchi et al. 2011). For example, TA systems are important for adaptation of M.tb to unfavourable environmental conditions inside the host and maybe required for triggering a non-replicating state (Lewis 2007). Mycobacteria often adopt a non-replicative persistent, inactive state, to avoid unfavourable stress conditions like hypoxia, oxidative stress, nutritional limitations, acidic pH, etc., within the host macrophages (Wu et al. 2012). There are 88 putative TA systems present in M.tbH37Rv that are conserved in M.tb complex (MTBC) but are few in other non-pathogenic mycobacteria, indicating a potential contribution to the pathogenic lifestyle of M.tb (Ramage et al. 2009). In response to starvation, several toxins within the cells are upregulated leading to inhibition of translation and selective degradation of mRNA. Contrary to starvation which leads to generalized upregulation of TA loci, low pH exposure leads to downregulation of few TA genes (Gupta et al. 2017).

15.4.2 Persistence

In majority of bacteria, there are set of genes that are involved in growth inhibition, and their overexpression may result in cell death, similar to programmed cell death of eukaryotes (Wen et al. 2014). It has been reported that TA systems are important for persister phenotype in E. coli (Tsilibaris et al. 2007). TA systems present in M.tb may regulate cell division during infection (Warner and Mizrahi 2006). After infection, M.tb initially grows and then acquires latency, a state of non-replicating cells in unfavourable conditions that can survive long periods with the potential to reactivate itself later whenever environment is favourable (Stewart et al. 2003, North and Jung 2004). In the case of persistence, a subpopulation of non-replicating bacteria becomes tolerant to antibiotics (Gomez and McKinney 2004). The presence of TA system in M.tb, as in other bacteria, imparts antibiotic tolerance and is one of the main reasons why long term of antibiotic therapy is required to cure tuberculosis (Keren et al. 2004; Ramage et al. 2009). It has been reported that mycobacterial MazF ribonucleases are involved in the drug tolerance, adaptation in oxidative stress, and nutrient depletion and virulence (Tiwari et al. 2015). M.tb RelE toxins are involved in formation of persisters specific to individual antibiotic and involved in drug-specific tolerance (Singh et al. 2010).

15.4.3 Biofilms

Planktonic bacteria can aggregate and attach themselves on biotic or abiotic surfaces to form biofilms. Formation of biofilms by pathogens is considered to be one of the 266 A. Kumar et al. main survival strategies to counter the host defence (Rybtke et al. 2011; Kumar et al. 2017a). TA systems have been shown to be involved in biofilm formation, with exceptions. It has been shown that type II TA system of E. coli involving MqsR protein is induced during biofilm formation and deletion of this gene resulted in the absence of biofilm formation (Kasari et al. 2010). The role of mqsRA TA system in biofilm formation is associated with motility and as autoinducer-2 quorum sensing system (Gonzalez Barrios et al. 2006). The yefM-yoeB and relBE TA systems of Streptococcus pneumonia are involved in the biofilm formation. It has been reported that mutant strains lacking yefM-yoeB or both yefM-yoeB and relBE show reduc- tion in biofilm formation (Chan et al. 2018). It has also been shown that deletion of mutants of mazF and relE gene homologue did not affect biofilm formation in Streptococcus mutans (Lemos et al. 2005).

15.4.4 Antiphage Activity

The competitive edge of pathogen against the host is driven by the enormous diversity and multiplicity of TA systems. Bacteria have evolved several defence mechanisms to protect themselves and survive against the onslaught of phage infection. Bacteria exhibit a wide array of mechanisms to resist bacteriophages that include inhibition of adsorption, exclusion of superinfection, cleavage of nucleic acids of the phages through restriction-modification or CRISPR-Cas systems, and abortive infection (Stern and Sorek 2012;Seed2015). The abortive infection systems trigger premature death of phage-infected bacteria and restrict the phage to replicate or spread, thereby protecting the uninfected bacterial population within the niche. Thus, there is a link between abortive infection and TA systems. Several TA systems have anti-phage activity including Hok-Sok, LsoAB and MazEF, ToxIN, and AbiEG (Short et al. 2018).

15.4.5 Bacterial Virulence and Pathogenecity

Presence of TA modules in the genome is directly related to the virulence of bacteria (Georgiades and Raoult 2011), for example, type I toxins are involved in lysis of host cell. The PepA1 toxin of S. aureus is a pore-forming peptide that causes bacterial cell death. When there is oxidative stress inside host cell, the PepA1 toxin is typically released from its SprA antitoxin. This is an example of altruistic behaviour, as the peptide drives erythrocyte lysis, resulting in release of slowly dividing cells that escape the immune system (Sayed et al. 2012). Deletion of VapBC homologues in Haemophilus influenzae results in remarkable decrease of virulence in animal models for otitis media, tissue, etc. (Ren et al. 2012). 15 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 267

15.4.6 Growth Regulation

M.tb encounters different types of unfavourable conditions during infection. To survive in stressful conditions, bacteria regulate its growth. There are several proteins that are involved in growth regulation other than TA systems such as DATIN, IciA, MSMEG_1878 (M.tbRv3241c orthologue in M. smegmatis), etc. M. tb DATIN and MSMEG_1878 inhibit protein synthesis by interacting with ribosome (Kumar et al. 2012; Li et al. 2018). M.tb IciA inhibits bacterial growth by inhibiting the opening of two strands of DNA during replication (Kumar et al. 2009; Kumar et al. 2017b). Toxin-antitoxin systems are activated during starvation stress. The RNase toxins, instead of being bactericidal, are usually bacteriostatic in nature. There is quick arrest of growth in response to starvation or other environmental stresses which help in their survival, and quicker resumption of growth occurs when the situation improves (Gerdes 2000; Gerdes et al. 2005).

15.4.7 Programmed Cell Death

Programmed cell death (PCD or apoptosis) is a physiological process and occurs mainly in multicellular, eukaryotic organisms during the process of embryonic development or tissue turnover. Dysregulation of PCD results in diseases like tumour formation, autoimmune diseases, or lysosomal disorders (Hayes 2003). Bacteria, being unicellular, do undergo PCD. However in natural environment, bacteria exist as biofilms that represent multicellular colonies and display coordina- tion as in multicellular organisms. Such immobilized bacteria maintain discrete and ordered spatial structures within the biofilm niche. There are several genes in bacteria that are homologues to eukaryotic genes involved in PCD (Koonin and Aravind 2002), and TA modules of E. coli are either PCD genes or mediators of reversible growth arrest, which alternatively might allow the cells to enter a dormant or a semi-dormant state. It has been shown that PCD in bacteria might allow surviving cells to scavenge nutrients from dead ones and may prevent spread of bacteriophages (Engelberg-Kulka and Glaser 1999).

15.5 Applications of Toxin-Antitoxin Systems

There is a growing need for new antimicrobial agents due to decrease in the effectivity of drugs being used currently arising out of increase in multi-drug resistance. The toxins of TA systems usually target various biological processes such as replication, transcription, translation, macromolecular synthesis, cell wall synthesis, phage infection, and cytoskeletal polymerization. The fact that some of these toxins also overlap with the targets of the antibiotics (Wen et al. 2014) provides an option to explore TA systems as drug target in bacteria. There are several antibiotics that act indirectly against the TA systems. The detailed study of the interaction between the toxin and the antitoxin may help in the formulation of new 268 A. Kumar et al. drugs. Mapping the precise location of different TA systems in bacterial chromo- some or plasmid will uncover fundamental insight into their possible applications as drug targets. Previously, IS-elements, DR-elements, variable number tandem repeats (VNTR), and single-nucleotide polymorphisms (SNPs) were used as genetic markers in housekeeping genes or other genes and were used for genotyping. The type II TA systems are involved in virulence, persistence, and survival of M.tb inside host macrophages (Zaychikova et al. 2015). Analyses of 173 sequenced genomes of M. tb for the genes of type II TA systems show genetic diversity (SNPs) that correlates with the specific genotype of M.tb strains (Zaychikova et al. 2015). This correlation between a genotype of particular strain and SNPs in different genes of type II TA system paved way to consider TA systems as a new biomarker for genotyping of tuberculosis caused by different strains of M.tb.

15.6 Conclusion

Genome sequencing and its annotation have suggested presence of significant number of TA systems in microorganisms. The classification of TA system is based on similarity of primary sequences and on the specificity of interaction between pair of toxin and antitoxin molecules. Type II system is structurally most flexible in terms of location of antitoxin gene either upstream or downstream of the toxin gene or regulation of repression activities that is encoded by different genes. The various types of TA system may shuffle among themselves in terms of function and can exhibit ‘mix and match’ phenomenon. This is exemplified by the structural similarity between GinI solitary toxins with type II HicA toxin. In some cases, toxins can also evolve from the antitoxin under selective pressure. As in the case of VapD sequences, which mediate defence against phages, TA system points to a common origin with CRISPR-cas system. There is consensus that the evolution of TA system in bacteria enabled these unicellular organisms into robust molecular machines that could withstand the onslaught of environment and various stress. The importance of TA system is also underlined by the fact that while many pathogenic and obligatory intracellular pathogens opted for reductive evolution in genome size and shed off the extra burden of many genes, yet they retained the genes associated with TA system. The distribution of TA system varies even among strains of the same species of bacteria. TA systems are highly mobile in nature and moves between genomes through horizontal gene transfer. The presence of TA genes within the transposons makes them refractory to gene efflux and stabilizes the TA system within the cells. The additive property of TA genes within the transposons facilitates increased stability and exclusion of foreign DNA. Although the activity of TA system may be subdued by host cell machinery, it may rescue its activity similar to the restriction modification system. TA systems are preferentially associated with genomic islands or plasmids which serve as a mechanism to maintain their structural integrity during stress and survival through post-segregational killing. M.tb possesses a huge number of TA systems that are usually located within distinct genomic islands and trigger 15 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 269 decrease in metabolic activity. In case the TA genes are integrated within the core genome, these accumulate mutation that may lead to loss of addictive property or deletion of TA system. In case of E. coli, type I hok-sok system are inactivated by IS sequences, gene rearrangements, and point mutations. TA system can evolve through integration with the host regulatory machinery and may replace the antitoxin molecules with signal transduction molecules. Toxins have many cellular functions including inhibition of protein synthesis, DNA replication, and synthesis of cell wall in response to unfavourable conditions. Some toxins in TA system act as ribonucleases, while some other toxins act as gyrase inhibitors and kinases. The TA genes express in different stressed conditions such as nutrient deficiency, antibiotic treatment, bacteriophage infection, host immune responses, oxidative stress, and high temperature. They are involved in persistence, slow cell growth, cell cycle arrest, or cell death. Proteins involved in TA systems may act as important targets for drug development that may help in reduction of treatment duration of tuberculosis and other infectious diseases. In spite of the diversity in structure of TA system, the function of the TA system is tightly regulated by other cellular networks such that the prolific activity of TA system is activated only as a response to cellular physiology and the toxins are unleashed for minimal activity. The activity of TA locus that regulate activation of signaling pathways involved in persister cell formation in biofilms are modules acting as effectors of persister cell formation. These are usually deeply integrated into cellular signaling pathways that tightly control their activation and use their characteristic auto-regulatory features to tune the induction, duration, and intensity of the phenotypic switch into dormancy. It is speculated that our current understanding of the toxin-antitoxin system is still redundant and a deeper understanding of the mechanistic significance of TA system will enable us to unravel the mysteries of how these unicellular organisms could dominate the environment. It is envisaged that unlocking the mechanism of TA system could allow us to shape better strategies for overcoming the harmful effects of TA system in clinical pathogenesis and in dealing with microbial drug tolerance.

Funding This project has been funded by “UGC-BSR Research Start-Up-Grant project No. F. 30-487/2019(BSR) sanctioned to Ashutosh Kumar”

References

Aakre CD, Phung TN, Huang D, Laub MT (2013) A bacterial toxin inhibits DNA replication elongation through a direct interaction with the β sliding clamp. Mol Cell 52(5):617–628. https://doi.org/10.1016/j.molcel.2013.10.014 Agarwal S, Tiwari P, Deep A, Kidwai S, Gupta S, Thakur KG, Singh R (2018) System-wide analysis unravels the differential regulation and in vivo essentiality of virulence-associated proteins B and C toxin-antitoxin Systems of Mycobacterium tuberculosis. J Infect Dis 217 (11):1809–1820. https://doi.org/10.1093/infdis/jiy109 270 A. Kumar et al.

Bernard P, Couturier M (1992) Cell killing by the F plasmid CcdB protein involves poisoning of DNA-topoisomerase II complexes. J Mol Biol 226(3):735–745. https://doi.org/10.1016/0022- 2836(92)90629-X Blower TR, Short FL, Rao F, Mizuguchi K, PeiXY FPC, Luisi BF, Salmond GPC (2012) Identifi- cation and classification of bacterial type III toxin–antitoxin systems encoded in chromosomal and plasmid genomes. Nucleic Acids Res 40:6158–6173 Bordes P, Cirinesi AM, Ummels R, SalaA SS, Bitter W, Genevaux P (2011) SecB like chaperone controls a toxin–antitoxin stress-responsive system in Mycobacterium tuberculosis. Proc Natl Acad Sci U S A 108:8438–8443 Brantl S (2012) Bacterial type I toxin-antitoxin systems. RNA Biol 9:1488–1490. https://doi.org/ 10.4161/rna.23045 Brantl S, Jahn N (2015) sRNAs in bacterial type I and type III toxin–antitoxin systems. FEMS Microbiol Rev 39:413–427 Brielle R, Pinel-Marie ML, Felden B (2016) Linking bacterial type I toxins with their actions. Curr Opin Microbiol 30:114–121 Chan WT, Domenech M, Moreno-Córdoba I, Navarro-Martínez V, Nieto C, Moscoso M, García E, Espinosa M (2018) The Streptococcus pneumoniae yefM-yoeB and relBE toxin-antitoxin operons participate in oxidative stress and biofilm formation. Toxins 10(9):E378. https://doi. org/10.3390/toxins10090378 Christensen-Dalsgaard M, Jorgensen MG, Gerdes K (2010) Three new RelE-homologous mRNA interferases of Escherichia coli differentially induced by environmental stresses. Mol Microbiol 75:333–348 Cole ST, Eiglmeier K, Parkhill J, James KD, Thomson NR (2001) Massive gene decay in the leprosy bacillus. Nature 409:1007–1011 Deep A, Tiwari P, Agarwal S, Kaundal S, Kidwai S, Singh R, Thakur KG (2018) Structural, functional and biological insights into the role of Mycobacterium tuberculosis VapBC11 toxin– antitoxin system: targeting a tRNase to tackle mycobacterial adaptation. Nucleic Acids Res 46 (21):11639–11655. https://doi.org/10.1093/nar/gky924 Diago-Navarro E, Hernandez-Arriaga AM, López-Villarejo J, Muñoz-Gómez AJ, Kamphuis MB, Boelens R, Lemonnier M, Díaz-Orejas R (2010) parD toxin-antitoxin system of plasmid R1 – basic contributions, biotechnological applications and relationships with closely-related toxin- antitoxin systems. FEBS J 277(15):3097–3117. https://doi.org/10.1111/j.1742-4658.2010. 07722.x Engelberg-Kulka H, Glaser G (1999) Addiction modules and programmed cell death and antideath in bacterial cultures. Annu Rev Microbiol 53:43–70 Engelberg-Kulka H, Amitai S, Kolodkin-Gal I, Hazan R (2006) Bacterial programmed cell death and multicellular behavior in bacteria. PLoS Genet 2(10):e135. https://doi.org/10.1371/journal. pgen.0020135 Fineran PC, Blower TR, Foulds IJ, Humphreys DP, Lilley KS, Salmond GP (2009) The phage abortive infection system, ToxIN, functions as a protein-RNA toxin-antitoxin pair. Proc Natl Acad Sci U S A 106:894–899 Fivian-Hughes AS, Davis EO (2010) Analyzing the regulatory role of the HigA antitoxin within Mycobacterium tuberculosis. J Bacteriol 192:4348–4356 Fozo EM, Hemm MR, Storz G (2008) Small toxic proteins and the antisense RNAs that repress them. Microbiol Mol Biol Rev 72:579–589. https://doi.org/10.1128/MMBR.00025-08 Fridman O, Goldberg A, Ronin I, Shoresh N, Balaban NQ (2014) Optimization of lag time underlies antibiotic tolerance in evolved bacterial populations. Nature 513(7518):418–421. https://doi.org/10.1038/nature13469 Georgiades K, Raoult D (2011) Genomes of the most dangerous epidemic bacteria have a virulence repertoire characterized by fewer genes but more toxin–antitoxin modules. PLoS One 6:e17962 15 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 271

Gerdes K (2000) Toxin-antitoxin modules may regulate synthesis of macromolecules during nutritional stress. J Bacteriol 182(3):561–572. https://doi.org/10.1128/JB.182.3.561-572.2000 Gerdes K, Maisonneuve E (2012) Bacterial persistence and toxin-antitoxin loci. Annu Rev Microbiol 66:103–123 Gerdes K, Christensen SK, Lobner-Olesen A (2005) Prokaryotic toxin-antitoxin stress response loci. Nat Rev Microbiol 3:371–382 Gomez JE, McKinney JD (2004) M. tuberculosis persistence, latency, and drug tolerance. Tuber- culosis (Edinb) 84:29–44 Gonzalez Barrios AF, Zuo R, Hashimoto Y, Yang L, Bentley WE, Wood TK (2006) Autoinducer 2 controls biofilm formation in Escherichia coli through a novel motility quorum-sensing regulator (MqsR, B3022). J Bacteriol 188(1):305–316. https://doi.org/10.1128/JB.188.1.305- 316.2006 Guglielmini J, Melderen LV (2011) Bacterial toxin-antitoxin systems- translation inhibitors every- where. Mob Genet Elem 1(4):283–290. https://doi.org/10.4161/mge.18477 Gupta A (2009) Killing activity and rescue function of genome-wide toxin-antitoxin loci of Mycobacterium tuberculosis. FEMS Microbiol Lett 290:45–53 Gupta M, Nayyar N, Chawla M, Sitaraman R, Bhatnagar R, Banerjee N (2016) The chromosomal parDE2 toxin–antitoxin system of Mycobacterium tuberculosis H37Rv: genetic and functional characterization. Front Microbiol 7:886. https://doi.org/10.3389/fmicb.2016.00886 Gupta A, Venkataraman B, Vasudevan M, Gopinath-Bankar K (2017) Co-expression network analysis of toxin-antitoxin loci in Mycobacterium tuberculosis reveals key modulators of cellular stress. Sci Rep 7(1):5868. https://doi.org/10.1038/s41598-017-06003-7 Hayes F (2003) Toxins-antitoxins: plasmid maintenance, programmed cell death and cell cycle arrest. Science 301(5639):1496–1499 Hazan R, Engelberg-Kulka H (2004) Escherichia coli mazEF-mediated cell death as a defense mechanism that inhibits the spread of phage P1. Mol Gen Genomics 272(2):227–234. https:// doi.org/10.1007/s00438-004-1048-y Hurley JM, Cruz JW, Ouyang M, Woychik NA (2011) Bacterial toxin RelE mediates frequent codon-independent mRNA cleavage from the 50 end of coding regions in vivo. J Biol Chem 286 (17):14770–14778. https://doi.org/10.1074/jbc.M110.108969 Kamada K, Hanaoka F (2005) Conformational change in the catalytic site of the ribonuclease YoeB toxin by YefM antitoxin. Mol Cell 19:497–509 Kasari V, Kurg K, Margus T, Tenson T, Kaldalu N (2010) The Escherichia coli mqsR and ygiT genes encode a new toxin-antitoxin pair. J Bacteriol 192(11):2908–2919. https://doi.org/10. 1128/JB.01266-09 Kawano M, Aravind L, Storz G (2007) An antisense RNA controls synthesis of an SOS-induced toxin evolved from an antitoxin. Mol Microbiol 64:738–754. https://doi.org/10.1111/j.1365- 2958.2007.05688.x Keren I, Shah D, Spoering A, Kaldalu N, Lewis K (2004) Specialized persister cells and the mechanism of multidrug tolerance in Escherichia coli. J Bacteriol 186:8172–8180 Keren I, Minami S, Rubin E, Lewis K (2011) Characterization and transcriptome analysis of Mycobacterium tuberculosis persisters. MBio 2(3):e00100–e00111. https://doi.org/10.1128/ mBio.00100-11 Koonin EV, Aravind L (2002) Origin and evolution of eukaryotic apoptosis: the bacterial connec- tion. Cell Death Differ 9(4):394–404 Korch SB, Hill TM (2006) Ectopic overexpression of wild-type and mutant hipA genes in Escherichia coli: effects on macromolecular synthesis and persister formation. J Bacteriol 188 (11):3826–3836. https://doi.org/10.1128/JB.01740-05 Kroll J, Klinter S, Schneider C, Voß I, Steinbüchel A (2010) Plasmid addiction systems: perspectives and applications in biotechnology. Microb Biotechnol 3(6):634–657. https://doi. org/10.1111/j.1751-7915.2010.00170.x 272 A. Kumar et al.

Kumar S, Farhana A, Hasnain SE (2009) In-vitro helix opening of M. tuberculosis oriC by DnaA occurs at precise location and is inhibited by IciA like protein. PLoS One 4(1):e4139. https://doi. org/10.1371/journal.pone.0004139 Kumar A, Majid M, Kunisch R, Rani PS, Qureshi IA, Lewin A (2012) Mycobacterium tuberculosis DosR regulon gene Rv0079 encodes a putative, ‘Dormancy Associated Translation Inhibitor (DATIN)’. PLoS One 7(6):e38709. https://doi.org/10.1371/journal.pone.0038709 Kumar A, Alam A, Rani M, Ehtesham NZ, Hasnain SE (2017a) Biofilms: survival and defense strategy for pathogens. Int J Med Microbiol 307(8):481–489. https://doi.org/10.1016/j.ijmm. 2017.09.016 Kumar A, Rani M, Ehtesham NZ, Hasnain SE (2017b) Commentary: modification of host responses by mycobacteria. Front Immunol 8:466. https://doi.org/10.3389/fimmu.2017.00466. eCollection 2017 Kumar A, Alam A, Tripathi D, Rani M, Khatoon H, Pandey S, Ehtesham NZ, Hasnain SE (2018) Protein adaptations in extremophiles: an insight into extremophilic connection of mycobacterial proteome. Semin Cell Dev Biol 84:147–157. https://doi.org/10.1016/j.semcdb.2018.01.003 Kussell E, Kishony R, Balaban NQ, Leibler S (2005) Bacterial persistence: a model of survival in changing environments. Genetics 169(4):1807–1814. https://doi.org/10.1534/genetics.104. 035352 Lee KY, Lee BJ (2016) Structure, biology, and therapeutic application of toxin–antitoxin systems in pathogenic bacteria. Toxins 8(10):305. https://doi.org/10.3390/toxins8100305 Lee IG, Lee SJ, Chae S, Lee KY, Kim JH, Lee BJ (2015) Structural and functional studies of the Mycobacterium tuberculosis VapBC30 toxin-antitoxin system: implications for the design of novel antimicrobial peptides. Nucleic Acids Res 43(15):7624–7637. https://doi.org/10.1093/ nar/gkv689 Lemos JA, Brown TA Jr, Abranches J, Burne RA (2005) Characteristics of Streptococcus mutans strains lacking the MazEF and RelBE toxin-antitoxin modules. FEMS Microbiol Lett 253 (2):251–257. https://doi.org/10.1016/j.femsle.2005.09.045 Leplae R, Geeraerts D, Hallez R, Guglielmini J, Dreze P, Van ML (2011) Diversity of bacterial type II toxin-antitoxin systems: a comprehensive search and functional analysis of novel families. Nucleic Acids Res 39:5513–5525 Lewis K (2007) Persister cells, dormancy and infectious disease. Nat Rev Microbiol 5:48–56. https://doi.org/10.1038/nrmicro1557 Li Y, Sharma MR, Koripella RK, Yang Y, Kaushal PS, Lin Q, Wade JT, Gray TA, Derbyshire KM, Agrawal RK, Ojha AK (2018) Zinc depletion induces ribosome hibernation in mycobacteria. Proc Natl Acad Sci 115(32):8191–8196. https://doi.org/10.1073/pnas.1804555115 Masuda H, Tan Q, Awano N, Wu KP, Inouye M (2012) YeeU enhances the bundling of cytoskele- tal polymers of MreB and FtsZ, antagonizing the CbtA (YeeV) toxicity in Escherichia coli. Mol Microbiol 84:979–989 Mine N, Guglielmini J, Wilbaux M, Van Melderen L (2009) The decay of the chromosomally encoded ccdO157 toxin-antitoxin system in the Escherichia coli species. Genetics 181 (4):1557–1566. https://doi.org/10.1534/genetics.108.095190 North RJ, Jung YJ (2004) Immunity to tuberculosis. Annu Rev Immunol 22:599–623 Pandey DP, Gerdes K (2005) Toxin-antitoxin loci are highly abundant in free-living but lost from host-associated prokaryotes. Nucleic Acids Res 33(3):966–976. https://doi.org/10.1093/nar/ gki201 Pedersen K (2003) The bacterial toxin RelE displays codon-specific cleavage of mRNAs in the ribosomal A site. Cell 112:131–140 Pimentel B, Madine MA, de la Cueva-Méndez G (2005) Kid cleaves specific mRNAs at UUACU sites to rescue the copy number of plasmid R1. EMBO J 24(19):3459–3469. https://doi.org/10. 1038/sj.emboj.7600815 Ramage HR, Connolly LE, Cox JS (2009) Comprehensive functional analysis of Mycobacterium tuberculosis toxin-antitoxin systems: implications for pathogenesis, stress responses, and evo- lution. PLoS Genet 5:e1000767 15 Toxin-Antitoxin (TA) Systems in Stress Survival and Pathogenesis 273

Ren D, Walker AN, Daines D (2012) Toxin–antitoxin loci vapBC-1 and vapXD contribute to survival and virulence in nontypeable Haemophilus influenza. BMC Microbiol 12:263 Rybtke MT, Jensen PO, Hoiby N, Givskov M, Tolker-Nielsen T, Bjarnsholt T (2011) The implication of Pseudomonas aeruginosa biofilms in infections. Inflamm Allergy Drug Targets 10:141–157 Sayed N, Nonin-Lecomte S, Réty S, Felden B (2012) Functional and structural insights of a Staphylococcus aureus apoptotic-like membrane peptide from a toxin–antitoxin module. Biol Chem 287:43454–43463 Seed KD (2015) Battling phages: how bacteria defend against viral attack. PLoS Pathog 11(6): e1004847 Sevin EW, Barloy-Hubler F (2007) RASTA-Bacteria: a web-based tool for identifying toxin- antitoxin loci in prokaryotes. Genome Biol 8:R155 Shao Y, Harrison EM, Bi D, Tai C, He X (2011) TADB: a web-based resource for type 2 toxin antitoxin loci in bacteria and archaea. Nucleic Acids Res 39:D606–D611 Short FL, Akusobi C, Broadhurst WR, Salmond GPC (2018) The bacterial type III toxin-antitoxin system, ToxIN, is a dynamic protein-RNA complex with stability-dependent antiviral abortive infection activity. Sci Rep 8:1013. https://doi.org/10.1038/s41598-017-18696-x Singh R, Barry CE, Boshoff HI (2010) The three RelE homologs of Mycobacterium tuberculosis have individual, drug-specific effects on bacterial antibiotic tolerance. J Bacteriol 192:1279–1291 Smollett KL, Fivian-Hughes AS, Smith JE, Chang A, Rao T, Davis EO (2009) Experimental determination of translational start sites resolves uncertainties in genomic open reading frame predictions – application to Mycobacterium tuberculosis. Microbiology 155:186–197 Stern A, Sorek R (2012) The phage-host arms-race: shaping the evolution of microbes. BioEssays 33:43–51 Stewart GR, Robertson BD, Young DB (2003) Tuberculosis: a problem with persistence. Nat Rev Microbiol 1:97–105 Tiwari P, Arora G, Singh M, Kidwai S, Narayan OP, Singh R (2015) MazF ribonucleases promote Mycobacterium tuberculosis drug tolerance and virulence in Guinea pigs. Nat Commun 6:6059. https://doi.org/10.1038/ncomms7059 Tsilibaris V, Maenhaut-Michel G, Mine N, Melderen LV (2007) What is the benefit to Escherichia coli of having multiple toxin-antitoxin Systems in its Genome? J Bacteriol 189(17):6101–6108. https://doi.org/10.1128/JB.00527-07 Van Melderen L, Saavedra De Bast M (2009) Bacterial toxin-antitoxin systems: more than selfish entities? PLoS Genet 5(3):e1000437. https://doi.org/10.1371/journal.pgen.1000437 Wang X, Lord DM, Hong SH, Peti W, Benedik MJ, Page R, Wood TK (2013) Type II toxin/ antitoxin MqsR/MqsA controls type V toxin/antitoxin GhoT/GhoS. Environ Microbiol 15 (6):1734–1744. https://doi.org/10.1111/1462-2920.12063 Warner DF, Mizrahi V (2006) Tuberculosis chemotherapy: the influence of bacillary stress and damage response pathways on drug efficacy. Clin Microbiol Rev 19:558–570 Wen Y, Behiels E, Devreese B (2014) Toxin–antitoxin systems: their role in persistence, biofilm formation, and pathogenicity. Pathog Dis 70:240–249. https://doi.org/10.1111/2049-632X. 12145 Winther KS, Gerdes K (2009) Ectopic production of VapCs from Enterobacteria inhibits translation and trans-activates YoeB mRNA interferase. Mol Microbiol 72:918–930 Wu Y, Vulić M, Keren I, Lewis K (2012) Role of oxidative stress in persister tolerance. Antimicrob Agents Chemother 56:4922–4926. https://doi.org/10.1128/AAC.00921-12 Yamaguchi Y, Park JH, Inouye M (2011) Toxin-antitoxin systems in bacteria and archaea. Annu Rev Genet 45:61–79. https://doi.org/10.1146/annurev-genet-110410-132412 Zaychikova MV, Zakharevich NV, Sagaidak MO, Bogolubova NA, Smirnova TG, Andreevskaya SN, Larionova EE, Alekseeva MG, Chernousova LN, Danilenko VN (2015) Mycobacterium tuberculosis type II toxin antitoxin systems: genetic polymorphisms and functional properties 274 A. Kumar et al.

and the possibility of their use for genotyping. PLoS One 10(12):e0143682. https://doi.org/10. 1371/journal.pone.0143682 Zhang Y, Zhang J, Hoeflich KP, Ikura M, Qing G, Inouye M (2003) MazF cleaves cellular mRNAs specifically at ACA to block protein synthesis in Escherichia coli. Mol Cell 12(4):913–923. https://doi.org/10.1016/S1097-2765(03)00402-7 Zhu L, Zhang Y, Teh JS, Zhang J, Connell N, Rubin H, Inouye M (2006) Characterization of mRNA interferases from Mycobacterium tuberculosis. J Biol Chem 281:18638–18643 www.nature.com/npjbiofilms

ARTICLE OPEN Peptidyl-prolyl isomerase-B is involved in Mycobacterium tuberculosis biofilm formation and a generic target for drug repurposing-based intervention

Ashutosh Kumar1,10, Anwar Alam1,2, Sonam Grover1, Saurabh Pandey3,11, Deeksha Tripathi2,4, Monika Kumari5, Mamta Rani6, Aditi Singh7, Yusuf Akhter 8, Nasreen Z. Ehtesham3 and Seyed E. Hasnain1,2,9

Tuberculosis (TB), a disease caused by Mycobacterium tuberculosis (M.tb), takes one human life every 15 s globally. Disease relapse occurs due to incomplete clearance of the pathogen and reactivation of the antibiotic tolerant bacilli. M.tb, like other bacterial pathogens, creates an ecosystem of biofilm formed by several proteins including the cyclophilins. We show that the M.tb cyclophilin peptidyl-prolyl isomerase (PpiB), an essential gene, is involved in biofilm formation and tolerance to anti-mycobacterial drugs. We predicted interaction between PpiB and US FDA approved drugs (cyclosporine-A and acarbose) by in-silico docking studies and this was confirmed by surface plasmon resonance (SPR) spectroscopy. While all these drugs inhibited growth of Mycobacterium smegmatis (M.smegmatis) when cultured in vitro, acarbose and cyclosporine-A showed bacteriostatic effect while gallium nanoparticle (GaNP) exhibited bactericidal effect. Cyclosporine-A and GaNP additionally disrupted M.tb H37Rv biofilm formation. Co-culturing M.tb in their presence resulted in significant (2–4 fold) decrease in dosage of anti-tubercular drugs- isoniazid and ethambutol. Comparison of the cyclosporine-A and acarbose binding sites in PpiB homologues of other biofilm forming infectious pathogens revealed that these have largely remained unaltered across bacterial species. Targeting bacterial biofilms could be a generic strategy for intervention against bacterial pathogens. npj Biofilms and Microbiomes (2019) 5:3 ; https://doi.org/10.1038/s41522-018-0075-0

INTRODUCTION mental illness, diabetes, malaria etc. target cellular pathways Biofilm associated diseases cause nearly 80% of the recalcitrant which are also utilised by M.tb for survival. Drug repurposing offers 3 hospital infections.1 Several non-pathogenic and pathogenic a viable option to fast track new therapies against other diseases. species of microorganism including mycobacteria make biofilm Previous studies showed the involvement of biofilm formation as one of the generic mechanisms to overcome stress. The matrix by Mycobacterium abscessus and Pseudomonas aeruginosa in cystic of the biofilm is composed of extracellular components consisting fibrosis and also as a virulence determinant in uropathogenic 4–6 of biopolymers that are essentially secreted by the microorgan- Escherichia coli isolates. The presence of extracellular M.tb isms and act as a physical barrier to drugs or against immune within biofilm like structure inside lung lesions of M.tb infected surveillance. In addition to the rapid emergence of drug resistance guinea pigs undergoing antibiotic treatment points to the in several strains of mycobacteria, the growing menace of drug possibility of biofilm formation within the host tissues.7 Peptidyl- tolerance has led to the requirement of higher doses of drugs for prolyl isomerases (PPIase), popularly known as cyclophilins, are effective management of diseases such as tuberculosis (TB).2 ubiquitously expressed protein foldases which aid in protein There is acute shortage of drugs that can be used against biofilm folding or refolding by accelerating the rate-limiting cis-trans and forming pathogens. The inherent ability of the pathogen to evolve trans-cis-conformational changes at Xaa-Pro bonds.8 M.tb Ppiases under selective drug pressure outpaces the rate of development is also involved in chaperonic activity, chromatin remodelling, of new drugs, and diminishes the efficacy of the drug by the time regulatory processes in the cell, RNA-mediated gene expression, it is commercially available. The eminent solution is to expedite modulating of infections etc.9,10 Several FDA approved drugs and new arsenal of drugs against classical and non-classical targets in nanoparticle based therapies are being repurposed against M.tb proteome or establish new roles for currently available drugs. biofilms and have shown promising results. Anti-helminth drug, Drugs licensed for other known disorders in humans related to niclosamide, has shown inhibitory effects against biofilms formed

1JH-Institute of Molecular Medicine, Jamia Hamdard, New Delhi, India; 2Molecular Infection and Functional Biology Lab, Kusuma School of Biological Sciences, Indian Institute of Technology-Delhi, New Delhi, India; 3National Institute of Pathology, Safdarjung Hospital Campus, New Delhi, India; 4Department of Microbiology, Central University of Rajasthan, Ajmer, Rajasthan, India; 5Centre for Computational Biology and Bioinformatics, School of Life Sciences, Central University of Himachal Pradesh, Himachal Pradesh, India; 6Department of Biochemical Engineering and Biotechnology, Indian Institute of Technology-Delhi, New Delhi, India; 7School of Biotechnology, Jawaharlal Nehru University, New Delhi, India; 8Department of Biotechnology, Babasaheb Bhimrao Ambedkar University, Lucknow, India and 9Dr Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Prof CR Rao Road, Hyderabad, India Correspondence: Nasreen Z. Ehtesham ([email protected]) or Seyed E. Hasnain ([email protected], [email protected]) 10Present address: Department of Microbiology, Tripura Central University, Suryamaninagar, Agartala, Tripura, India 11Present address: Department of Biochemistry, Jamia Hamdard, New Delhi, India These authors contributed equally: Ashutosh Kumar, Anwar Alam, Sonam Grover, Saurabh Pandey. Received: 1 June 2018 Accepted: 15 October 2018

Published in partnership with Nanyang Technological University A. Kumar et al. 2 dosage of anti-TB drugs and can be used as conjunct drug/agent for targeting biofilm associated diseases involving other bacteria.

RESULTS Recombinant M.smegmatis expressing M.tb PpiB show increased biofilm formation in vitro M.smegmatis vector control (Ms_VC) lacking either M.tb_PpiA or M. tb_PpiB genes were used as control to examine the role of M.tb PpiA and PpiB in biofilm formation. Ms_VC and recombinant M. smegmatis (Ms_PpiA and Ms_PpiB) were induced by culturing cells in the absence and presence of anhydrotetracycline, as described in methods. Results in Fig. 1a show that Ms_VC and Ms_PpiA express basal level of biofilm, indicating that M.tb PpiA gene does not contribute to biofilm formation. It could be seen (Fig. 1b) that Ms_PpiB exhibited nearly 1.5-fold increase (p < 0.005) in biofilm formation as compared to either Ms_VC or Ms_PpiA cells. These results demonstrate the involvement of M.tb PpiB in biofilm formation, thereby modulating the cell surface properties of the pathogen.

Cyclosporine-A, acarbose, or GaNP binding sites in PpiB Fig. 1 M.tb PpiB protein enhances biofilm and pellicle formation in homologues in biofilm forming bacterial species have largely recombinant M.smegmatis: M. tb PpiA and PpiB genes were cloned in remained unaltered; Evidence of physical interaction M.smegmatis, as described in methods, to generate Ms_PpiA and Ms_PpiB strains, respectively. M.smegmatis vector control (Ms_VC), A comparison of amino acid sequences using BLAST showed that Ms_PpiA and Ms_PpiB were grown in static culture and induced M.tb PpiB exhibits at least 30% similarity with PpiB homologues in biofilm causing bacteria (supplementary Fig. S1). Results in Fig. 2a 1234567890():,; with anhydrotetracycline for expressing M.tb PPIases. a Pellicle formed in Ms_VC, Ms_PpiA and Ms_PpiB cultures from a represen- show the multiple sequence alignment of amino acid groups in tative experiment. The pellicle formed at the liquid air interface was the docking site of PpiB in biofilm forming bacteria that have quantified as described in methods. Values shown in b are mean [±s. remained conserved and could putatively interact with acarbose, e.m] from three independent experiments and depict the cyclosporine-A or dimeric atomic gallium.17 Pro162 and Arg184 in amount of biofilm formed in Ms_VC, Ms_PpiA and Ms_PpiB cultures. ’ M.tb PpiB are conserved for acarbose and cyclosporine-A binding, ***p < 0.005 vs. Ms_VC (Student s t-test) respectively and also present in other biofilm forming bacteria. Similarly for binding of dimer of atomic gallium,17 Gly203 in M.tb 11 by P.aeruginosa. Nanoparticles, by virtue of their small size and PpiB is conserved in all pathogens and the adjacent Thr204, charge have also been effective as antimicrobial agents. Silver present in the binding groove, is conserved in most of the nanoparticles have shown promising results as an alternative pathogens. Homologues of M.tb PpiB are also present in several 12 agent to inhibit bacterial biofilms. well known pathogenic bacteria such as Staphylococcus aureus, The biology of M.tb biofilm formation and its clinical relevance Staphylococcus epidermidis, Staphylococcus intermedius, Streptococ- is scant in the literature. Pellicles formed at the liquid–air interface cus mutans, Staphylococcus saprophyticus, Streptococcus constella- of a static culture are working model for in vitro studies on tus, Pseudomonas aeruginosa that are known to make biofilm. It biofilms.13 In the present study we elucidate the role of M.tb PpiB was therefore, investigated whether the amino acid residues and identify drug repurposing-based biofilm inhibitors. Recombi- present in the active site of PpiB, involved in interaction with nant M.smegmatis cells carrying M.tb PpiB gene under anhydrote- acarbose or cyclosporine-A and dimeric atomic gallium, are tracycline inducible promoter, was used as a model for biofilm common to other PpiB homologues in biofilm forming bacteria. studies. We show that heterologous expression of M.tb PpiB in M. The modelled structure of PpiB was found to have overall 98% smegmatis exhibited enhanced biofilm formation as compared to residues in the allowed regions. Our model scored −1.23 in the wild type M.smegmatis, pointing to its likely role in developing MolProbity Clashscore was greater than the recommended Global drug tolerance. Previous studies14 pointed to the possible Z-score values of −3, suggestive of being an adequate model. interaction of PpiB with cyclosporine-A rendering it a possible High throughput virtual screeing (HTVS), as described in methods, candidate among US FDA approved drug for inhibition of biofilms. was done to study the probable interaction of modelled PpiB Recent reports suggest that gallium, a FDA approved agent used sturucture with US FDA approved drugs. These drugs were ranked in cancer related hypercalcemia and cancer diagnostics, has been in order of their docking score with PpiB (supplementary Table 1). Acarbose, with the highest docking score (−13.3), was selected for repurposed for antimicrobial therapies.15,16 In-silico studies inhibition studies. Cyclosporine-A, despite having a lesser score supported by SPR data showed that acarbose, a FDA approved (−5.2), was also selected as putative drug in view of its known drug against diabetes, and cyclosporine-A, a FDA approved function as cyclophilin inhibitor.18 Prokaryotic cyclophilins bind immunosuppressant used in patients undergoing organ trans- with cyclosporine-A with weak affinity,14 although not much is fi plantation, interact with PpiB and inhibit bio lm forming activity known about PpiB cyclosporine-A interaction.19 The binding site in of PpiB. A comparison of PpiB homologues in different groups of PpiB used for docking analysis of cyclosporine-A was taken from fi bio lm forming pathogens reveals that the binding residues that the homologous structure of PpiB, which showed conserved interact with cyclosporine-A or acarbose have largely remained docking site residue in the catalytic site.20 We used the Arg184 conserved, thereby pointing to its efficacy as a putative candidate conserved residue from the catalytic centre of M.tb PpiB- for targeting biofilms across a wide genre of microorganisms. To cyclosporine-A docked complex which generated high potential our knowledge, the present study proves that PpiB is a suitable energy and therefore we assumed that this protein-drug complex candidate to target biofilm forming organisms. We also demon- may represent a real entity. Multiple interactions between M.tb strate that cyclosporine-A, acarbose or GaNP can reduce the PpiB and cyclosporine-A can be seen (Fig. 2b). Molecular docking

npj Biofilms and Microbiomes (2019) 3 Published in partnership with Nanyang Technological University A. Kumar et al. 3

carried out using alternate platform also confirmed high binding S3, panel c). Similarly, Gly in PpiB homologous proteins interacts energies of acrabose with M.tb PpiB. Estimated inhibition constant with dimeric atomic gallium in most of the biofilm forming (Ki) was in the range of 16–20 µM for acarbose. Figure 2d and pathogens (supplementary Table 2). While the interaction of supplementary Fig. S2 show the interaction plot of acarbose with gallium to PpiB is based on dimeric nature of bonding of PpiB protein. Molecular interaction studies show that Thr204 gallium,17 it remains to be shown whether the same will be true in interacts with dimeric atomic gallium (Fig. 2f, supplementary Fig. a preparation of nanoparticles. Reports exist where atomic/

Published in partnership with Nanyang Technological University npj Biofilms and Microbiomes (2019) 3 A. Kumar et al. 4 Fig. 2 Multiple sequence alignment of M.tb PpiB in biofilm forming bacteria and interaction of PpiB with cyclosporine-A, acarbose and GaNP. a M.tb PpiB (Rv2582) exhibits homology with proteins from other biofilm forming bacteria and possesses similar amino acids Arg and Pro at the binding site of cyclosporine-A (highlighted in green box) and acarbose (highlighted in red box), respectively. A dimer of atomic gallium17 similarly binds to Gly residue (highlighted in black box), which is conserved within the PpiB binding site of all biofilm-forming bacteria. b, d, f Interaction of cyclosporine-A, acarbose and dimer of atomic gallium with PpiB was tested by molecular docking analysis. b The docked complex of cyclosporine-A and PpiB. The protein (pink) is shown in surface view whereas interacting residues (grey) and ligand (green) is represented in stick model. Hydrogen bond (yellow) is shown in dotted lines. d Interactions of PpiB with acarbose showing various hydrogen and hydrophobic interactions. f The docked complex of dimer of atomic gallium and PpiB. The protein (pink) is shown in surface view whereas interacting residues (green) and ligand (red) is represented in stick model. Hydrogen bond (black) is shown in dotted lines. c, e, g SPR analysis was performed as described in methods. Response units (RU) of the interaction of PpiB with cyclosporine-A (c) or acrabose e, or GaNP g from representative experiment are shown

Fig. 3 Effect of cyclosporine-A, acarbose or GaNP on induction of biofilm in M. smegmatis (Crystal violet assay). Ms_VC and Ms_PpiB cells were cultured in absence [( )VC tet-, ( ) PpiB tet-] or presence [( ) VC tet+,( ) PpiB tet+] of anhydrotetracycline, as described in methods. Cells were treated with cyclosporine-A (0, 10, 100, 1000 μg/ml) a or acarbose (0, 1, 10, 100, 500, 1000 μg/ml) b or GaNP (0, 10, 50, 100, 1000 μg/ml) c and incubated for 7 days. At the end point, biofilm was quantified as described in methods. Values shown from a representative experiment are means [±s.e.m] of biofilm formed.*p < 0.05, **p < 0.01, ***p < 0.005 (Student’s t test)

molecule level docking involving specific amino acids in the target Having shown in-silico binding of cyclosporine A or acarbose to groove have been extrapolated to aggregation/complex/nano- PpiB, the actual physical interaction between purified recombi- particle of the same (21,22 and references therein). These results, nant M.tb PpiB and cyclosporine-A or acarbose, was tested using based on in-silico docking and high binding capacity with PpiB, SPR spectroscopy. SPR analyses show that cyclosporine-A (Fig. 2c), point to the possibility of acarbose, cyclosporine-A and gallium in acarbose (Fig. 2e) or GaNP (Fig. 2g) interact with M.tb PpiB in a acting as inhibitors of M.tb PpiB. Drug docking and molecular dose dependent manner and bind with high affinity. These results simulations studies of PpiB with homologues proteins present in suggest that cyclosporine-A, acarbose or GaNP, by virtue of their biofilm forming bacteria, annotated as WP_061736025.1, ability to bind to PpiB, could modulate the activity of PpiB. WP_049374178.1, WP_019168288.1, WP_019320573.1, WP_048792681.1, WP_006270079.1, CRQ97127.1, were also per- M.smegmatis expressing M.tb PpiB show reduced biofilm formed with acarbose, cyclosporine A and dimeric atomic gallium. formation in presence of cyclosporine-A or acarbose or GaNP Our results (supplementary Table 2) show that amino acid residues Given the earlier observation (Fig. 1) that M.tb PpiB activity is of Pro and Arg that interact with acarbose and cyclosporine-A, essential for biofilm formation, we speculated that modulation of respectively are largely conserved across the PpiB homologous PpiB activity upon binding with cyclosporine-A, acarbose or GaNP proteins, in some cases it is present at different positions. Presence could affect biofilm formation. A threshold concentration of 100, of conserved amino acids at the cyclosporine-A, acarbose and 1000, 50 µg/ml of cyclosporine-A, acarbose and GaNP, respectively gallium binding site in PpiB homologues of several biofilm at which viability of PpiB expressing M.smegmatis was not forming bacteria indicate that these have also largely remained significantly affected (supplementary Fig. S4) and did not showed unaltered and hence could prove to be an excellent putative bactericidal effect (supplementary Fig. S5) was validated using target across bacterial species. alamar blue assay. A decrease in biofilm formation may be a result of decreased cell number per-se, so it was important to ascertain

npj Biofilms and Microbiomes (2019) 3 Published in partnership with Nanyang Technological University A. Kumar et al. 5

Fig. 4 Effect of anti-TB drugs on the survival of M. smegmatis in the presence and absence of cyclosporine-A or acarbose or GaNP. Ms_VC and Ms_PpiB cells were induced with anhydrotetracycline to express ppiase in absence and presence of cyclosporine-A (100 μg/ml) a, d or acarbose (1000 μg/ml) b or GaNP (50 μg/ml) c, e. Cells were incubated in static culture to allow biofilm formation. At the end of 7 days Ms_VC and Ms_PpiB, cultured in absence of anhydrotetracycline [( )VC tet-, ( ) PpiB tet-] or presence of anhydrotetracycline [( )VC tet+,( ) PpiB tet+], were treated either with isoniazid (0, 8, 16, 32, 64 μg/ml) or ethambutol (0, 0.25, 1, 4, 16 μg/ml) and further incubated for 68 h. Susceptibility of M. smegmatis to isoniazid in absence and presence of biofilm was scored by assessing the viability of cells in a 4 h alamar blue assay. Values shown from a representative experiment are means [±s.e.m] of percent cell viability the dose of cyclosporine-A, acarbose and GaNP that does not in VC tet- and PpiB tet- cells.23–25 Significant reduction in biofilm affect the overall growth of M.smegmatis. We accordingly assessed formation compared to control (no treatment) was observed in their effect on biofilm formation by M.smegmatis expressing PpiB, presence of GaNP at 50 µg/ml (Fig. 3c) (p < 0.05). As expected, in as described in methods. It is evident (Fig. 3) that in the absence of the absence of ppib gene in Ms_VC cells, only basal levels of anhydrotetracycline induction, Ms_PpiB cells (PpiB tet-) do not biofilm formation occurred in either presence or absence of develop significant levels of biofilm. Upon induction with cyclosporine-A, acarbose or GaNP. However, it was intriguing to anhydrotetracycline, Ms_PpiB cells (PpiB tet+) developed biofilm. observe that the levels of biofilm formation in Ms_PpiB cells, It is apparent (Fig. 3a) that 100 µg/ml of cyclosporine-A resulted in uninduced by anhydrotetracycline (PpiB tet-), did not exhibit any significant decrease (p < 0.05) in biofilm formation while complete significant change to either cyclosporine-A, acarbose or GaNP inhibition of biofilm formation (p < 0.005) was evident at a administration. In Ms_PpiB cells, induction with anhydrotetracy- concentration of 1000 µg/ml, and this was comparable to basal cline (PpiB tet+) showed significant decrease in biofilm formation levels of biofilm formation in Ms_VC cells (VC tet- or VC tet+). As at 100, 1000, and 50 µg/ml of cyclosporine-A, acarbose or GaNP, shown in Fig. 3b there was a significant inhibition in the biofilm respectively. These in vitro biofilm inhibition results lend support formation in the presence of 500 and 1000 µg/ml of acarbose in to the earlier in-silico SPR results, thereby demonstrating that Ms_PpiB cells, compared to no acarbose. However, at lower cyclosporine-A, acarbose or GaNP physically interact with M.tb concentrations of acarbose treatment, a transient increase in PpiB in a dose dependent manner and suppress the activity of biofilm formation, attributed due to “Hormetic effect”, was noted PpiB protein resulting in inhibition of biofilm formation.

Published in partnership with Nanyang Technological University npj Biofilms and Microbiomes (2019) 3 A. Kumar et al. 6

Fig. 5 Effect of cyclosporine-A or GaNP on biofilm forming potential of M.tuberculosis H37Rv cells: H37Rv cells expressing PpiB were grown in static culture in absence or presence of cyclosporine-A (100 µg/ml) or GaNP (25, 50 µg/ml) for 7 days, as described in materials and methods. Representative figure of the pellicle formed at the liquid-air interface is shown. Suppressive effect of cyclosporine-A a and GaNP b on biofilm formation in M.tuberculosis H37Rv strains is evident. c Pretreatment or posttreatment of H37Rv cells with GaNP and the resultant suppression in biofilm formation

Cyclosporine-A or acarbose or GaNP co-treatment with anti-TB thereby enabling greater access of anti-TB drugs to cells. The drugs increases susceptibility of mycobacteria to these drugs consequent reduced viability of Mycobacterium in the presence of We next investigated the impact of reduced biofilm formation in cyclosporine-A, acarbose, or GaNP points to their potential use as terms of susceptibility to anti-TB drugs. Isoniazid and ethambutol adjunct therapy. are front line antibiotics that are normally effective at a dosage of 16 µg/ml and 1 µg/ml, respectively. Since M.tb PpiB expression results in enhanced biofilm formation, it could abrogate drug Cyclosporine-A or GaNP inhibit biofilm formation in M.tb sensitivity of M.smegmatis and hence the MIC of drugs is altered. In the absence of cyclosporine-A or acarbose or GaNP, PpiB tet+ While the experiments described so far were carried out on a non- cultures, induced to form biofilm, develop physical barrier over pathogenic strain of M.smegmatis, the eventual objective of our cells and prevent exposure to anti-TB drugs. PpiB tet- cultures, that study was to examine if PPIase is involved in biofilm formation in were not induced to express PpiB proteins, are unaffected by the virulent M.tb strains. Static culture of H37Rv cells were incubated in inhibitory action of cyclosporine-A or acarbose of GaNP on absence/presence of cyclosporine-A or GaNP in a BSL-3 contain- biofilms and hence are exposed directly to anti-TB drugs. Results ment facility. Results show that while untreated control H37Rv cells in Fig. 4a show that in the presence of cyclosporine-A (100 µg/ml), formed pellicle, treatment with cyclosporine-A (100 µg/ml) dosage of isoniazid was reduced from 64 to 32 μg/ml. Similarly, resulted in significant reduction (Fig. 5a) in biofilm formation, the dosage of isoniazid in the presence of acarbose (500 µg/ml), reduction was more pronounced in the presence of 50 µg/ml was reduced (Fig. 4b) from 64 to 32 μg/ml for PpiB (tet+). GaNP (Fig. 5b). The role of GaNP in suppressing biofilm formation Likewise, dosage of isoniazid in presence of GaNP (50 µg/ml), was was further examined in two clinically relevant scenarios. H37Rv 16 µg/ml (Fig. 4c), a four-fold decrease as compared to control. cells, pretreated for 6 and 24 h with GaNP (25, 50 µg/ml), were Similar experiments were carried out for ethambutol. Results incubated to allow biofilm formation. In another set of experi- (Fig. 4d) show that in the absence of cyclosporine-A, PpiB tet+ ments, H Rv cells were allowed to form biofilm and GaNP (25, cultures exhibited dosage of 16 μg/ml for ethambutol but this 37 50 µg/ml) treatment was carried out post-biofilm formation. decreased four-fold (4 μg/ml, p < 0.05) in the presence of Pretreatment of H37Rv cells with GaNP (Fig. 5c) resulted in dose cyclosporine-A (100 µg/ml). Results (Fig. 4e) show that dosage of fi ethambutol in the presence of GaNP (50 µg/ml), decreased from dependent inhibition of bio lm formation that correlated with the 16 to 1 µg/ml, as compared to control. The efficacy of acarbose in duration for which the cells were pre-treated, the suppression in fi decreasing the dosage of ethambutol was insignificant (data not bio lm formation being enhanced when H37Rv cells were pre- shown). These results clearly demonstrate that cyclosporine-A treated with GaNP for 24 h as compared to 6 h. Treatment with (100 μg/ml) or GaNP (50 µg/ml) inhibit the activity of PpiB protein GaNP post-biofilm formation resulted in disintegration of pellicle which in turn negatively impacts the ability of the bacterium to at the liquid-air interface. form biofilm efficiently, resulting in reduced percent viability

npj Biofilms and Microbiomes (2019) 3 Published in partnership with Nanyang Technological University A. Kumar et al. 7 DISCUSSION Cyclophilin inhibitors, as a unique tool in therapeutic biology, 39 The current study, directed to address the problem of biofilms are showing promising results in several diseases. Previous 19 using Mycobacterium as model organism, explores proteins that studies pointed that PpiA is a Cyclosporine-A binding cyclophilin may aid in biofilm formation and their putative inhibitors from the and treatment with cyclosporine-A sensitises drug-tolerant biofilm 40,41 databank of US FDA approved drugs. Biofilm formation involves a of Candida albicans to various antifungal drugs. We used M. complex process that exhibits heterogeneity in terms of the key smegmatis over expressing M.tb PpiB as a model to evaluate the pathways or mechanisms among different groups of microorgan- effect of known and unknown inhibitors of cyclophilins. In-silico isms. While factors such as PrfA and SinR regulate biofilm docking analysis of interaction of cyclosporine-A with M.tb PpiB formation in Listeria and Bacillus subtilis respectively, several other homologue revealed that PipB possesses conserved amino acid factors such as RNA regulatory proteins RsmA in P. aeruginosa, groups in the binding pocket. These molecular docking studies, in chaperonic proteins GroEL-1 in M.smegmatis, cell wall proteins sync with previous studies, show that Cyclosporine-A can PE11 in M.tb have also been shown to play key role in biofilm stereochemically bind with PpiB. Among the FDA approved drugs, formation.26 There is little consensus on a single protein or factor acarbose exhibited greatest docking score and could potentially that may act as a master molecule for biofilm formation. We interact with PpiB. The physical interaction of cyclosporine-A, fi therefore, started to identify unique protein that could act as acarbose and GaNP with PpiB was experimentally con rmed putative candidate affecting biofilm formation across species. through SPR studies. Consistent with these results, we showed that cyclosporine-A, acarbose or GaNP could suppress biofilm Except in Mycoplasma genitilium and some members of archaea, fi all microorganisms possess a highly conserved and ubiquitously formation. A biphasic dose response for bio lms has been expressed group of proteins known as cyclophilins. Cyclophilins, reported for many inhibitors/antibiotics/chemicals/drugs/ligands etc. Such a response, termed as “Hormetic response”,is such as peptidyl-prolyl isomerases (PPIases; EC 5.2.1.8), catalyse fi the cis/trans isomerization of peptidyl-prolyl bonds and are characterised by stimulation of bio lm formation at lower dose and inhibition of the same at higher dose. Some reports have also therefore important for correct folding or refolding of nascent pointed out to antibiotic acting as antagonist of biofilm formation proteins that in turn regulate interacting partner proteins to form 27 at low levels, agonists at higher levels and once again antagonist complexes. The role of several PPIases in biofilm formation, – at still higher level.23 25 This is exactly what we have observed: stress tolerance and pathogenesis of bacteria are already – acarbose at lower concentration (upto 100 µg/ml) showed known.28 30 increase in biofilm formation and at 500 µg/ml and above M.tb possesses two types of cyclophilins, PpiA and PpiB, of inhibited biofilm formation. While cyclosporine A and acarbose these only PpiB is essential for the survival of the pathogen as exhibit bacteriostatic activity at the concentration reported knockout variants fail to survive.31 Previous studies9 showed that (supplementary Fig. S5), only at higher concentration GaNP M.tb PpiB possess chaperonic activity and aid in intracellular exhibits bactericidal effect. survival of M.tb. M.tb PpiA or PpiB genes under the control of A comparison of the amino acid domains in the binding pocket anhydrotetracycline inducible promoter were cloned in M. of PpiB homologues expressed in biofilm forming microorganisms smegmatis. Our results (Fig. 1) demonstrated that M.smegmatis fi interestingly showed that PpiB possess similar amino acids that overexpressing M.tb PpiB, not PpiA, developed signi cantly can interact with either cyclosporine-A or acarbose or GaNP. The greater biomass of pellicle as compared to a basal expression in structure of gallium nanoparticle has not been reported so far, we control cells. It is interesting to mention that M.smegmatis PpiB therefore used dimeric form of atomic gallium17 that could act as displays 64% homology with M.tb PpiB (supplementary Fig. S6). A the building block of gallium nanoparticle. This clearly positions fi distinct increase in bio lm formation, when compared with vector PpiB as a unique protein that can be targeted to inhibit biofilm control, was expectedly seen in M.smegmatis over expressing M.tb formation across bacterial species, more so when several mixed PpiB. Glycopeptidolipids, like PpiB, are component of the species of microorganisms exist in the biofilm. Each of these membrane fraction and are also part of the secretome and are heterogenous species develop biofilms using varying cellular fi 32–34 known to play important role in bio lm formation. Our results pathways. Although some antibiotics act as anti-biofilm agents, clearly demonstrate a direct involvement of M.tb PpiB in biofilm 9 however, such a drug that may be effective against a putative formation. That PpiB also acts as a chaperone is in agreement protein involved in biofilm formation may not be as effective in with reports of staphylococcus trigger factor having roles in stress 32 other organisms either due to the absence of the protein target or tolerance and biofilm formation. redundancy in the metabolic pathway. The presence of conserved Our next step was to identify suitable drug(s) that could act as amino acid Arg, Pro, Gly at the binding site of Cyclosporine-A, an inhibitor of PpiB protein. Developing new drugs is a long acarbose and GaNP, respectively highlights that PpiB could prove process taking about 10–15 years. Drug repurposing is gaining to be a unique target in controlling biofilms, thereby providing a popularity as it allows bypassing of the cumbersome clinical trial possible generic mechanism for treatment of infections caused by of drugs for which the parameters of toxicity and effectiveness other biofilm producing pathogens. have already been tested and approved. The effectiveness of While there are global efforts to develop new drugs against TB, osteoarthritis drug Celebrex in decreasing polyp formation in efforts are needed to reduce the duration of the drug regimen. colon cancer patients, anti-malarial drug chloroquine in improving Using first line anti-TB drugs (ethambutol and isoniazid), we have outcome of cancer drug Erlotinib, anti-diabetic drug metformin in shown that the reduced mycobacterial biofilm formation in the 35 lowering morbidity of TB patients are some examples of drug. presence of cyclosporine-A, acarbose or GaNP (Fig. 3) results in An inhibitor of human phosphodiesterase, sildenafil originally dosage reductions for these anti-TB drugs (Fig. 4). While increased used in case of erectile dysfunction, has shown encouraging dosage of anti-TB drugs results in drug tolerance of the pathogen, results in animal studies and is now being deliberated as an it also has a negative impact on patients in terms of toxicity. Our adjuvant host directed therapy to curtail the duration of TB drug results show that treatment with cyclosporine-A or acarbose help regimen.36 Nanoparticles are also emerging as key modulators in reducing the dosage of anti-TB drugs by at least two-fold. This against several human pathogens. FDA approved gallium has has wide implications as it provides proof of principle that shown promising efficacy against M.tb due to similar charge as Fe, cyclosporine-A, a known immunosuppressant that affects T cells, thereby allosterically competing with Fe to bind Fe- can be repurposed as a conjunct therapy against biofilm siderophores.37 This results in disruption of iron metabolism, associated diseases (Fig. 6). One can argue that treatment with leading to failure of microbial cells to grow in presence of cyclosporine-A may activate latent TB by suppressing immunity. It gallium.38 is conceivable that the concentration of cyclosporine-A at which it

Published in partnership with Nanyang Technological University npj Biofilms and Microbiomes (2019) 3 A. Kumar et al. 8

Fig. 6 Schematic overview of the effect of repurposed drugs on biofilm and its outcome on tuberculosis treatment: Under stress like conditions Mycobacteria secrete exogenous layer of matrix that forms a physical barrier for entry of drugs. The cells within the matrix continuously secrete to develop a biomass of biofilm that enables the cells to withstand high minimum inhibitory concentration (MIC) of drugs. As a result, higher dosage of drugs is required to kill the cells. Cells at the core of the biofilm matrix are least affected by drugs and evolve in due time so as to withstand even higher concentration of drugs. This confers drug tolerance and leads to drug toxicity, increased treatment cost and mortality. Cyclosporine-A, acarbose and GaNP inhibit the activity of PpiB that play crucial role in biofilm formation. Treatment with these drugs suppresses formation of biofilm and the bacterium is exposed directly to the drugs. As a result the drug is effective at low MIC values. Treatment with these drugs also reduces the MIC of existing anti-tubercular drugs resulting in decreased toxicity. The end result is that patient mortality and treatment cost may be reduced significantly. Regular and dotted arrows in the figure denote confirmed and putative roles respectively

inhibits biofilm formation can be reduced further to minimal infected with both M.tb and HIV.45,46 These studies support our concentrations by using this drug with suitable adjuvants, thereby results and point to the possibility that GaNP would be an reducing chances of its immunosuppressive effects to prevail over effective intervention against bacterial biofilms. It has not escaped its efficacy as biofilm inhibitor. Our results related to the use and our attention that cyclosporine-A as an adjunct to existing anti- efficacy of cyclosporine-A is in line with previous reports that tubercular drugs could be a potential strategy to address the suggest that it acts synergistically to improve the efficacy of problem of eradicating latent TB by first activating the bacterium, antifungals against C.parapsilosis.42 Cyclosporine-A in combination by virtue of its immune-suppressive action, followed by the with azole antifungal flucanizole has been shown to be effective biofilm inhibition reported in our study. against biofilms formed by C.albicans43 and also imparts sensitivity We also show that acarbose treatment resulted in a reduction of to C.albicans towards fluconazole by involving multiple path- dosage of anti-TB drugs such as isoniazid. Acarbose is widely used ways.44 GaNP is known to facilitate phagosome maturation, inhibit for the management of type 2 Diabetes mellitus as well. The growth of M.tb in macrophages, inhibit HIV infection through efficacy of acarbose to block maltose importer and consequently release of interferons and can be targeted to human macrophages suppress growth of E. coli is known.47 Given the fact that TB and

npj Biofilms and Microbiomes (2019) 3 Published in partnership with Nanyang Technological University A. Kumar et al. 9 diabetes exhibit distinct correlation in patients and synergistically pellicle was stained by adding 125 µl (w/v) 0.1% Crystal Violet solution. The affect the clinical outcome of each other in patients, the efficacy of stained pellicle was washed thrice with water followed by addition of 30% acarbose as a medicament in reducing the dosage of anti-TB acetic acid. The samples were subsequently incubated for 10–15 min at drugs could prove to be beneficial. There are several bacteria that room temperature to dissolve the stain and absorbance was spectro- are involved in the biofilm of cystic fibrosis, wounds, contact photometrically recorded at 550 nm. lenses, orthopaedic implants, breast implants, dental biofilm, pacemakers, prosthetic heart valves etc. and have similar protein Percent viability of biofilm induced culture 48 fi to M.tb PpiB. Our results can be extrapolated to test the ef cacy Biofilm formation by M.smegmatis Ms_VC and Ms_PpiB induced with or of cyclosporine-A or acarbose or GaNP in reducing the dosage of without anhydrotetracycline was performed in presence of FDA approved other drugs or for diseases caused by biofilm-forming micro- agents, as described above. At the end of 7 days of incubation period organisms as well. isoniazid (8, 16, 32, 64 µg/ml) or ethambutol (0.25, 1, 4, 16 µg/ml) was Taken together, these results conclusively demonstrate that added to the wells of the microtiter plate and the plate was incubated PpiB is a potential drug target involved in Mycobacterium biofilm further for 68 h. Cell viability in presence of isoniazid and ethambutol were formation and cyclosporine-A, acarbose and GaNP directly bind to assessed using Alamar Blue assay.50 Briefly, 0.01% alamar blue reagent was PpiB and disrupt biofilm formation. The consequent reduction in added to each well of microtiter plate and the plates were further dosage values for anti-TB drugs ethambutol and isoniazid points incubated for 3–4 h. Conversion of resazurin (blue) to resorufin (pink) was to their ability to act as therapeutic interventions to counter drug monitored at 570 nm and 600 nm, respectively to score the viability of the tolerance and also possibly reduce dosage of existing anti- cells. All assays were performed in triplicate. tubercular drugs with implications in reducing drug-induced toxicity and also treatment duration. It will also be interesting to In-silico amino acid sequence alignment and similarity search evaluate PpiB as a drug target given the fact that PpiB is not only fi The amino acids sequences of M.tb PpiA (GenBank accession number: an essential gene of M.tb, but is involved in bio lm formation. The CCP42731.1), M.tb PpiB (GenBank accession number: CCE38048.1 were conservation of drug binding sites within PpiB across pathogenic downloaded from the NCBI. The sequence homology search of M.tb PpiB bacteria biofilm tempts us to suggest that PpiB-targeted biofilm was done using BLASTp in known biofilm forming bacteria on NCBI disruption could prove to be a masterstroke for combatting website. M.tb PpiB amino acids sequence was used as queries in BLASTp biofilm-related infections across microbial species. analyses against the NCBI non-redundant protein database of the specific bacteria to find their similar homologues. METHODS Modelling of PpiB structure and molecular dynamics (MD) Reagents simulations M.smegmatis mc2155, initially obtained from ATCC, was maintained in our Crystallographic structure of PpiB, being unavailable at Protein Data Bank, laboratory as glycerol stocks. M.tb H37Rv (gift from Prof R. K. Bhatnagar, Jawaharlal Nehru University, New Delhi, India), was cultured in BSL-3 homology modelling techniques involving multiple bioinformatics tools or facility. Growth media (Middlebrook 7H9) and OADC supplement were servers such as the MODELLER version 9.11 or Phyre2, respectively were obtained from BD, USA. Glycerol, Tween 80, acetic acid, acarbose and used to generate PpiB model structure. Protein sequence of PpiB from M.tb cyclosporine-A were procured from Sigma-Aldrich, India. GaNP (purity > (strain ATCC 25618/H37Rv) was obtained from UniprotKB database 99.9%) was obtained from Nanoshel, India. All other reagents such as [P9WHW1]. Protein structure model validation was carried out using Alamar Blue, Crystal Violet, Isoniazid, and Ethambutol were of analytical protein structure validation software suite (PSVS). grade and obtained from Himedia, India. Molecular docking analysis of M.tb cyclophilin (PpiB) in complex Constructs and recombinant strains used in the study with cyclosporine-A, acarbose, or gallium Recombinant strains of M.smegmatis expressing M.tb Ppiases were Molecular docking analysis of cyclosporine-A was carried out to study the generated using E. coli-mycobacterium shuttle vector pST2K and specific interactions and affinity with the PpiB protein using AutoDock Tools 1.5.6 10 fl oligonucleotide primers, as detailed elsewhere. Brie y, pST_ppiA, (open access). 3D structure of cyclosporine-A was obtained from chemical pST_ppiB and pST2K vector containing anhydrotetracycline inducibe structure database ChemSpider. AutoDockVina 1.1.2 program was used for promoter were electroporated in wild type M.smegmatis (Ms_WT) and docking of cyclosporine-A at the docking site of PpiB protein. The location transformed strains were designated as Ms_PpiA, Ms_PpiB, and Ms_VC, of the catalytic site was mapped and deduced from the structure-based respectively. alignment of related proteins reported earlier.20 Alternatively, Glide module of Schrodinger was used to screen other Cell culture and biofilm formation compounds from FDA library as described previously.51 Briefly, Drug library fi M.smegmatis and M.tb H37Rv were maintained in growth media was prepared using Ligprep module applying OPLS 2005 force eld and supplemented with 10% OADC, 0.001% glycerol, and 0.05% Tween-80. docking was performed using HTVS and XP (extra precision) docking to Cultures were incubated at 37°C in a shaker incubator and diluted to OD of filter out the compounds with low binding energy. Compounds having a 0.08 in growth media prior to sub-culturing in 96-wells or in test tubes for docking score greater than −5 in HTVS were used for XP docking protocol. induction of biofilms. 20 ng/ml anhydrotetracycline was used to induce An XP score greater than −8 was scored as strong binding. The dynamic PPIase expression in recombinant M.smegmatis, resulting in biofilm nature of interaction between PpiB and acarbose was studied using formation. Recombinant M.smegmatis and M.tb H37Rv were cultured in GROMACS version 4.6.5 and above assigning GROMOS96 43a1 force field static phase in Tween 80-free growth media for 7 days and 4 weeks, as per standard protocols. respectively to allow formation of pellicle at the liquid air interface. The chemical structure of elemental gallium was obtained from PubChem (CID 5464084). Using the elemental gallium, a dimer structure Crystal violet assay for atomic gallium was created using Maestro interface available from Cyclosporine-A, acarbose, and GaNP effect on biofilm formation was Schrodinger. Molecular orbital analysis of gallium in dimeric state showed that gallium dimer is the essential building block for the formation of assessed by quantifying the pellicle formed at the liquid air interface, using 17 crystal violet.49 Anhydrotetracycline induced Ms_PpiB and Ms_VC cells gallium clusters. Further molecular docking study of dimeric atomic were cultured in the presence of various concentrations of cyclosporine-A gallium was carried out with PpiB protein and its homologues using 52 (0, 10, 100, and 1000 µg/ml), acarbose (0, 1, 10, 100, 500, 1000 µg/ml) or PatchDock algorithm. Ligplot was used for visualisation of the interac- GaNP (0, 10, 50, 100, 1000 µg/ml) in sterile flat bottom 96-well microtiter tions between protein-ligand complex in 2D schematic representations. plate (Thermo Scientific, India). At the end of 7 days of static phase culture, PyMol and Chimera were used for preparing cartoon representations of the the media beneath the pellicle were aspirated out and remaining solid structures.

Published in partnership with Nanyang Technological University npj Biofilms and Microbiomes (2019) 3 A. Kumar et al. 10 Surface plasmon resonance (SPR) study non-ST131 lineages from India. Antimicrob. Agents Chemother. 58, 7240–7249 In-vitro interactions of M.tb PpiB with cyclosporine-A, acarbose and GaNP (2014). were studied using SPR technique employing Autolab ESPRIT analyzer, as 7. Lenaerts, A. J. et al. Location of persisting mycobacteria in a Guinea pig model of per manufacturer’s instructions. Briefly, M.tb rPpiB9 (expressed in E. coli tuberculosis revealed by r207910. Antimicrob. Agents Chemother. 51, 3338–3345 BL21(DE3) rosetta strain) was diluted in PBS upto a concentration of (2007). 100 μg/ml and passed through a CM5 sensor chip for immobilisation. 8. Davis, T. L. et al. Structural and biochemical characterization of the human During association phase (300 s) cyclosporine-A, acarbose and GaNP were cyclophilin family of peptidyl-prolyl isomerases. PLoS Biol. 8, e1000439 (2014). diluted to different concentration in running buffer (PBS) and were allowed 9. Pandey, S. et al. Mycobacterium tuberculosis peptidyl-prolyl isomerases also to pass over the immobilised PpiB. During dissociation phase (150 s) PBS exhibit chaperone like activity in-vitro and in-vivo. PLoS One 11, e0150288 (2016). was applied to sensor chip and the sensor chip was regenerated between 10. Pandey, S. et al. Mycobacterium tuberculosis peptidyl-prolyl isomerases are each binding experiment with NaOH. immunogenic, alter aytokine profile and aid in intracellular survival. Front. Cell. Infect. Microbiol. 7, 38 (2017). 11. Imperi, F. et al. New life for an old drug: the antihelminthic drug niclosamide Statistical analysis inhibits P aerugonisa quorum sensing. Antimicrob. Agents Chemother. 57, Statistical analyses using a two-way analysis of variance (ANOVA) and 996–1005 (2013). Student’s t-test were performed using Sigmaplot and Sigmastat software 12. Markowska, K. et al. Silver nanoparticles as an alternative strategy against bac- (Systat Software Inc, San Jose, CA). A p-value (p < 0.05) was considered terial biofilms. Acta Biochem. Pol. 60, 523–530 (2013). significant. 13. Ojha, A. K. et al. Growth of Mycobacterium tuberculosis biofilms containing free mycolic acids and harbouring drug-tolerant bacteria. Mol. Microbiol. 69, 164–174 (2008). DATA AVAILABILITY 14. Unal, C. M. & Steinert, M. Microbial peptidyl prolyl cis/trans isomerase (PPIases): The authors declare that all the data supporting the findings of this study are virulence factors and potential alternative drug targets. Microbiol. Mol. Biol. Rev. – available within the article, or upon request from the corresponding author. 78, 544 571 (2014). 15. Warrel, R. P. Clinical trials of gallium nanoparticle in patients with cancer related hypercalcemia. Semin. Oncol. 18,26–31 (1991). ACKNOWLEDGEMENTS 16. Bonchi, C. et al. Repurposing of gallium based drugs for antibacterial therapy. Biofactors 40, 303–312 (2014). This work was supported by Centre of Excellence Grant (BT/PR12817/COE/ 34/23/ 17. Tonner, R. & Gaston, N. The dimeric nature of bonding in gallium: from small 2015) and (BT/PR23099/NER/95/632/2017) from Department of Biotechnology (DBT), clusters to the alpha-gallium phase. Phys. Chem. Chem. Phys. 16, 24244–24249 Ministry of Science and Technology, Government of India. S.E.H. is a JC Bose National (2014). Fellow, Department of Science and Technology, Government of India and Robert 18. Kim et al. Inhibition of cyclophilin A by cyclosporine A promotes retinal ganglion Koch Fellow, Robert Koch Institute, Berlin, Germany. We thank Prof Rakesh Bhatnagar cell survival by preventing mitochondrial alteration in ischemic injury. Cell Death from the School of Biotechnology, Jawaharlal Nehru University, New Delhi, India for Dis. 5, e1105 (2014). use of BSL 3 containment Facility. S.G. was a recipient of DHR Women Scientist 19. Takahashi, N., Hayano, T. & Suzuki, M. Peptidyl prolyl cis trans isomerase is the Fellowship. A.A. was a recipient of DBT Research Associateship grant from the cyclosporine A binding protein cyclophilin. Nature 337, 473–475 (1989). Department of Biotechnology. Department of Biotechnology (DBT), Ministry of 20. Henriksson, L. M. et al. X‐ray structure of peptidyl‐prolyl cis–trans isomerase A Science and Technology, Government of India supported project grants (BT/ from Mycobacterium tuberculosis. FEBS J. 271, 4107–4113 (2004). PR12817/COE/ 34/23/2015) and (BT/PR23099/NER/95/632/2017). Research in Y.A. 21. Asl, B. A. et al. Probing the interaction of zero valent iron nanoparticles with lab is funded by DBT and Indian Council of Medical Research (ICMR). We blood system by biophysical, docking, cellular, and molecular studies. Int. J. Biol. acknowledge the support for Article Processing Charges from intramural funds by Macromol. 1, 639–650 (2018). Indian Council of Medical Research (ICMR), Ministry of Health and Family Welfare, 22. Sekar, G. et al. Exploring the interaction between iron oxide nanoparticles (IONPs) Government of India. and human serum albumin (HSA): spectroscopic and docking studies. J. Mol. Liq. 241, 793–800 (2017). 23. Calabrese, E. J. & Baldwin, L. A. Hormesis: U-shaped dose responses and their AUTHOR CONTRIBUTIONS centrality in toxicology. Trends Pharmacol. Sci. 22, 285–291 (2001). The study was designed by S.E.H., N.Z.E., A.A. and A.K. A.A., A.K., S.G., S.P., D.T., M.K., M. 24. Davies, J., Spiegelman, G. B. & Yim, G. The world of subinhibitory antibiotic R., A.S., and Y.A. carried out the experiments. A.K., A.A., S.G., S.P. are co-first authors. A. concentrations. Curr. Opin. Microbiol. 9, 445–453 (2006). A., A.K., N.Z.E., S.G., and S.E.H. wrote the manuscript. 25. Kendig, E. L., Le, H. H. & Belcher, S. M. Defining hormesis: evaluation of a complex concentration response phenomenon. Int. J. Toxicol. 29, 235–246 (2010). 26. Kumar, A., Alam, A., Rani, M., Ehtesham, N. Z., & Hasnain, S. E. Biofilms: survival ADDITIONAL INFORMATION and defense strategy for pathogens. Int. J. Med. Microbiol. https://doi.org/10.1016/ Supplementary information accompanies the paper on the npj Biofilms and j.ijmm.2017.09.016 (2017). Microbiomes website (https://doi.org/10.1038/s41522-018-0075-0). 27. Gothel, S. F. & Marahiel, M. A. Peptidyl-prolyl cis-trans isomerases, a superfamily of ubiquitous folding catalysts. Cell. Mol. Life Sci. 55, 423–436 (1999). Competing interests: The authors declare no competing interests. 28. Roset, M. S., Garcia, F. L., DelVecchio, V. G. & Briones, G. Intracellularly induced cyclophilins play an important role in stress adaptation and virulence of Brucella abortus. Infect. Immun. 81, 521–530 (2013). ’ Publisher s note: Springer Nature remains neutral with regard to jurisdictional claims 29. Skagia, A. et al. Cyclophilin PpiB is involved in motility and biofilm formation via fi in published maps and institutional af liations. its functional association with certain proteins. Genes Cells 21, 833–851 (2016). 30. Skagia, A. et al. Functional analysis of the cyclophilin PpiB role in bacterial cell division. Genes Cells 22, 810–824 (2017). REFERENCES 31. Sassetti, C. M., Boyd, D. H. & Rubin, E. J. Genes required for mycobacterial growth 1. Davies, D. Understanding biofilm resistance to antibacterial agents. Nat. Rev. Drug defined by high density mutagenesis. Mol. Microbiol. 48,77–84 (2003). Discov. 2, 114–122 (2003). 32. Wen, Z. T., Suntharaligham, P., Cvitkovitch, D. G. & Burne, R. A. Trigger factor in 2. Kumar, A. et al. Redox homeostasis in mycobacteria: the key to tuberculosis Streptococcus mutans is involved in stress tolerance, competence development, control? Expert. Rev. Mol. Med. 13, e39 (2011). and biofilm formation. Infect. Immun. 73, 219–225 (2005). 3. Novac, N. Challenges and opportunities of drug repositioning. Trends Pharmacol. 33. Recht, J. & Kolter, R. Glycopeptidolipid acetylation affects sliding motility and Sci. 34, 267–272 (2013). biofilm formation in Mycobacterium smegmatis. J. Bacteriol. 183, 5718–5724 4. Hoiby, N., Ciofu, O. & Bjarnsholt, T. Pseudomonas aeruginosa biofilms in cystic (2001). fibrosis. Future Microbiol. 5, 1663–1674 (2010). 34. Schorey, J. S. & Sweet, L. The mycobacterial glycopeptidolipids: structure, func- 5. Qvist, T. et al. Chronic pulmonary disease with Mycobacterium abscessus com- tion, and their role in pathogenesis. Glycobiology 18, 832–841 (2008). plex is a biofilm infection. Eur. Respir. J. 46, 1823–1826 (2015). 35. Singhal, A. et al. Metformin as adjunct antituberculosis therapy. Sci. Transl. Med. 6, 6. Hussain, A. et al. Genotypic and phenotypic profiles of Escherichia coli isolates 263ra159 (2014). belonging to clinical sequence type 131 (ST131), clinical non-ST131, and fecal

npj Biofilms and Microbiomes (2019) 3 Published in partnership with Nanyang Technological University A. Kumar et al. 11 36. Maiga, M. et al. Successful shortening of tuberculosis treatment using adjuvant 48. Singh, P. K. et al. Quorum-sensing signals indicate that cystic fibrosis lungs are host-directed therapy with FDA-approved phosphodiesterase inhibitors in the infected with bacterial biofilms. Nature 407, 762–764 (2000). mouse model. PLoS One 7, e30749 (2012). 49. O’Toole, G. A. Microtiter dish biofilm formation assay. J. Vis. Exp. https://doi.org/ 37. Wilson, B. R., Bogdan, A. R., Miyazawa, M., Hashimoto, K. & Tsuji, Y. Siderophores in 10.3791/2437 2437 (2011). iron metabolism: From mechanism to therapy potential. Trends Mol. Med. 22, 50. Taneja, N. K. & Tyagi, J. S. Resazurin reduction assays for screening of anti- 1077–1090 (2016). tubercular compounds against dormant and actively growing Mycobacterium 38. Olakanmi, O. et al. Gallium disrupts iron metabolism of mycobacteria residing tuberculosis, Mycobacterium bovis BCG and Mycobacterium smegmatis. J. Anti- within human macrophages. Infect. Immun. 689, 5619–5627 (2000). microb. Chemother. 60, 288–293 (2007). 39. Hopkins, S. & Gallay, P. Cyclophilin inhibitors: an emerging class of therapeutics 51. Singh, A. et al. Computational identification of novel piperidine derivatives as for the treatment of chronic hepatitis C infection. Viruses 4, 2558–2577 (2012). potential HDM2 inhibitors designed by fragment-based QSAR, molecular docking 40. Mitra, D., Mukherjee, S. & Das, A. K. Cyclosporin A binding to Mycobacterium and molecular dynamics simulations. Struct. Chem. 27, 993–1003 (2016). tuberculosis peptidyl-prolyl cis-trans isomerase A-investigation by CD, FTIR and 52. Schneidman-Duhovny, D., Inbar, Y., Nussinov, R. & Wolfson, H. J. PatchDock & fluorescence spectroscopy. FEBS Lett. 580, 6846–6860 (2006). SymmDock: servers for rigid and symmetric docking. Nucleic Acids Res. 33, 41. Shinde, R. B., Chauhan, N. M., Raut, J. S. & Karuppayil, S. M. Sensitization of W363–W367 (2005). Candida albicans biofilms to various antifungal drugs by cyclosporine A. Ann. Clin. Microbiol. Antimicrob. 11, 27 (2012). 42. Corderio, R. D. A. et al. The calcineurin inhibitor cyclosporine A exhibits synergism Open Access This article is licensed under a Creative Commons with antifungals against C parasilopsis species complex. J. Med. Micro. 63, Attribution 4.0 International License, which permits use, sharing, 936–944 (2014). adaptation, distribution and reproduction in any medium or format, as long as you give 43. Uppuluri, P. et al. Synergistic effect of calceneurin inhibitors and fluconazole appropriate credit to the original author(s) and the source, provide a link to the Creative against C albicans biofilms. Antimicro. Agents Chemother. 52, 1127–1132 (2008). Commons license, and indicate if changes were made. The images or other third party 44. Jia, W. et al. The calceneurin inhibitor cyclosporine A synergistically enhances the material in this article are included in the article’s Creative Commons license, unless susceptibility of C albicans biofilms to flucanozole by multiple mechanisms. BMC indicated otherwise in a credit line to the material. If material is not included in the Microbiol. 16, 113 (2016). article’s Creative Commons license and your intended use is not permitted by statutory 45. Choi et al. GaNP inhibits growth of both mycobacteria and HIV and release of regulation or exceeds the permitted use, you will need to obtain permission directly Interleukin-6 and IL-8 in coinfected macrophages. Antimicro. Agents Chemother. from the copyright holder. To view a copy of this license, visit http://creativecommons. 61, e02505–e02516 (2017). org/licenses/by/4.0/. 46. Choi et al. GaNP facilitate phagosome maturation and inhibit growth of virulent M. tb in macrophages. PLos One https://doi.org/10.1371/journal.pone.0177987 (2017). © The Author(s) 2019 47. Brunkhorst, C., Andersen, C. & Schneider, E. Acarbose, a pseudooligosaccharide, is transported but not metabolized by the maltose-maltodextrin system of Escherichia coli. J. Bacteriol. 181, 2612–2619 (1999).

Published in partnership with Nanyang Technological University npj Biofilms and Microbiomes (2019) 3

201 8

b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480

ht tp://www.bjmicrobiol.com.br/

Environmental Microbiology

Study of pandrug and heavy metal resistance

among E. coli from anthropogenically influenced

Delhi stretch of river Yamuna

a,1 b,1 c a

Mudsser Azam , Arif Tasleem Jan , Ashutosh Kumar , Kehkashan Siddiqui ,

a a,∗

Aftab Hossain Mondal , Qazi Mohd. Rizwanul Haq

a

Department of Biosciences, Jamia Millia Islamia, New Delhi, India

b

School of Biosciences and Biotechnology, Baba Ghulam Shah Badshah University, Rajouri, India

c

Kasuma School of Biological Sciences, Indian Institute of Technology, New Delhi, India

a r t i c l e i n f o a b s t r a c t

Article history: Escalating burden of antibiotic resistance that has reached new heights present a grave

Received 7 October 2016 concern to mankind. As the problem is no longer confined to clinics, we hereby report iden-

Accepted 21 November 2017 tification of a pandrug resistant Escherichia coli isolate from heavily polluted Delhi stretch of

Available online 12 February 2018 river Yamuna, India. E. coli MRC11 was found sensitive only to tobramycin against 21 antibi-

otics tested, with minimum inhibitory concentration values >256 ␮g/mL for amoxicillin,

Associate Editor: Elizabeth Marques

carbenicillin, aztreonam, ceftazidime and cefotaxime. Addition of certain heavy metals

at higher concentrations were ineffective in increasing susceptibility of E. coli MRC11 to

Keywords:

antibiotics. Withstanding sub-optimal concentration of cefotaxime (10 ␮g/mL) and mercuric

Antibiotic resistance

chloride (2 ␮g/mL), and also resistance to their combinatorial use, indicates better adapt-

ESBLs

ability in heavily polluted environment through clustering and expression of resistance

Heavy metal resistance

genes. Interestingly, E. coli MRC11 harbours two different variants of blaTEM (blaTEM-116

Polluted environment

and blaTEM-1 with and without extended-spectrum activity, respectively), in addition to mer

Resistance genes

operon (merB, merP and merT) genes. Studies employing conjugation, confirmed localization

of blaTEM-116, merP and merT genes on the conjugative plasmid. Understanding potentiali-

ties of such isolates will help in determining risk factors attributing pandrug resistance and

strengthening strategic development of new and effective antimicrobial agents.

© 2018 Sociedade Brasileira de Microbiologia. Published by Elsevier Editora Ltda. This is

an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).

∗

Corresponding author at: Q.M.R. Haq, Department of Biosciences, Jamia Millia Islamia, New Delhi, India.

E-mail: [email protected] (Q.M. Haq).

1

Authors contributed equally to this work.

https://doi.org/10.1016/j.bjm.2017.11.001

1517-8382/© 2018 Sociedade Brasileira de Microbiologia. Published by Elsevier Editora Ltda. This is an open access article under the CC

BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

472 b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480

towards wide range of antibiotics and heavy metals. Genetic

Introduction

determinants imparting ESBL positive phenotype, conjugation

frequency and biofilm formation in different media composi-

Pollution of the aquatic environment through anthropogenic

tion were studied for the isolate. Additionally, growth kinetics

activities, such as discharge of municipal sewage and

and susceptibility pattern to antibiotic and heavy metal either

untreated or partially treated industrial waste, creates means

singly or in combination were also performed.

for selection, proliferation and dissemination of resistant

traits among bacteria. Dissemination of resistance genes

that diminishes the treatment options of infectious diseases Methodology

has compromised human and animal health against mul-

1

tidrug resistant bacteria. Pollution of aquatic environments Isolation and characterization of bacteria

is common in developing countries like India, one such case

occurs with the largest tributary of river Ganga – Yamuna, Upon collection of the water samples from thirteen diverse

having its origin from Yamunotri Glacier of Uttar Kashi in locations – spread across Delhi stretch during March–April and

2

Uttarakhand. It enters the capital territory Delhi at Palla vil- September–October, 2012 and 2013 – bacterial screening was

lage 15 km upstream of Wazirabad barrage, that acts as a done by spreading 100 ␮L of serially diluted samples on Mac-

reservoir accounting for more than 70% of Delhi’s water sup- Conkey agar plates supplemented with cefotaxime (4 ␮g/mL).

ply. Moreover, it is mainly the Delhi stretch of river Yamuna Non-duplicate bacterial colonies with distinct colony mor-

that receives high amount of discharge in terms of domestic phology were selected and subjected to characterization by

sewage and industrial waste from urban centres and there are IMViC test and 16S rRNA gene sequence analysis.

also various industrial settings established in and around the

3

capital city. Providing a better niche for their establishment, Antimicrobial susceptibility test

pollution of aquatic environment selects bacteria by causing

exchange of genetic determinants as well as acquisition of The antibiotic susceptibility test was done by disc diffu-

resistance traits crucial for it to thrive under the increasing sion method following Clinical and Laboratory Standards

18

pressure of pollutants. There are several reports pertaining to Institute guidelines. All isolates were subjected to drug

the dissemination of antimicrobial resistance genes, and it is susceptibility test against 21 antibiotics as described

19

known that their related genetic factors originate and are later previously. Representing 13 different categories that

transferred, from one bacteria to other bacteria present at a includes cephalosporins, carbapenems, aminoglycosides and

4–6

distant place. fluoroquinolones. Multiple Antibiotic Resistance (MAR) index

20

Development of resistance in bacteria makes it potent was calculated, as described by Krumperman. Categoriza-

to compete with the sensitive ones, particularly in selec- tion of bacterial isolates into, multidrug resistant, extensively

tive environments. Co-resistance to antibiotics and metals drug resistant and pandrug resistant, was strictly done on

share common structural and functional strategies, con- the basis of their resistance profiles following guidelines

7,8

21

ferred by chromosomal or mobile genetic elements. There of Magiorakos et al. Minimum inhibitory concentration

is an increasing concern regarding potential of metal con- (MIC) of antibiotics was determined by broth micro-dilution

taminated environment, acting as a pool, for sequestering method using Luria Bertani (LB) broth and E-test following

antibiotic resistance genes, both in environmental and clinical CLSI guidelines with minor modifications in the culture

9,10

isolates. Studies reporting assembly of resistance determi- medium. Similarly, resistance tests against heavy metals

nants on the same genetic element, insinuates towards the were determined by broth micro-dilution method by using

need to elucidate the role of polluted environment in acqui- different concentrations of heavy metal salts of cadmium,

sition of resistance in both pathogenic and non-pathogenic copper, chromium, mercury, lead and zinc. The cultures

11–14 ◦

isolates of aquatic habitats. Exploring the resistance pro- (96-well culture plates) were incubated at 37 C for 14–18 h

files among microbes inhabiting the aquatic ecosystem in and Optical Density (OD) was measured at 600 nm. The

urban areas, is of significant importance, as they directly or minimum concentration of each antibiotic and heavy metal

indirectly influence the sanitation outlook of the area. The salt inhibiting the growth of isolate was considered as its

presence of mercury and other heavy metals in water samples MIC. An isolate designated in this study as MRC11, showing

collected from river Yamuna has been reported in the studies broader resistance to different class of antibiotics and heavy

15 16

of Sehgal et al. and Malik et al. Tolerance of mercury at dif- metals, was selected for further studies.

ferent concentrations, suggests operation of different modes

of detoxification encoded by genes located on mer operon

Phenotypic screening for ESBL production

among bacterial inhabitants of the polluted environment.

Additionally, mer operon genes have been frequently observed

Following susceptibility test to 3rd generation cephalosporins

12,17

to be genetically linked to antibiotic resistance genes.

[ceftazidime (CAZ), cefotaxime (CTX) and ceftriaxone (CTR)],

Taken together, investigation of mer operon genes as a rep-

selected bacterial isolates were screened for the production

resentation of resistance to mercury, becomes pre-requisite

of extended spectrum ␤-lactamase enzyme(s) by Phenotypic

to determine its role in selection and survival of bacterial

Disc Confirmatory Test (PDCT) on Mueller-Hinton Agar plates

isolates in the polluted environments. Thus, in the present 19

as described earlier. E. coli ATCC 25922 and Klebsiella pneumo-

study, cefotaxime resistant Escherichia coli isolates from Delhi

niae ATCC 700603 were used as negative and positive controls,

stretch of river Yamuna were assessed for its susceptibility respectively.

b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480 473

Determination of ESBL and mer operon genes from trans-conjugant was used as template for PCR to check

the presence of ESBLs and mer operon genes.

The cefotaxime resistant isolate was screened for ESBL

determinants (bla , bla and bla ) and carbapenem

TEM SHV CTX-M Growth kinetics

resistance genes (blaNDM-1, blaOXA-23 and blaOXA-48) along with

the determinants of mer operon genes (merB, merP and merT),

Effect of antibiotic and heavy metal, separately and also in

that confer resistance against mercury. Using gene specific

combination, was studied by measuring growth parameters at

primers (Table 1), PCR reactions were carried out as described

regular intervals in nutrient media and subsequent plotting of

19

earlier. Amplified PCR products were sequenced directly

values in form of growth curves. For analysis of the growth pat-

using automated DNA sequencer at Xcelris Labs Ltd., Gujarat,

tern, secondary culture of the isolate was setup in four culture

India. DNA sequence data was analyzed for homology using

flasks containing (a) LB broth, (b) LB broth supplemented with

BLAST online search engine at NCBI (National Centre for

10 ␮g/mL cefotaxime, (c) LB broth supplemented with 2 ␮g/mL

Biotechnology Information).

HgCl2, and (d) LB broth supplemented with 10 ␮g/mL cefo-

taxime + 2 ␮g/mL HgCl2. The cultures were grown aerobically

till saturation, in an automated incubator shaker (Scigenics,

Localization of resistance genes by plasmid typing

◦

India) set at 150 rpm and 37 C. Similar experimental condi-

tions were set up for the test isolate, its trans-conjugant, also

Localization of resistance genes was performed by conju-

for negative (E. coli strain ATCC 25922) and positive controls

gation studies followed by screening through PCR based

(E. coli isolate positive for ESBL production).

amplification. ESBL producing test isolate was designated as

donor, and a plasmid free, sodium azide resistant strain of

R

E. coli (J53 Az ) sensitive to 21 antibiotics, served as recip- Biofilm formation assay

ient in the conjugation studies. Briefly, donor and recipient

cells in a ratio of 1:2 were mixed in 5 mL LB broth and incu- Biofilm formation was investigated by crystal violet assay

◦ 22

bated for 48 h at 37 C without shaking. The trans-conjugants following protocol described by O’Toole and Kolter. Briefly,

were selected on nutrient agar plates supplemented with bacterial culture with OD600 of 0.1 was diluted in tryptic soy

cefotaxime (10 ␮g/mL) and sodium azide (100 ␮g/mL). The fre- broth (TSB) by adding 100 ␮L broth to 100 ␮L culture (1:1 ratio).

quency of plasmid transfer was calculated with respect to Polystyrene microtitre plates (96-well) were inoculated with

◦

the donor cells. An E. coli isolate positive for ESBL produc- 200 ␮L of diluted culture per well and incubated at 37 C for

tion was used as a control. In order to study co-transfer 48 h. Following this, cultures were discarded and wells were

of multiple resistance markers in the conjugant, antibiotic gently washed four times with 200 ␮L of sterile phosphate

profiling was performed against 21 antibiotics. Transfer of buffer saline (PBS, pH 7.2) to remove loosely associated bacte-

complete conjugative plasmid was confirmed from test iso- ria. Cells adhered to the wall were stained with 200 ␮L of

late and trans-conjugant by plasmid isolation using alkaline 0.1% (w/v) crystal violet at room temperature for 20 min. It

lysis method, followed by size determination by performing was followed by a wash with PBS. The crystal violet that

agarose gel electrophoresis. Furthermore, the isolated plasmid stained the cells was solubilized in 200 ␮L of 95% (v/v) ethanol.

Table 1 – Primer sequences used for gene amplification.

Primer Sequence Product size (bp) Reference

 

ID-F 5 -GGCGGACGGGTGAGTAATG-3 649 [19]  

ID-R 5 -ATCCTGTTTGCTCCCCACG-3  

TEM-F 5 -ATGAGTATTCAACATTTCCGTGT-3 861  

TEM-R 5 -TTACCAATGCTTAATCAGTGAGG-3  

SHV-F 5 ATGCGTTATATTCGCCTGTGTATTATCTCCC-3 860  

SHV-R 5 -TTAGCGTTGCCAGTGCTCGATCAG-3  

CTX-MF 5 -SCVATGTGCAGYACCAGTAA-3 480  

CTX-MR 5 -GCTGCCGGTYTTATCVCC-3  

merPF 5 -ATGAAGAAACTGTTTGCCTCC-3 276  

merPR 5 -TCACTGCTTGACGCTGGACG-3  

merTF 5 -TTAATAGAAAAATGGAACGAC-3 351  

merTR 5 -ATGTCTGAACCACAAAACGGG-3 merBF 5-ATGAAGCTCGCCCCATATATTTTAG-3 667  

merBR 5 -TCACGGTGTCCTAGATGACATGG-3  

NDM1F 5 -GCATAAGTCGCAATCCCCG-3 194 This  

NDM1R 5 -GGTTTGATCGTCAGGGATGG-3 study  

OXA23F 5 -GAAGCCATGAAGCTTTCTG-3 200  

OXA23R 5 -GTATGTGCTAATTGGGAAACA-3  

OXA48F 5 -CCAATAGCTTGATCGCCCTC-3 209  

OXA48R 5 -CCATAATCGAAAGCATGTAGC-3

474 b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480

Table 2 – Antimicrobial profiling of donor E. coli MRC11 strain and E. coli J53 trans-conjugant.

S. no. Antibiotic MRC11 MRC11tra ns-conjugant

ZOI R/I/S ZOI R/I/S (mm ) (mm)

1. Ampicillin (AMP) 6 R 6 R

2. Ampicillin + Sulbactam (A/S) 6 R 15 I

3. Amikacin (AK) 16 I 26 S

4. Aztreonam (AT) 8 R 8 R

5. Ceftazidime (CAZ) 6 R 21 S

6. Ceftazidime + Clavulanic Acid (CAC) 8 R 27 S

7. Cefotaxime (CTX) 6 R 14 R

8. Cefotaxime + Clavulanic Acid (CEC) 8 R 19 S

9. Cefotetan (CTN) 6 R 17 S

10. Cefoxitin (CX) 6 R 6 R

11. Cefuroxime (CXM) 6 R 6 R

12. Cefazolin (CZ) 6 R 6 R

13. Ciprofloxacin (CIP) 6 R 15 R

14. Ertapenem (ETP) 17 R 6 R

15. Imipenem (IPM) 20 I 12 R

16. Levofloxacin (LE) 10 R 33 S

17. Ofloxacin (OF) 6 R 32 S

18. Polymyxin B (PB) 9 R 18 S

19. Rifampicin (RIF) 9 R 20 S

20. Trimethoprim (TR) 6 R 18 S

21. Tobramycin (TOB) 15 S 23 S

ZOI, Zone of Inhibition; S, sensitive is represented by Black color; I, intermediate as Green; R, resistant by Red color.

Samples were incubated for 20 min at room temperature and Binary toxicity assay

biofilm formation was quantified by measuring OD600 in an

ELISA reader (Thermo Scientific, MultiscanGo). Change in the The multidrug resistant isolate was further subjected to

amount of biofilm formation under different media compo- combined toxic effect of heavy metals and antibiotics. The

sition was observed by supplementing TSB with cefotaxime petriplates containing MHA supplemented with different con-

␮ ␮

(2 g/mL), mercuric chloride (0.2 g/mL) and their combina- centrations of heavy metals was inoculated and spread with

tion, and the wells containing TSB alone were used as blanks. the bacterial culture. Combined toxicity was determined by

The percentage of biofilm formation was calculated by the measuring the zone of inhibition after placing antibiotic

formula: discs at appropriate distance and incubation for 12–14 h at

  ◦

[Test OD − Blank OD ] 37 C. On the basis of susceptibility, four antibiotics with

600 nm 600 nm

Percentage of biofilm formation = × 100

Blank OD

600 nm three different modes of action: amikacin (aminoglycoside),

ciprofloxacin (fluoroquinolone), cefoxitin (chephamycin) and

Taking a note of sensitivity to antibiotics, E. coli isolate

imipenem (carbapenem), were included in this study. Each

known for ESBL production was used as a positive control and

antibiotic was used at four different concentrations, in com-

Pseudomonas aeruginosa ATCC 9027, exhibiting sensitivity to

bination with six heavy metal salts [CdCl2 (175, 75, 50, 5 mg/L),

antibiotics along with biofilm formation capability, was used

CuCl2 (250, 100, 50, 5 mg/L), CrO3 (125, 30, 15, 5 mg/L), HgCl2

as a negative control, in assessing the behaviour of the test

(15, 7.5, 1, 0.1 mg/L), Pb(CH COO) (250, 125, 75, 10 mg/L) and isolate. 3 2

b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480 475

ZnSO4 (500, 250, 100, 50 mg/L)]. The concentrations of heavy operon (merB, merP and merT) genes associated in imparting

metals were selected on the basis of MIC obtained for each resistance against organic as well as inorganic mercury, gives

metal. an indication about the presence of both broad and narrow

spectrum mer operon in the isolate. Results

Plasmid profiling

The bacterial isolate −2

Conjugation efficiency of the magnitude of 3.5 × 10 and

−5

6.1 × 10 trans-conjugant per donor cell was observed for

A total of 75 non-duplicate E. coli isolates were collected

E. coli MRC11 and control, respectively. Such high conjuga-

from water samples in heavily polluted Delhi stretch of

tion efficiency gives an indication of higher possibility of

river Yamuna. Screening for ESBL production revealed 34

transfer of plasmid borne genes. Successful transfer of resis-

E. coli isolates to be positive for ESBL production. Out of

tance determinants leads to the transformation of sensitive

these ESBL positive E. coli isolates, an isolate (annotated as

recipient into well acquaint resistant phenotype (Table 2).

MRC11) collected on March 26, 2013 from Old Railway Bridge,

An ∼22 kb plasmid DNA isolated from trans-conjugant

100–150 m downstream of Civil Mill Drain falling into the river

◦   ◦   confirms complete transfer of conjugative plasmid via con-

(77 14 51.42 E and 28 39 41.77 N), displayed high incidences

jugation. PCR showed positive amplification corresponding

of resistance, therefore it was chosen for MDR studies against

to blaTEM-116 along with mer operon (merP and merT) genes

wide range of antibiotics and heavy metals. In addition to the

from trans-conjugant plasmid. The trans-conjugant was

identification by assessment of the biochemical parameters,

resistant against cephalosporins, monobactams, carbapen-

recognition based on 16S ribosomal RNA gene sequence, con-

ems and ciprofloxacin. Higher MIC values of cephalosporins

firmed the isolate to be an E. coli (NCBI Accession # KT428597).

(>256 ␮g/mL), carbenicillins (>512 ␮g/mL) and ciprofloxacin

(5 ␮g/mL) demonstrate high resistance range of the trans-

Antimicrobial susceptibility

conjugant (Table 3).

By checking susceptibility of the bacterial isolates against

Growth studies

antibiotics, it was observed that E. coli MRC11 exhibits

resistance against broad range of antibiotics represent-

Growth kinetic studies reflect the effect of antimicrobials

ing ␤-lactams, monobactams, fluoroquinolones, polymyxins,

(antibiotics and heavy metals), either alone or in combina-

rifampicins and trimethoprim (Table 2). Phenotypic determi-

tion, on the growth pattern of E. coli MRC11. Growth of E. coli

nation of ESBL production revealed complete absence of zone

MRC11 under normal conditions showed a lag phase of about

of inhibition for third generation cephalosporins, individually,

2 h followed by a sharp increase in biomass during the log

as well as in combination with the inhibitor i.e. clavulanic

phase. On supplementing media with cefotaxime, no signifi-

acid. The isolate was found to be sensitive to tobramycin

cant change in growth was observed for the isolate (Fig. 1A and

and amikacin. In summary, resistance displays peaks for

B). However, supplementing media with mercuric chloride, an

amoxicillin and carbenicilin (MIC >512 ␮g/mL) followed by

extended lag phase of 9 h was observed for both test isolate

aztreonam, cefotaxime and ceftazidime (MIC >256 ␮g/mL),

and its trans-conjugant. Contrarily, to the expected extension

ofloxacin (32 ␮g/mL) and ciprofloxacin (30 ␮g/mL), respectively

in lag phase following supplementation of both cefotaxime

(Table 3). The isolate showed complete resistance to ofloxacin

and mercuric chloride, a decline in the lag phase by about 2 h (a

and trimethoprim, comprising choice of drugs available for

net lag phase of 7 h) was observed, which may be attributed to

the treatment of UTI infections. Besides, MIC value of 2 ␮g/mL

the resistance determinants conferring co-resistance (Fig. 1A).

(susceptible range) for amikacin, ZOI was observed to be in +

However, growth in the case of positive control (E. coli ESBL )

intermediate category (16 mm). Higher MAR index (0.95) for

was compromised following treatment with mixture of cefo-

the E. coli MRC11 gives an indication of its portrayal as a thera-

taxime and mercuric chloride (Fig. 1C). Compared to this,

peutic challenge. Simultaneously, by looking at the resistance

negative control displays no growth patterns after the appli-

against different heavy metals, it was observed that E. coli

cation of cefotaxime and mercuric chloride, either alone or in

MRC11 exhibits a very high resistance profile for metals, even

combination (Fig. 1D).

higher than the prescribed permissible limit which is consid-

ered tolerable for humans (Table 4).

Biofilm formation assay

ESBL and mer operon determinants Biofilm formation in nutrient media was around 59% in E. coli

MRC11 and 79% in P. aeruginosa ATCC 9027 (Fig. 2). When

Analysis of DNA sequence data revealed presence of two dif- assessed under different medium conditions, compared to the

ferent variants of blaTEM (blaTEM-1 and blaTEM-116; NCBI control, biofilm formation was found to follow a different pat-

Accession # KU376497 and KT428598) gene in E. coli MRC11 tern for E. coli MRC11 isolate. Maximum biofilm formation in

that is credited towards contribution of resistance to ␤- the medium supplemented with mercuric chloride was seen

lactams. As amplification could not be achieved for other two in E. coli MRC11, also biofilm formation for the control isolate is

ESBL (blaSHV and blaCTX-M) genes and carbapenem resis- maximum in the media which is supplemented with mercuric

tance imparting genes (blaNDM-1, blaOXA-23 and blaOXA-48), it was chloride and cefotaxime. The growth patterns of the biofilm

concluded that the isolate is devoid of them. Presence of mer formation for control is, higher in media supplemented with

476 b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480

Table 3 – MIC values of antibiotics towards E. coli MRC11 isolate and E. coli J53 trans-conjugant.

Antibiotic E. co li MRC11 E. coli J53 E. coli MRC11trans-

con jugant

MIC S/I/R MIC S/I/R MIC S/I/R (µg/mL) (µg/mL) (µg/mL)

Amikacin (AK) 2.0 S 0.5 S 0.1 S

Kanamycin 64 R 1.0 S 1.0 S (KAN) Ciprofloxacin 30 R 0.016 S 5.0 R

(CIP)

Carbenicil lin R 64 R R

(CRB)

Amox icillin R 2 S R (AMX) Rifampicin (RIF) 64 R 2 S 0.25 S

Ertapenem (ETP) 8.0 R 0.03 S 6.0 R

Imipenem (IPM) 4.0 R 0.25 S 4.0 R

Ceftriazon e 16 R 0.1 S R

(CTR)

Oflox acin (OF) 32 R 0.06 S 0.06 S

Aztreonam (AT) R 0.094 S 2.0 S

Cefotaxime R 0.125 S R

(CTX)

Ceftazidime R 0.5 S R

(CAZ)

MIC, Minimum Inhibitory Concentration; S, sensitive is represented by Black color; I, intermediate as Green; R, resistant by Red color.

mixture of mercuric chloride and cefotaxime, which is fol-

Table 4 – MIC values of heavy metals towards E. coli

lowed by the growth seen in presence of mercuric chloride

MRC11 isolate.

(68%) and the least growth (65.6%) occurs in the presence of

Heavy metal MIC (mg/L)

cefotaxime alone. However, highest growth of the biofilm can

be seen in the media supplemented with metal alone, which Cadmium chloride 250

Copper chloride 500

followed by the growth in media supplemented with cefo-

Chromium (III) oxide 125

taxime (56.2%) and the least growth is seen in the presence

Mercuric chloride 7.8

of mercuric chloride and cefotaxime (Fig. 2).

Lead acetate 500

Zinc sulphate 1000

Dual resistance towards antibiotics and heavy metals

MIC, Minimum Inhibitory Concentration.

Keeping in mind that the treatment with metal and antibi-

of E. coli MRC11 towards sensitive range. Zinc in combination

otics effects bacterial susceptibility, combined toxicity assays

with imipenem, showed a positive impact at 500 mg/L, with lit-

for E. coli MRC11 were performed. The detailed results

tle or no effect at 250 mg/L, 100 mg/L and 50 mg/L. On one side,

of the combined toxicity test are summarized in Fig. 3.

where imipenem showed positive correlation with lead and

E. coli MRC11 isolate which showed intermediate phenotype

zinc, no differences were observed at varying concentrations

towards amikacin, exhibits sensitive phenotype in presence

of chromium and mercury. Higher concentrations of cadmium

of cadmium, copper and chromium. Increase in the zone of

and copper have an antagonistic effect when used in com-

inhibition was observed when amikacin and copper, both

bination with imipenem. The susceptibility pattern changed

were used together, at a concentration of 5 mg/L and 50 mg/L,

from sensitive to resistant in case of cadmium and from sen-

respectively. While as no such change in the zone diameter

sitive to intermediate for copper. Additionally, E. coli MRC11

was observed at 100 mg/L and 250 mg/L concentration, which

isolate displayed complete phenotypic resistance to different

gives a clear indication about their synergistic effect towards

concentrations of cadmium, copper, chromium, mercury, lead

E. coli MRC11. In a similar manner, combination of amikacin

and zinc, when used in combination with ciprofloxacin and

with zinc showed increase in its activity as observed by the

cefoxitin.

increase of zone diameter at 500 mg/L concentration, with lit-

tle or no change at the lower values. Contrary to this, mercury

at a concentration of 1 mg/mL reduced the zone diameter from Discussion

16 mm to 14 mm, thereby indicating that the two hold antag-

onistic effect. Sheer imprudence in the use of antibiotics has resulted in

Similarly, combination of lead (75 mg/L) with imipenem the emergence and dissemination of resistance determinants.

showed increase in zone diameter; that shifted the activity Recognized as a major challenge in the healthcare systems,

b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480 477 A B

2.0 2.0 1.8 1.8

1.6 1.6 1.4 1.4

1.2 1.2

600 1.0 600 1.0

OD 0.8 OD 0.8

0.6 0.6 MRC11T

MRC11

µ MRC11T + CTX 10 g/mL

MRC11 + CTX 10µg/mL

0.4 0.4 µ

MRC11T + Hg 2 g/mL

MRC11 + Hg 2µg/mL

µ µ

MRC11T + CTX 10 g/mL+ Hg 2 g/mL 0.2 MRC11 + CTX 10µg/mL+ Hg 2µg/mL 0.2

0.0 0.0

0 1 2 3 4 5 6 7 8 9 10 11 12 13 1415 16 17 18 19 20 21 22 23 0 1 2 3 4 5 6 7 8 9 1011121314151617181920212223 Time (h) Time (h)

CD+Control

2.0 +Control + CTX 10µg/mL 2.0

+Contro + Hg 2µg/mL

1.8 +Control + CTX 10µg/mL+ Hg 2µg/mL 1.8 1.6 1.6 1.4 1.4

1.2 1.2

600 1.0 600 1.0

OD 0.8 OD 0.8

0.6 0.6 E. coil ATCC 25922 E. coil ATCC 25922 + CTX 10 µg/mL

0.4 0.4 E. coil ATCC 25922 + Hg 2 µg/mL E. coil ATCC 25922 + CTX 10µg/mL + Hg 2µg/mL 0.2 0.2

0.0 0.0

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 0 1 2 3 4 5 6 7 8 9 10 11 12 13 1415 16 17 1819 20 2122 23

Time (h) Time (h)

Fig. 1 – Growth pattern of bacteria under different culture conditions. (A) Isolate E. coli MRC11, (B) isolate E. coli J53

trans-conjugant, (C) positive control (i.e., isolate resistant to ␤-lactams and heavy metals) and (D) sensitive control E. coli

ATCC 25922. The X axis depicts time in hours where as Y axis represents bacterial growth presented in terms of OD at

600 nm.

85

80

75

70

65 P. aeruginosa ATCC 9027 60 E. coil MRC11 55 Positive control Biofilm formation(%)

50

45

40

S1 S2 S3 S4

Fig. 2 – Biofilm formation in different medium conditions. S1, Luria broth; S2, Luria broth supplemented with 10 mg/L

cefotaxime; S3, Luria broth supplemented with 2 mg/L mercuric chloride; S4, Luria broth supplemented with 10 mg/L

cefotaxime + 2 mg/L mercuric chloride.

bacteria have developed mechanisms that render antibiotics the development of newer gene variants with broader activ-

useless against infections. Polluted environment exacerbates ity, ultimately increasing the resistance burden in different

19,23,24

the situation by creating a selective pressure favouring acqui- environmental conditions. One such location identified

sition of resistance genes and point mutations that leads to as Delhi stretch of river Yamuna, representing only 2% of the

478 b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480

80 60 Ciprofloxacin Cefoxitin Amikacin Imipenem Ciprofloxacin Cefoxitin Amikacin Imipenem S 50 S 60 S I I I I I I 40 I S S S S 40 S I I I I 30

20 20

0 10 0 -20 R R R R RR R R

R R

-10

Changerate of ZOI (%) Changerate

WM C1 C2 C3 C4 Changerate of ZOI (%) Changerate -40 R R R R R R R R R R -20 R WM C1 C2 C3 C4 -60 -30 Concentration of cadmium chloride (mg/L) Concentration of copper chloride (mg/L)

50 50 Ciprofloxacin Cefoxitin Amikacin I Imipenem Ciprofloxacin Cefoxitin Amikacin Imipenem I I 40 I I I 40 I S I I I I I 30 30 20

20 10 NG NG 10 0

WM C1 C2 C3 C4

NG -10

0 R R R R R R

Changerate of ZOI (%) Changerate -20 Changerate of ZOI (%) Changerate

-10

-30

R R R R R R R R WM C1 C2 C3 C4 -20 R -40 Concentration of chromium (III) oxide (mg/L) Concentration of mercuric chloride (mg/L)

50 60 Ciprofloxacin Cefoxitin Amikacin Imipenem Ciprofloxacin Cefoxitin Amikacin Imipenem S S S 50 I I I I S

40 I

40

S S I I I I I I I I 30 30 20 WM C1 C2 C3 C4 20 10

10 0

WM C3

C1 C2 C4 -10

R R R R R R R R R R 0 -20 Changerate of ZOI (%) Changerate

Changerate of ZOI (%) Changerate -30

-10

R R R R R R R R R R -40 R -20 -50

Concentration of lead acetate (mg/L) Concentration of zinc sulphate (mg/L)

Fig. 3 – Effect of heavy metal on antibiotic resistance pattern of E. coli MRC11. R (resistant), I (intermediate) and S (sensitive)

category as per CLSI criteria were labelled on each column. WM represented sensitivity in absence of heavy metal; C1, C2,

C3 and C4 were concentrations of heavy metals in the medium, that were, Cd (175, 75, 50, 5 mg/L), Cu (250, 100, 50, 5 mg/L),

Cr (125, 30, 15, 5 mg/L), Hg (15, 7.5, 1, 0.1 mg/L), Pb (250, 125, 75, 10 mg/L) and Zn (500, 250, 100, 50 mg/L) respectively. NG

showed no growth of E. coli MRC11 isolate at that particular concentration of heavy metal.

total catchment area of the river, gets discharge from around grown to an alarming condition, troubling clinicians treating

2 26

22 sewage and 28 industrial clusters. This stretch is heavily them. Associated with urinary tract infections (UTIs), E. coli,

eutrophicated at several locations, where concentration of which predominates among members of Enterobacteriaceae,

27,28

almost all major heavy metals and antibiotics is well above seems to be a true community ESBL-producing pathogen.

16,25

the maximum permissible limits. Rather than acting as Delhi stretch of river Yamuna is crammed with multidrug

a graveyard for bacterial isolates where multiple antimicro- resistant E. coli isolates, out of which 45% were ESBL produc-

bial agents work simultaneously, the polluted river Yamuna ers. Among these, one isolate, E. coli MRC11 exhibits resistance

is crammed with isolates harbouring clusters of resistance against 20 antibiotics tested. Having an MAR index of 0.95

genes. and high MIC values (>256 ␮g/mL) to different antibiotics,

Extended spectrum ␤-lactamase enzymes are reported consortium of available drugs seems inadequate in dealing

as the main contributing factors to high level of resistance with its infections. Similar to study carried out in Thailand

observed among majority of the multidrug resistant bacte- where Pornsinchai et al. reported co-presence of blaTEM-1 and

rial isolates. As such, rapid emergence of multidrug resistance blaTEM-116 genes in clinical isolates of E. coli and K. pneumo-

among opportunistic pathogens that cause endocarditis, bac- niae, E. coli MRC11 isolate was found to harbour two different

29

teraemia, sepsis, urinary tract and other infections, has now variants of blaTEM gene. Carbapenems being a choice of drug

b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480 479

30

against ESBL producing isolates, E. coli MRC11 exhibits resis- E. coli MRC11, compared to control isolate. The plausible expla-

tance against ertapenem and imipenem. Absence of amplicon nation for these observations could come from the fact that

for blaNDM-1, blaOXA-23 and blaOXA-48 genes suggest presence of potential expression of the ESBL gene on conjugative plas-

other determinants imparting carbapenem resistance. Simi- mid, might confer resistance to cephalosporins-resulting in

lar to study of Rath et al. reporting prevalence of amikacin well adapted phenotype – so that it can function or grow in a

resistance among nosocomial and community acquired E. coli normal manner, without any effect on the growth pattern and

isolates, E. coli MRC11 in our study was seen to be resistant to also without significant induction of biofilm formation. Taken

31

amikacin. Fluoroquinolones constitute the most widely used together, detection of E. coli MRC11 in aquatic environment

anti-bacterials that are effective in treating UTIs caused by presents a risk factor for dissemination of resistance genes

32

E. coli. Wide spread resistance to levofloxacin and ofloxacin to co-inhabitants including potential pathogens which pose

in E. coli MRC11 often renders fluoroquinolones of no use. serious threat to mankind. Understanding the potentialities

Resistance against one or more agents from thirteen different of multiresistant isolates to withstand effect of antimicrobials

classes of antibiotics portrays pandrug resistance phenotype will help in strategic development of new and more effective

to E. coli MRC11 isolate. This is in concordance with the studies antimicrobial agents so as to “reset the clock” for resistance

that advocate high incidence of multidrug resistance among level and infection control. To achieve this, further studies are

bacterial isolates collected from water streams influenced by required to evaluate the role of polluted environment in resis-

33,34

industrial pollution and municipal sewage. tance spread and to determine the factors responsible that

Clustering of the resistance genes and their mobilization attribute bacteria with the pandrug resistant phenotype.

through horizontal gene transfer armours bacteria with resis-

35,36

tance against wide range of antibiotics. Complying with

Conflicts of interest

the reports, trans-conjugant of E. coli MRC11 that apprehends

presence of resistance determinants on conjugative plasmid

Authors declare that no conflict of interest exist with this pub-

exhibits resistance to wide variety of antibiotics, in addition

lication.

of having ESBL phenotype possibly attributed to presence of

other resistance genes that are not examined in this study.

Moreover, in our work it was observed that, trans-conjugant

Acknowledgement

of E. coli MRC11 carries mer operon (merP and merT) genes.

These results are in corroboration with the work of McIn-

Mudsser Azam acknowledges Council of Scientific & Industrial

tosh et al. reporting genetic linkage between mer operon

17 Research (CSIR), India for financial support as Senior Research

with that of antibiotic resistance genes. In combined tox-

Fellowship (09/466(0136)/2011-EMR-I).

icity test, amikacin in amalgamation with cadmium chloride

and chromium (III) oxide showed synergistic effect possibly

due to chemical reaction between the effective concentra- r e f e r e n c e s

tions of heavy metals and antibiotics, and/or presence of

heavy metal that elicit stress response, impairing protein

(enzyme) synthesis, that impart resistance to the bacteria.

1. Bush K, Jacoby GA. Updated functional classification of

Contrarily, when antimicrobial activity of the combination ␤-lactamases. Antimicrob Agents Chemother.

(antibiotics plus metal) is no more equal to or greater than par- 2010;54(3):969–976.

2. Dhillon MK, George MP, Mishra S. Water quality of River

ent substance, and/or coagulation occurs, it is considered as

Yamuna – Delhi stretch. Int J Environ Sci. 2013;3(5):1416–1423.

antagonistic effect as observed in the case of imipenem with

3. Misra AK. A River about to Die: Yamuna. J Water Resour Prot.

cadmium chloride (175 mg/L) and zinc (50 mg/L), and amikacin

2010;2:489–500.

with mercuric chloride (1 mg/L) combinations. These results

4. Mazodier P, Davies J. Gene transfer between distantly related

are complying with the observations of studies carried out bacteria. Annu Rev Genet. 1991;25:147–171.

37

by Zhou et al. that reports changed susceptibility of Psu- 5. Allen HK, Donato J, Wang HH, et al. Call of the wild:

domonas fluorescens to antibiotics when co-exposed to heavy antibiotic resistance genes in natural environments. Nat Rev

metals. Microbiol. 2010;8:251–259.

6. Costa PMD, Loureiro L, Matos AJF. Transfer of

Knowing the fact that blaTEM genes impart resistance to

multidrug-resistant bacteria between intermingled

␤-lactams, growth of the isolate E. coli MRC11 under study

ecological niches: the interface between humans, animals

was found insensitive to cefotaxime (10 ␮g/mL). Accordingly,

and the environment. Int J Environ Res Public Health.

E. coli MRC11 transformant harbouring resistance genes on 2013;10(1):278–294.

plasmid follows almost similar growth pattern. While, pres- 7. Roberts MC. Update on acquired tetracycline resistance

genes. FEMS Microbiol Lett. 2005;245:195–203.

ence of heavy metal in media reduces the microbial activity

8. Baker-Austin C, Wright MS, Stepanauskas R, et al.

to a large extent, thereby increases lag phase, later replenish-

Co-selection of antibiotic and metal resistance. Trends

ing the growth possibly due to co-resistance or detoxification

Microbiol. 2006;14:176–182.

strategies generated in the cell by the proteins encoded by

9. Stepanauskas R, Glenn TC, Jagoe CH, et al. Co-selection for

the set of mer operon genes. Similar growth patterns, with-

microbial resistance to metals and antibiotics in freshwater

out any effect of antibiotic in medium, on the growth kinetics microcosms. Environ Microbiol. 2006;8:1510–1514.

of antibiotic resistant bacteria, were also demonstrated by 10. Seiler C, Berendonk TU. Heavy metal driven co-selection of

38 antibiotic resistance in soil and water bodies impacted by

Drummond et al. Also, no significant increase in biofilm for-

agriculture and aquaculture. Front Microbiol. 2012;3(399):1–10.

mation was observed in presence of antimicrobial agent for

480 b r a z i l i a n j o u r n a l o f m i c r o b i o l o g y 4 9 (2 0 1 8) 471–480

11. Summers AO, Wireman J, Vimy MJ, et al. Mercury released 24. Kummerer K. Antibiotics in the aquatic environment – a

from dental silver fillings provokes an increase in review – Part I. Chemosphere. 2009;75:417–434.

mercury-resistant and antibiotic resistant bacteria in oral 25. Mutiyar PK, Mittal AK. Occurrences and fate of selected

and intestinal floras of primates. Antimicrob Agents human antibiotics in influents and effluents of sewage

Chemother. 1993;37(4):825–834. treatment plant and effluent-receiving river Yamuna in

12. Wireman J, Liebert CA, Smith T, et al. Association of mercury Delhi (India). Environ Monit Assess. 2014;186(1):541–557.

resistance with antibiotic resistance in Gram-negative fecal 26. Theophel K, Schacht VJ, Schlüter M, et al. The importance of

bacteria of primates. Appl Environ Microbiol. growth kinetic analysis in determining bacterial

1997;63:4494–4503. susceptibility against antibiotics and silver nanoparticles.

13. Ghosh S, Mahapatra NR, Ramamurthy T, et al. Plasmid Front Microbiol. 2014;5(544):1–10.

curing from an acidophilic bacterium of the genus Acidocella. 27. Pitout JD, Nordmann P, Laupland KB, et al. Emergence of

FEMS Microbiol Lett. 2000;183:271–274. Enterobacteriaceae producing extended-spectrum

14. Davis IJ, Richards H, Mullany P. Isolation of silver- and beta-lactamases (ESBLs) in the community. J Antimicrob

antibiotic-resistant Enterobacter cloacae from teeth. Oral Chemother. 2005;56:52–59.

Microbiol Immunol. 2005;20:191–194. 28. Pitout JD, Laupland KB. Extended-spectrum

15. Sehgal M, Garg A, Suresh R, Dagar P. Heavy metal beta-lactamase-producing Enterobacteriaceae: an emerging

contamination in the Delhi segment of Yamuna basin. public-health concern. Lancet Infect Dis. 2008;8:159–216.

Environ Monit Assess. 2012;184:1181–1196. 29. Pornsinchai P, Chongtrakool P, Diraphat P, et al. Emergency

16. Malik D, Singh S, Thakur J, et al. Heavy metal pollution of the room: an unrecognized source of extended-spectrum

Yamuna river: an introspection. Int J Curr Microbiol Appl Sci. ␤-lactamase producing Escherichia coli and Klebsiella

2014;3(10):856–863. pneumoniae. Southeast Asian J Trop Med Public Health.

17. McIntosh D, Cunningham M, Ji B, et al. Transferable, multiple 2015;46(1):51–62.

antibiotic and mercury resistance in Atlantic Canadian 30. Kumar D, Singh AK, Ali MR. Antimicrobial susceptibility

isolates of Aeromonas salmonicida subsp. salmonicida is profile of extended spectrum ␤-lactamase (ESBL) producing

associated with carriage of an IncA/C plasmid similar to the Escherichia coli from various clinical samples. Infect Dis: Res

Salmonella enterica plasmid pSN254. J Antimicrob Chemother. Treat. 2014;7:1–8.

2008;61:1221–1228. 31. Rath S, Dubey D, Sahu MC, et al. Surveillance of ESBL

18. Clinical and Laboratory Standard Institute. Performance producing multidrug resistant Escherichia coli in a teaching

Standards for Antimicrobial Testing; Twenty-Fifth Information hospital in India. Asian Pac J Trop Dis. 2014;4(2):140–149.

Supplement. Document M100-S25; 2015. 32. World Health Statistics. WHO 2014; 2014. http://apps.who.

19. Azam M, Jan AT, Haq QMR. blaCTX-M-152, a novel variant of int/iris/bitstream/10665/112738/1/9789240692671 eng.pdf.

CTX-M-group-25, identified in a study performed on the 33. McArthur JV, Tuckfield RC. Spatial patterns in antibiotic

prevalence of multidrug resistance among natural resistance among stream bacteria: effects of industrial

inhabitants of river Yamuna, India. Front Microbiol. pollution. Appl Environ Microbiol. 2000;66(9):3722–3726.

2016;7(176):1–13. 34. Korzeniewska E, Harnisz M. Extended-spectrum

20. Krumperman PH. Multiple antibiotic resistance indexing of beta-lactamase (ESBL)-positive Enterobacteriaceae in

Escherichia coli to identify high-risk sources of fecal municipal sewage and their emission to the environment. J

contamination of foods. Appl Environ Microbiol. Environ Manag. 2013;128:904–911.

1983;46(1):165–170. 35. Rowe-Magnus DA, Mazel D. The role of integrons in

21. Magiorakos AP, Srinivasan A, Carey RB, et al. antibiotic resistance gene capture. Int J Med Microbiol.

Multidrug-resistant, extensively drug-resistant and 2002;292(2):115–125.

pandrug-resistant bacteria: an international expert proposal 36. Nordmann P, Poirel L, Walsh TR, et al. The emerging NDM

for interim standard definitions for acquired resistance. Clin carbapenemases. Trends Microbiol. 2011;19:588–595.

Microbiol Infect. 2012;18(3):268–281. 37. Zhou Y, Xu YB, Xu JX, et al. Combined toxic effects of heavy

22. O’Toole GA, Kolter R. Initiation of biofilm formation in metals and antibiotics on a Pseudomonas fluorescens strain

Pseudomonas fluorescens WCS365 proceeds via multiple, ZY2 isolated from swine wastewater. Int J Mol Sci.

convergent signalling pathways: a genetic analysis. Mol 2015;16(2):2839–2850.

Microbiol. 1998;28(3):449–461. 38. Drummond LJ, Smith DGE, Poxton IR. Effects of sub-MIC

23. Sarmah AK, Meyer MT, Boxall AB. A global perspective on the concentrations of antibiotics on growth of and toxin

use, sales, exposure pathways, occurrence, fate and effects production by Clostridium difficile. J Med Microbiol.

of veterinary antibiotics (VAs) in the environment. 2003;52:1033–1038.

Chemosphere. 2006;65(5):725–759. MOLECULAR MEDICINE

Bench to Bedside and Beyond

Mycobacterium indicus pranii

The continuing journey of Professor G. P. Talwar Festschrift at Ninety Molecular Medicine : Bench to Bedside and Beyond

Functional characterization of mycobacterial protein PE18/Rv1788 and its ortholog MIP_03868: A comparative analysis

1 1 Anwar Alam , Ashutosh Kumar , Sakshi Kohli1, Yadvir Singh1, Khushbu Sharma1, Nasreen Z. Ehtesham2 and 1,3,4 Seyed E. Hasnain * 1Molecular Infection and Functional Biology Lab, Kusuma School of Biological Sciences, Indian Institute of Technology-Delhi, New Delhi 2National Institute of Pathology, Safdarjang Hospital Campus, New Delhi 3JH-Institute of Molecular Medicine, Hamdard Nagar, New Delhi 4Dr Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Professor CR Rao Road, Hyderabad Email: [email protected] *Corresponding author Abstract Gradual expansion of PE/PPE family of proteins from non-pathogenic to pathogenic mycobacteria and their association with other virulent factors supports their vivid role in mycobacterial virulence and pathogenesis. The evolution of PE/PPE family proteins is found to be synchronized to that of Esx or Type VII secretion systems from the saprophytic slow growing mycobacteria to the pathogenic fast growing mycobacteria down the geological timescale. The recently evolved Esx5, an exclusive feature of the slow growing pathogenic mycobacteria, serves as the channel for the transport of various associated as well as non- associated PE/PPE proteins which lack classical signal peptides. The cell wall association or secretion of PE/PPE proteins contributes to their multiple roles in mycobacterial virulence including host immune evasion, immune-modulation and antigenic variation. The esx-5 region of Mycobacterium tuberculosis (M. tb) (Rv1782- Rv1798) harbors 2 pe (pe18, 19) and 3 ppe (ppe25, 26, 27) genes. A perturbed state is achieved in determining the potent but complex role of these family of proteins because of gene cooption as the orthologs and paralogs may hold significantly distinct role in mycobacterial species other than M. tb complex. M-coffee alignment showed a difference of 8 amino acid residues in orthologous proteins PE18/Rv1788 and MIP_03868 respectively. Out of 8 amino acids 3 were found to be conserved whereas 4 amino acids were semi-conserved. Furthermore, phosphorylation sites were found to be altered between the two homologs as determined by NetPhos software. String analysis suggests interaction of PE18 with PPE25 and PPE26 of Esx5. The localization of both the proteins was found to be different at sub-cellular level by fluorescence microscopy, where PE18 is found to migrate to mitochondria with likely functional involvement in respiration and intermediary metabolism, on

107 Seyed E. Hasnain et al. the contrary its ortholog MIP_03868 was not found to co-localize to the mitochondria. Such analyses provide clues to explore the differences in the functionality of these proteins based on the differences in just 8 amino acids or their combinations. Introduction Mycobacterium tuberculosis infects nearly one third of the global population causing clinical symptoms of Tuberculosis (TB) in 10.4 million people annually and accounts for 1.4 million deaths [1]. In 80% of the infected individuals TB persists in latent form, and emerges into active form under immunocompromised conditions. The risk factors for TB, such as under nutrition, HIV, smoking, air pollution and diabetes mellitus are prevalent in India and we are amongst the top six countries that account for 60% of all new cases. Rapid emergence of strains which are resilient towards current chemotherapeutic drugs adds new dimensions to the existing challenges such as poor prognosis of TB due to prolonged drug treatment and side effects. Further, drug tolerance enables phenotypic variant subpopulation of persister cells within Mycobacterial biofilms to sustain higher minimum inhibitory concentration (MIC) and is responsible for recalcitrance of TB infections in clinical relapse cases [2]. A higher success rate in treating TB requires early and correct diagnosis of M. tb. Diagnostic tools such as Bactec 460TB system provided initial leads about association of MDR with mutations in genes for rpoB (rifampicin), katG, inhA (isoniazid), gyrA, gyrB (ofloxacin) and rpsS (streptomycin) [3]. Subsequently, DNA fingerprinting methods using IS6110 and a combination of two rapid PCR-based typing methods- spoligotyping and VNTR, helped in identifying important genotypes, such as East African Indian, Delhi type, Beijing, Haarlem etc., prevalent among Indian populations [4]. Whole genome sequencing of Mycobacterium proved to be a major milestone and provided impetus to annotate newer roles of genes in MDR or establish phylogenetic relationship among different mycobacterial species [5]. A critical understanding of species which are at the borderline of non- pathogenic and pathogenic forms along the phylogenetic tree could churn out crucial data as to how and why mycobacterial species acquired pathogenicity during the course of evolution. Mycobacterium indicus pranii (MIP), a saprophytic, non-pathogenic mycobacterium, possesses homologs of virulence factors of M. tb and is positioned at the transitionary position along the evolution of pathogenicity. Gene cooption and moonlighting provide the basis of acquisition of pathogenicity along the course of mycobacterial evolution [6]. Understanding the molecular mechanisms that led to acquisition of pathogenicity is of paramount importance to derive strategies for successful interventions against TB. In pursuance of this aim our group has been working on several fronts to decipher the molecular basis of pathogenicity in Mycobacterium using MIP as a model bacterium. It would help unravel host pathogen interaction and the role of host genome in conferring heterogeneity in severity of virulence, immune response and therapeutic efficacy [7]. PE/PPE proteins are also being seen as players for developing therapeutic drugs due to its distinct evolutionary

108 Molecular Medicine : Bench to Bedside and Beyond advantage in terms of antigenicity [8]. We discuss the importance of PE multigene family, unique to Mycobacterium genus, which has expanded during the evolution of pathogenicity and plays a key role in niche adaptation in mycobacteria. Non pathogenic and pathogenic mycobacteria Mycobacterium genus exemplifies host tropism and habitat diversification in terms of its variation as non-pathogenic, opportunistic and pathogenic species. These include saprophytes such as M. smegmatis, immunomodulators such as M. vaccae, opportunist pathogens such as M. avium and strict intracellular pathogens such as M. tuberculosis or M. leprae. The free living non-pathogenic bacteria which were programmed to tackle environmental pressure essentially evolved in order to adapt to niche locations within the host where the challenges are different in terms of attachment to cellular surfaces and complex immune responses. Non pathogenic bacteria that coexisted with pathogenic partners in the ecological niche acquired new genes to gain access to the host. An understanding to the genetic basis of habitat diversification and pathogenicity could prove to be effective in identifying disease related genes and vaccine candidates which may help in controlling mycobacterial infections. Complete genome sequencing of pathogenic mycobacteria have shown that evolution from saprophytic predecessor involved loss of non-essential genes and gain of few other genes that confer advantage in establishing niche within the host [9]. Non-pathogenic mycobacteria possess 1-2 times higher number of metabolism related genes, transport and regulatory protein genes as compared to the pathogenic members [10]. It is widely accepted that evolution of the tubercle bacilli from ancient environmental Proteobacteria involved events of substantial Horizontal gene transfer (HGT) [11]. Yet comparing the genome sizes of various mycobacterial species indicates gene loss is far more extensive than gene acquisition in evolution of the obligate pathogens. Mycobacterium indicus pranii: The connecting link between non-pathogens and pathogenic mycobacteria In late 80s the bacterium, initially termed as Mycobacterium ‘w’, was shown to evoke cell mediated immune responses against M. leprae [12]. Extensive work carried out by the group led by Prof. G. P. Talwar resulted in its use as an immunotherapeutic vaccine ‘Immuvac’ against leprosy and also pointed to its possible use as an adjunct to chemotherapy in category-II TB patients [13]. Whole genome sequencing of this bacterium positioned it as a distinct species bearing genomic homology to non-pathogenic fast growers like M. smegmatis and pathogenic slow growers like M. tb [14], and was subsequently renamed as Mycobacterium indicus pranii (MIP). Although, MIP shares 80% virulence factors of M. tb (more than other non-pathogenic species), yet absence of crucial virulence associated genes encoded by mce2, mce3 operons renders it non-infectious. In-silico comparative proteomic analyses showed that RD1 region of H37Rv containing ESAT-6, CFP-10 are absent in MIP, pointing to their likely role in pathogenicity [15- 17]. MIP occupies an intermediate position in evolutionary hierarchy from free living mycobacteria to pathogenic forms and provides insights to mechanism that might have led to acquisition of pathogenicity in free living mycobacteria. Lateral

109 Seyed E. Hasnain et al. gene transfer has been shown to play a role in augmenting immunogenicity of majority of MIP proteins. Supporting this notion was the observation that out of 44 M. tb virulence factors that had homologs in MIP, 27 were laterally acquired. These included members of phospholipase C family (Rv2349c, Rv1755c, Rv2351c and Rv2350c) and glycosyl transferases (Rv2958c, Rv2962c) which are involved in fatty acid synthesis in M. tb and execute post translational modification of proteins, respectively inside host cells [18]. Thus, such virulence factors might fall into the category of useful acquisitions down the evolution to help gain access to the host niche. Mechanisms of acquisition of pathogenicity in mycobacteria Conservation of virulence factors across mycobacterium species suggests a driving force beyond gene acquisition and gene loss in mycobacterial evolution. Gene cooption has been shown as an evolutionary mechanism in rhodococcus, indicating that adaptive changes in certain proteins of non-pathogenic bacteria and horizontal acquisition of few critical genes during the process of evolution can give rise to virulence factors in pathogenic bacteria. Gene cooption leads to gain of new gene function by duplication or without duplication (Fig.1). Gene duplication involves either sub-functionalization (conservation of function partially) or neo- functionalization (gain of a new function) of the paralogous genes. Without gene duplication, genes can be co-opted for altered function by change in coding sequences and gain of novel domains [19]. In concordance with the concept of gene cooption in evolution, sequence variation between virulence factors of M. tb and their homologs in MIP, alter globularity and hydrophobicity. Few of the proteins like Ag85C (Rv0129c), MmpS4

Fig. 1: Role of expansion of PE gene family in Mycobacterium evolution. Gene cooption leads to gain of new gene function with or without duplication. PE proteins have evolutionarily expanded through duplication of esx genes that formed part of ancestral secretion system of M. tb.

110 Molecular Medicine : Bench to Bedside and Beyond

(Rv0451c), Ag85B (Rv1886c), mbth (Rv2377c), mycP4 (Rv3449) and mycP1 (Rv3883c) were found to be completely disordered proteins in M. tb (H37Rv), but their homologs in MIP were globular, whereas O-methyltransferase (Rv2956), WhiB3 (Rv3416) displayed the opposite change in globularity [20, 21]. Disordered regions of a protein not only participate in protein-protein or protein-DNA interactions, but also form aggregates that may aid in pathogenesis [22]. Aggregated proteins also show enhanced immune response as compared to monomers which explains higher immunogenicity of MIP proteins and its success as anti-leprosy vaccine. Another interesting observation is that although 18 virulence factors of M. tb (H37Rv) share greater than 90% sequence similarity with MIP, these factors do not confer virulence in MIP. Such contrasting virulent properties of homologs in H37Rv and MIP are attributed to an altered number of globular domains. For example, Rv1736c has been shown to be expressed by M.tb during dormancy and possess two globular domains, whereas its homolog in MIP has seven globular domains [20]. Additional globular domains might render altered functions to the protein. Seventy eight other proteins exhibited changes in number of globular domains between M. tb and MIP [21]. Another observation in support of gene cooption in mycobacterium evolution stems from changes in MCE family proteins between M. tb and MIP. MCE family is important for entry of M. tb into macrophages. Atleast six of the MCE family proteins in M. tb showed higher GRAVY values than their MIP homologs. The increased hydrophobic nature of MCE family proteins in M. tb (H37Rv) as compared to MIP might be a strategy employed by M. tb for easy recognition by phagocytic cells and gain access through the lipid bilayer [20, 21]. It is therefore evident that adaptive changes in niche factors of non-pathogenic mycobacterium propagated by gene cooption mechanisms might be responsible for origin of virulence factors in pathogenic Mycobacterium. PE/PPE as novel targets for drug and vaccine against Mycobacterium tuberculosis. Complete genome sequencing of M. tb revealed that nearly 10% of genome consist of unique family of 107 PE and 69 PPE proteins, found exclusively among this genus and has evolutionarily expanded through duplication of esx genes that formed part of ancestral secretion system of M. tb [23, 24]. The bicistronic PE/PPE genes are named due to the presence of Pro-Glu (PE) and Pro-Pro-Glu (PPE) signature motifs near the N-terminus [25, 26]. The PE family is classified into two subgroups: PE and PE_PGRS.PE_PGRS antigens of M. tb strain H37Rv share homology with PE protein of MIP (MIP_06644) and are highly stable and show extended C-terminal domain [6]. PGRS domain is absent in non-pathogenic and opportunistic mycobacteria. Recombinant proteins obtained from fusion of MIP¬_06644 domains of MIP with PGRS domain of M. tb Rv1818c proteins (PE¬_PGRS33) clearly show that PGRS domain of M. tb effects localization of PE domains within cells, stability and humoral response in humans [27]. PPE family can be sub grouped into PPE (no distinctive feature), PPE_SVP (C-terminal GxxSVPxxW motif), PPE_MPTR (major polymorphic tandem repeats) and PPE_PPW (C-terminal PxxPxxW motif) [26]. Most of these proteins are found to be localized to cell surface and/or secreted, play arole in regulating pathogenesis, in modulating the host immune response [8, 28-30], and in generating antigenic

111 Seyed E. Hasnain et al. variations, possibly due to the higher number of single-base substitution in these genes as compared to the rest of the genome [21, 23]. The polymorphism in this gene family is further involved in intra-species variability in H37Rv and H37Ra. Nucleotide and amino acids sequence of PE gene family vary between these H37Rv and H37Ra that led to changes in hydrophobicity, number of globular domains and potential serine threonine phosphorylation sites [20]. NetPhosBac analysis of this PE_PGRS protein Rv3508, involved in hypoxia and nitric oxide stress response of H37Rv, showed loss of phosphorylation sites. As phosphorylation plays a role in signaling process, loss of such sites may render protein insensitive to such stress responses leading to attenuation [31]. As shown in Fig. 2, PE/PPE protein family contributes to M. tb pathogenicity and modulates host immune response by a plethora of mechanisms and thus could be novel targets for anti-TB drugs. Remarkably, PE_PGRS62 inhibits maturation of phagosomes and PE_PGRS30 is associated with inhibiting phagosome lysosome fusion. PE_PGRS11, PE_PGRS17, PPE18, PPE38, PPE17 are shown to bind to TLRs and alter the downstream cascade and modulate the immune response [28, 29]. PE_PGRS33 has been shown to interact with host cell mitochondria and mediates host cell death [32].

Fig. 2: Role of PE/PPE proteins in M. tb virulence. Secreted PE/PPE proteins interfere with the host defence mechanisms whereas mycobacterial membrane bound PE/PPE members are responsible for generating drug resistance and antigenic variation in the pathogen.

There have been recent reports suggesting an alluring link between mycobacterial lipid metabolism and PE/PPE gene family. LipY (PE_PGRS63) and LipX (PE11) are shown to be up-regulated following starvation along with other M. tb lipase genes [33]. Further understanding of the host interacting partners of these families of proteins will illustrate key cues in host pathogen interactions critical for virulence and pathogenesis of M. tb. Futuristic generic tools for targeting against tuberculous and non-tuberculous bacteria Research on virulence factors acquired by mycobacteria to facilitate host invasion and pathogenicity holds epicenter in designing anti-TB drugs. Growing evidences of gene cooption and horizontal gene transfer suggest that non-tuberculous mycobacteria have propensity to acquire virulence due to

112 Molecular Medicine : Bench to Bedside and Beyond their coexistence with tuberculous mycobacteria in biofilms, as in case of cystic fibrosis where presence of different species of Mycobacterium and Pseudomonas have been reported [34]. The non-tuberculous bacteria, which have until now been characterized as key players in environmental biofilms and as opportunistic pathogens in immunocompromised individuals could gain etiological significance within biofilms of medical devices due to increasing use of invasive diagnostic and treatment procedures. On the other hand, hydrophobicity of M. tb cell surface which is an important factor for aerosolization, transmission and virulence could also favour bacterial adhesion on surfaces promoting biofilms. Biofilms can therefore, be conceptualized as interesting component for drug targeting because they are a common mechanism for drug tolerance in nontuberculous as well as tuberculous bacteria [35] and could prove to be the masterstroke required for tackling a wide spectrum of microorganisms. We have characterized important genes including PpiB (peptidyl prolyl isomerase), that play key role in biofilm formation or PE11, which is involved in cell wall remodeling and virulence [36, 37]. Our preliminary results show that targeting PpiB using FDA approved drugs suppresses biofilm formation and reduces MIC of first line anti-TB drugs. The hydrophobic cell wall of M. tb which seems to be impregnable to anti-TB drugs may be targeted with inhibitors of PE11 and could allow easy access to drugs thereby reducing dosing. Disturbing the growth of pathogenic microorganism through selective pressure of antibiotics can allow the normal microbiome to overpopulate, compete with pathogens for binding sites, food, space and neutralize toxins released by pathogens [38]. It is speculated that rapid evolution and emergence of newer strains or acquisition of pathogenicity among non-tuberculous species may soon outpace the efficacy of existing portfolio of TB drugs and has necessitated the need to either develop new drugs or repurpose currently available drugs. Repurposing preexisting FDA approved drugs circumvents steps in preclinical trials and allows clearance of safety/ethical concerns [39]. The bacterial genes that are involved in virulence, drug resistance, survival inside host and drug tolerance by biofilm formation have been identified through genome comparison [34]. Inhibitors against these proteins may be a key to antibacterial strategies as these proteins may perform their function alone or in combinations. Drugs used against some of the other known disorders in human being related to lysosomal disorders, diabetes, typhoid, malaria etc. target cellular pathways commonly utilized by M. tb for survival. For example, administration of hypertension drug Verapamil reduces risk towards TB by acting on efflux pumps [39]. Other candidate targets for drugs within M. tb genome have be screened across drug libraries. Several M. tb proteins such as Rv0079, Rv2626c, Rv2430c and Rv2608 modulate effector functions of macrophages, B cell and T cell response in host cell. These proteins might be used as target for inhibitor designing and screening that can help in enhancement of host cellular machinery to defend itself against M. tb [40, 41]. M. tb utilizes multifactorial pathways for survival and can transiently shut off metabolic pathways targeted by one drug and reroute pathways for survival. We envisage that such analyses would

113 Seyed E. Hasnain et al.

help identify strategic “target set” for polypharmacology and enable targeting multiple targets simultaneously for achieving higher therapeutic efficacy than with one target at a time. M. tb continues to evolve into a clever bug and holds an edge over survival pressures in terms of virulence, immune evasion and transmission. With the advent of high throughput detection techniques, genomic database and concerted efforts along multi dimensional networks, it is hoped that efforts can be expedited to overcome TB. If MIP turns out to be a successful vaccine against not just category II TB [13] but all forms of TB, this will be the biggest gift to humanity from Prof G. P. Talwar who continues to ignite passion for science among the emerging researchers in India and across the world. A TB free society, envisioned by WHO, will be the greatest leap of mankind in this century. Acknowledgements SEH is a JC Bose National Fellow, Department of Science and Technology, Government of India and Robert Koch Fellow, Robert Koch Institute, Berlin, Germany. This work was supported by a Centre of Excellence Grant to SEH and NZE from the Department of Biotechnology, Ministry of Science and Technology, Govt. of India. AA is a recipient of DBT Post Doctoral Fellowship. YS, SK and KS received Fellowships from IIT/ICMR. Declaration Authors declare no conflict of interests. References 1. WHO Global TB report, World Health Organization, Geneva. 2016. 2. Brauner A, Fridman O, Gefen O, Balaban NQ. Distinguishing between resistance, tolerance and persistence to antibiotic treatment. Nat Rev Microbiol 14: 320-330 (2016). 3. Siddiqi N, Shamim M, Jain NK, Rattan A, Amin A, Katoch VM, Sharma SK, Hasnain SE. Molecular genetic analysis of multi-drug resistance in Indian isolates of Mycobacterium tuberculosis. Mem Inst Oswaldo Cruz 93: 589-594 (1998). 4. Ahmed N, Ehtesham NZ, Hasnain SE. Ancestral Mycobacterium tuberculosis genotypes in India: Implications for TB control programmes. Infect Genet Evol 9: 142-146 (2009). 5. Hasnain SE, O'Toole RF, Grover S, Ehtesham NZ. Whole genome sequencing: a new paradigm in the surveillance and control of human tuberculosis. Tuberculosis (Edinb) 95: 91-94 (2015). 6. Singh Y, Kohli S, Sowpati DT, Rahman SA, Tyagi AK, Hasnain SE. Gene cooption in mycobacteria and search for virulence attributes: comparative proteomic analyses of Mycobacterium tuberculosis, Mycobacterium indicus pranii and other mycobacteria. Int J Med Microbiol 304: 742-748 (2014). 7. Cobat A, Orlova M, Barrera LF, Schurr E. Host genomics and control of tuberculosis infection. Public Health Genomics 16: 44-49 (2013). 8. Akhter Y, Ehebauer MT, Mukhopadhyay S, Hasnain SE. The PE/PPE multigene family codes for virulence factors and is a possible source of mycobacterial antigenic variation: perhaps more? Biochimie 94: 110-116 (2012). 9. Ahmed N, Dobrindt U, Hacker J, Hasnain SE. Genomic fluidity and pathogenic bacteria: applications in diagnostics, epidemiology and intervention. Nat Rev Microbiol 6: 387-394 (2008). 10. Prasanna AN, Mehra S. Comparative phylogenomics of pathogenic and non-pathogenic mycobacterium. PLoS One 8: e71248 (2013). 11. Veyrier F, Pletzer D, Turenne C, Behr MA. Phylogenetic detection of horizontal gene transfer during the step-wise genesis of Mycobacterium tuberculosis. BMC Evol Biol 9: 196 (2009).

114 Molecular Medicine : Bench to Bedside and Beyond

12. Ganju L, Mukherjee R, Batra HV, Talwar GP. Immuno-blot analysis of antigens of Mycobacterium w: a candidate anti-leprosy vaccine using monoclonal antibodies and patient sera. Zentralbl Bakteriol 273: 378-385 (1990). 13. Sharma SK, Katoch K, Sarin R, Balambal R, Jain NK, Patel N, Murthy KJR, Singla N, Saha PK, Khanna A, Singh U, Kumar S, Sengupta A, Banavaliker JN, Chauhan DS, Sachan S, Wasim M, Tripathi S, Dutt N, Jain N, Joshi N, Penmesta SRR, Gaddam S, Gupta S, Khamar B, Dey B, Mitra DK, Arora SK, Bhaskar S, Rani R. Efficacy and Safety of Mycobacterium indicus pranii as an adjunct therapy in Category II pulmonary tuberculosis in a randomized trial. Sci Reports: accepted (2017). 14. Saini V, Raghuvanshi S, Talwar GP, Ahmed N, Khurana JP, Hasnain SE, Tyagi AK. Polyphasic taxonomic analysis establishes Mycobacterium indicus pranii as a distinct species. PLoS One 4: e6263 (2009). 15. Aguilar LD, E Infante MV, Bianco A, Cataldi FB, Pando RH. Immunogenicity and protection induced by Mycobacterium tuberculosis mce-2 and mce-3 mutants in a Balb/c mouse model of progressive pulmonary tuberculosis. Vaccine 24: 2333-2342 (2006). 16. Ahmed N, Saini V, Raghuvanshi S, Khurana JP, Tyagi AK, Tyagi AK, Hasnain SE. Molecular analysis of a leprosy immunotherapeutic bacillus provides insights into Mycobacterium evolution. PLoS One 2: e968 (2007). 17. Saini V, Raghuvanshi S, Khurana JP, Ahmed N, Hasnain SE, Tyagi AK. Massive gene acquisitions in Mycobacterium indicus pranii provide a perspective on mycobacterial evolution. Nucleic Acids Res 40: 10832-10850 (2012). 18. Srinivas M, Rajakumari S, Narayana Y, Joshi B, Katoch VM, Rajasekharan R, Balaji KN. Functional characterization of the phospholipase C activity of Rv3487c and its localization on the cell wall of Mycobacterium tuberculosis. J Biosci 33: 221-230 (2008). 19. True JR, Carroll SB. Gene co-option in physiological and morphological evolution. Annu Rev Cell Dev Biol 18: 53-80 (2002). 20. Kohli S, Singh Y, Sharma K, Mittal A, Ehtesham NZ, Hasnain SE. Comparative genomic and proteomic analyses of PE/PPE multigene family of Mycobacterium tuberculosis H37Rv and H37Ra reveal novel and interesting differences with implications in virulence. Nucleic Acids Res 40: 7113- 7122 (2012). 21. Rahman SA, Singh Y, Kohli S, Ahmad J, Ehtesham NZ, Tyagi AK, Hasnain SE. Comparative analyses of nonpathogenic, opportunistic, and totally pathogenic mycobacteria reveal genomic and biochemical variabilities and highlight the survival attributes of Mycobacterium tuberculosis. MBio 5: e02020 (2014). 22. Uversky VN, Oldfield CJ, Dunker AK. Intrinsically disordered proteins in human diseases: introducing the D2 concept. Annu Rev Biophys 37: 215-246 (2008). 23. Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C, Harris D, Gordon SV, Eiglmeier K, Gas S, Barry CE 3rd, Tekaia F, Badcock K, Basham D, Brown D, Chillingworth T, Connor R, Davies R, Devlin K, Feltwell T, Gentles S, Hamlin N, Holroyd S, Hornsby T, Jagels K, Krogh A, McLean J, Moule S, Murphy L, Oliver K, Osborne J, Quail MA, Rajandream MA, Rogers J, Rutter S, Seeger K, Skelton J, Squares R, Squares S, Sulston JE, Taylor K, Whitehead S, Barrell BG. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393: 537-544 (1998). 24. Gey van Pittius NC, Sampson SL, Lee H, Kim Y, van Helden PD, Warren RM. Evolution and expansion of the Mycobacterium tuberculosis PE and PPE multigene families and their association with the duplication of the ESAT-6 (esx) gene cluster regions. BMC Evol Biol 2006 6: 95 (2006). 25. Tundup S, Akhter Y, Thiagarajan D, Hasnain SE. Clusters of PE and PPE genes of Mycobacterium tuberculosis are organized in operons: evidence that PE Rv2431c is co-transcribed with PPE Rv2430c and their gene products interact with each other. FEBS Lett 580: 1285-93 (2006). 26. Mohareer K, Tundup S, Hasnain SE. Transcriptional regulation of Mycobacterium tuberculosis PE/PPE genes: a molecular switch to virulence? J Mol Microbiol Biotechnol 21: 97-109 (2011). 27. Cascioferro A, Daleke MH, Ventura M, Donà V, Delogu G, Palù G, Bitter W, Manganelli R. Functional dissection of the PE domain responsible for translocation of PE_PGRS33 across the mycobacterial cell wall. PLoS One 6: e27713 (2011). 28. Nair S, Ramaswamy PA, Ghosh S, Joshi DC, Pathak N, Siddiqui I, Sharma P, Hasnain SE, Mande SC,

115 Seyed E. Hasnain et al.

Mukhopadhyay S. The PPE18 of Mycobacterium tuberculosis interacts with TLR2 and activates IL-10 induction in macrophage. J Immunol 183: 6269-6281 (2009). 29. Nair S, Pandey AD, Mukhopadhyay S. The PPE18 protein of Mycobacterium tuberculosis inhibits NF- κB/rel-mediated proinflammatory cytokine production by upregulating and phosphorylating suppressor of cytokine signaling protein. J Immunol 186: 5413-24 (2011). 30. Sampson SL. Mycobacterial PE/PPE proteins at the host-pathogen interface. Clin Dev Immunol 2011:497203 (2011). 31. Yu G, Fu X, Jin K, Zhang L, Wu W, Cui Z, Hu Z, Li Y. Integrative analysis of transcriptome and genome indicates two potential genomic islands are associated with pathogenesis of Mycobacterium tuberculosis. Gene 489: 21-29 (2011). 32. Bhat KH, Das A, Srikantam A, Mukhopadhyay S. PPE2 protein of Mycobacterium tuberculosis may inhibit nitric oxide in activated macrophages. Ann N Y Acad Sci 1283: 97-101 (2013). 33. Deb C, Daniel J, Sirakova TD, Abomoelak B, Dubey VS, Kolattukudy PE. A novel lipase belonging to the hormone-sensitive lipase family induced under starvation to utilize stored triacylglycerol in Mycobacterium tuberculosis. J Biol Chem 281: 3866-3875 (2006). 34. Qvist T, Pressler T, Høiby N, Katzenstein TL. Shifting paradigms of nontuberculous mycobacteria in cystic fibrosis. Respir Res 15: 41 (2014). 35. Olive AJ, Sassetti CM. Metabolic crosstalk between host and pathogen: sensing,adapting and competing. Nat Rev Microbiol 14: 221-34 (2016). 36. Pandey S, Sharma A, Tripathi D, Kumar A, Khubaib M, Bhuwan M, Chaudhuri TK, Hasnain SE, Ehtesham NZ. Mycobacterium tuberculosis Peptidyl-Prolyl Isomerases also exhibit chaperone like activity In-Vitro and In-Vivo. PLoS One 11: e0150288 (2016). 37. Singh P, Rao RN, Reddy JR, Prasad RB, Kotturu SK, Ghosh S, Mukhopadhyay S. PE11, a PE/PPE family protein of Mycobacterium tuberculosis in involved in cell wall remodeling and virulence. Sci Rep 6: 21624 (2016). 38. Singh Y, Ahmad J, Musarrat J, Ehtesham NZ, Hasnain SE. Emerging importance of holobionts in evolution and in probiotics. Gut Pathog 5: 12 (2013). 39. Szumowski JD, Adams KN, Edelstein PH, Ramakrishnan L. Antimicrobial efflux pumps and Mycobacterium tuberculosis drug tolerance: evolutionary considerations. Curr Top Microbiol Immunol 374: 81-108 (2013). 40. Choudhary RK, Mukhopadhyay S, Chakhaiyar P, Sharma N, Murthy KJ, Katoch VM, Hasnain SE. PPE antigen Rv2430c of Mycobacterium tuberculosis induces a strong B-cell response. Infect Immun 71: 6338-6343 (2003). 41. Chakhaiyar P, Nagalakshmi Y, Aruna B, Murthy KJ, Katoch VM, Hasnain SE. Regions of high antigenicity within the hypothetical PPE major polymorphic tandem repeat open-reading frame, Rv2608, show a differential humoral response and a low T cell response in various categories of patients with tuberculosis. J Infect Dis 190: 1237-1244 (2004).

116

201 7

General Commentary published: 28 April 2017 doi: 10.3389/fimmu.2017.00466

Commentary: modification of Host responses by mycobacteria

Ashutosh Kumar 1, Mamta Rani 2, Nasreen Z. Ehtesham 3 and Seyed E. Hasnain 1,4,5*

1 Molecular Infection and Functional Biology Laboratory, Kusuma School of Biological Sciences, Indian Institute of Technology-Delhi, New Delhi, India, 2 School of Life Sciences, Jawaharlal Nehru University, New Delhi, India, 3 Inflammation Biology and Cell Signaling Laboratory, National Institute of Pathology, New Delhi, India, 4 Jamia Hamdard, Institute of Molecular Medicine, New Delhi, India, 5 Dr Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Hyderabad, India

Keywords: Mycobacterium tuberculosis, stress adaptation, virulence factors, host pathways, persisters

A commentary on

Mycobacterial dormancy systems and host responses in tuberculosis by Peddireddy V, Doddam S, Ahmed N. Front Immunol (2017) 8:84. doi: 10.3389/fimmu.2017.00084

Mycobacterium tuberculosis (M.tb), an obligate slow-growing human pathogen, resides within the macrophage after phagocytosis and develops strategies to escape immune surveillance. It can cause active disease or can persist in a latent stage depending on the host immune responses. The mycobacterial cell wall consists of complex layers of arabinogalactan, peptidoglycan, and unusually long branched mycolic acids that are covalently linked with each other. The cell wall of mycobacteria, containing high proportion of lipids, has 15 times less density of pores in comparison to the outer membrane of Gram-negative bacteria (1). This low density of pores might cause more difficulty in absorption of nutrients and could contribute to slow growth of mycobacteria. The other reasons for slow growth are higher GC content of the promoters, differential orientation of the genes in relation

Edited by: to the direction of replication, low RNA/DNA ratio in growing mycobacteria, and presence of a Hridayesh Prakash, single ribosomal RNA operon present apart from the oriC. Proteins involved in the formation of the University of Hyderabad, India substrate-specific energy-dependent transporters ABC transport systems (ATP-binding cassette) are

Reviewed by: coded by only 2.5% of the M.tb genome that is very less compared to 5% in case of the Escherichia Elsa Anes, coli genome (2). Universidade de Lisboa, Portugal During infection, M.tb targets several host pathways such as induction of glycolytic flux (3),

*Correspondence: endoplasmic reticulum stress (4, 5), disruption of mitochondrial membrane (6), inhibition of apop- Seyed E. Hasnain tosis (7), induction of necrosis (3), phagosome maturation, suppressing host signaling pathways (8), [email protected], and regulate autophagy to survive within host cell (9). Inside the granuloma, both the mycobacteria [email protected] and the macrophages survive under stress conditions because of limitation of nutrients. To persist under such unfavorable conditions, both bacteria and macrophages have to conserve their energy by Specialty section: decreasing metabolic rate to allocate available resources toward the production of dedicated stress This article was submitted management proteins. Stress granules formation is a major adaptive defense mechanism through to Inflammation, translation repression for stress survival of host cell infected with mycobacteria (4, 5). a section of the journal Intracellular mycobacteria are found in different vacuolar compartments in distinct physiol­ Frontiers in Immunology ogical state, gene expression, and survival (10, 11). It has been shown that mycobacterial infection Received: 31 January 2017 activated phagocytes to secrete different cytokines after triggering several host receptors such as type Accepted: 05 April 2017 C lectins such as DC-SIGN (12, 13), NOD/NACHT receptors (14), mannose receptors (15), and Published: 28 April 2017 toll-like receptor 2 (16). Mincle receptor [macrophage inducible Ca2+-dependent (C-type) lectin] is Citation: a calcium-dependent lectin that is a receptor for mycobacterial cord factor, trehalose-6,6’-dimycolate Kumar A, Rani M, Ehtesham NZ and (TDM). Mincle expression on neutrophils is required for TDM infiltration that binds to both the Hasnain SE (2017) Commentary: Modification of Host Responses sugar portion of the glycolipid and the hydrocarbon tail (17). by Mycobacteria. We have reported several key proteins of M.tb that may be functionally important for pathogen- Front. Immunol. 8:466. esis and survival. Prominent among these are M.tb PE/PPE proteins that have multiple role in terms doi: 10.3389/fimmu.2017.00466 of providing antigenic variation to the pathogen, acting as a molecular switch toward virulence

Frontiers in Immunology | www.frontiersin.org 1 April 2017 | Volume 8 | Article 466 Kumar et al. Mycobacterial Adaptation and Host Responses and altering Th1/Th2 host immune response for survival18 ( –20), level of replication (31), transcription (32), and translation immune quorum sensing (21), etc. Interaction of M.tb virulence (33, 34) (Figure 1). A recent report (35) has described different factor RipA with chaperone MoxR1 was required for transport mycobacterial strategies against host immune responses such through TAT secretion system (22). Inhibition of M.tb chaper- as manipulation of the TLR responses, host cytokine responses, onic proteins such as PpiA and PpiB can derail protein folding antigen presentation by MHCs, inhibition of phagolysosomal machinery in M.tb (23) and reticence intracellular bacterial fusion, and resistance to reactive nitrogen intermediates. The survival through alteration of host cytokine profile 24( ). PpiB role of toxin antitoxins systems in mycobacterial growth regu- also regulates formation of biofilm and can contribute to drug lation in unfavorable conditions and role of Clp proteases in tolerance. Several M.tb proteins, such as DATIN, modulate host reactivation of latent bacilli have been described in detail. It has cytokine profile by interacting with TLR-2 (25), Rv2626c induce been shown that arrest of protein synthesis induces formation of the production of pro-inflammatory cytokines through NF-κB persisters (36) that may have similar metabolic and physiologi- (26). Rv2430c induces strong B-cell response (27), while Rv2608 cal state as the dormant bacteria (37). The persisters are drug induces different humoral and T-cell response in various catego- tolerant non-grower bacteria, genetically similar sibling of drug ries of TB patients (28). Inhibitors of these proteins can help boost susceptible bacteria but physiologically resistant (persistent) host immune system within host and provide an unfavorable against various bacterial drugs (38). Comparative genomic environment for M.tb to survive. M.tb ORF Rv1475c encoded analyses revealed genes associated with survival, virulence, aconitase is an iron binding protein that has conserved residues of antibiotic resistance, and biofilm formation (39). Many of these the iron-responsive class of proteins and binds to iron-responsive genes can act alone or in combination with other genes and thus elements in case of iron depletion (29). It is one of the several M.tb inhibitors against such genes can prove vital in targeting the proteins identified in 30-day infected guinea pig lungs indicating virulence and survival of M.tb. Drug re-purposing is an emerg- its role in host–pathogen interaction (30). ing strategy where drugs already in clinical use or approved by There are several proteins present in mycobacteria which US FDA for treatment of mental illness, diabetes, malaria, etc. help in its survival inside host by slowing down growth at the are being tested against some of the pathogen targets described

Figure 1 | Growth regulation by Mycobacterium for adaptation to stress/dormancy. Mycobacterium tuberculosis (M.tb) IciA (inhibitor of chromosome initiation) binds to the A + T rich oriC region of the M.tb genome and inhibits helix opening resulting in the arrest of chromosomal DNA replication (31). Activated toxin–antitoxin (TA) systems cleave mRNA to shut down metabolic activity (32). Peddireddy et al. (35) have also described the role of TA systems in M.tb and Mycobacterium smegmatis to remain in non-replicating phase that help bacteria in antibiotic tolerance. Highly expressed protein DATIN/RafH of Mycobacterium inhibits translation by binding with the ribosome under conditions of stress (33, 34). Confirmed and putative roles are indicated with continuous and dashed arrows, respectively.

Frontiers in Immunology | www.frontiersin.org 2 April 2017 | Volume 8 | Article 466 Kumar et al. Mycobacterial Adaptation and Host Responses above. Targeting those host cellular pathways that are also FUNDING commonly utilized by M.tb for its survival is yet another mode of developing new drugs. SH and NE thank the DBT, Ministry of S&T, Government of India, for a Centre of Excellence Grant (BT/PR12817/COE/34/23/2015). AUTHOR CONTRIBUTIONS SH is a JC Bose National Fellow of the Department of Science and Technology (S&T), Ministry of S&T, Government of India, SH and AK conceived the idea behind this commentary; AK and and a Robert Koch Fellow of the Robert Koch Institute, Berlin, MR wrote the draft; and NE and SH finalized the manuscript. Germany.

REFERENCES 16. Udgata A, Qureshi R, Mukhopadhyay S. Transduction of functionally contrasting signals by two mycobacterial PPE proteins downstream of TLR2 1. Niederweis M. Mycobacterial porins – new channel proteins in unique outer receptors. J Immunol (2016) 197(5):1776–87. doi:10.4049/jimmunol.1501816 membranes. Mol Microbiol (2003) 49:1167–77. doi:10.1046/j.1365-2958. 17. Feinberg H, Jégouzo SA, Rowntree TJ, Guan Y, Brash MA, Taylor ME, 2003.03662.x et al. Mechanism for recognition of an unusual mycobacterial glycolipid 2. Braibant M, Gilot P, Content J. The ATP binding cassette (ABC) transport sys- by the macrophage receptor mincle. J Biol Chem (2013) 288(40):28457–65. tems of Mycobacterium tuberculosis. FEMS Microbiol Rev (2000) 24:449–67. doi:10.1074/jbc.M113.497149 doi:10.1111/j.1574-6976.2000.tb00550.x 18. Akhter Y, Ehebauer MT, Mukhopadhyay S, Hasnain SE. The PE/PPE mul- 3. Mehrotra P, Jamwal SV, Saquib N, Sinha N, Siddiqui Z, Manivel V, et al. tigene family codes for virulence factors and is a possible source of myco- Pathogenicity of Mycobacterium tuberculosis is expressed by regulating meta- bacterial antigenic variation: perhaps more? Biochimie (2012) 94(1):110–6. bolic thresholds of the host macrophage. PLoS Pathog (2014) 10(7):e1004265. doi:10.1016/j.biochi.2011.09.026 doi:10.1371/journal.ppat.1004265 19. Mohareer K, Tundup S, Hasnain SE. Transcriptional regulation of 4. Seimon TA, Kim MJ, Blumenthal A, Koo J, Ehrt S, Wainwright H, et al. Mycobacterium tuberculosis PE/PPE genes: a molecular switch to virulence? Induction of ER stress in macrophages of tuberculosis granulomas. PLoS One J Mol Microbiol Biotechnol (2011) 21(3–4):97–109. doi:10.1159/000329489 (2010) 5(9):e12772. doi:10.1371/journal.pone.0012772 20. Khubaib M, Sheikh JA, Pandey S, Srikanth B, Bhuwan M, Khan N, et al. 5. Lim YJ, Choi JA, Choi HH, Cho SN, Kim HJ, Jo EK, et al. Endoplasmic Mycobacterium tuberculosis co-operonic PE32/PPE65 proteins alter host reticulum stress pathway-mediated apoptosis in macrophages contributes to immune responses by hampering Th1 response. Front Microbiol (2016) 7:719. the survival of Mycobacterium tuberculosis. PLoS One (2011) 6(12):e28531. doi:10.3389/fmicb.2016.00719 doi:10.1371/journal.pone.0028531 21. Tundup S, Mohareer K, Hasnain SE. Mycobacterium tuberculosis PE25/ 6. Chen M, Gan H, Remold HG. A mechanism of virulence: virulent PPE41 protein complex induces necrosis in macrophages: role in virulence Mycobacterium tuberculosis strain H37Rv, but not attenuated H37Ra, causes and disease reactivation? FEBS Open Bio (2014) 4:822–8. doi:10.1016/j. significant mitochondrial inner membrane disruption in macrophages fob.2014.09.001 leading to necrosis. J Immunol (2006) 176:3707–16. doi:10.4049/jimmunol. 22. Bhuwan M, Arora N, Sharma A, Khubaib M, Pandey S, Chaudhuri TK, et al. 176.6.3707 Interaction of Mycobacterium tuberculosis virulence factor ripa with chaper- 7. Keane J, Remold HG, Kornfeld H. Virulent Mycobacterium tuberculosis one MoxR1 is required for transport through the TAT secretion system. MBio strains evade apoptosis of infected alveolar macrophages. J Immunol (2000) (2016) 7(2):e02259. doi:10.1128/mBio.02259-15 164:2016–20. doi:10.4049/jimmunol.164.4.2016 23. Pandey S, Sharma A, Tripathi D, Kumar A, Khubaib M, Bhuwan M, et al. 8. Koul A, Herget T, Klebl B, Ullrich A. Interplay between mycobacteria and Mycobacterium tuberculosis peptidyl-prolyl isomerases also exhibit chap- host signalling pathways. Nat Rev Microbiol (2004) 2:189–202. doi:10.1038/ erone like activity in-vitro and in-vivo. PLoS One (2016) 11(3):e0150288. nrmicro840 doi:10.1371/journal.pone.0150288 9. Kumar D, Nath L, Kamal MA, Varshney A, Jain A, Singh S, et al. Genome- 24. Pandey S, Tripathi D, Khubaib M, Kumar A, Sheikh JA, Sumanlatha G, et al. wide analysis of the host intracellular network that regulates survival Mycobacterium tuberculosis peptidyl-prolyl isomerases are immunogenic, of Mycobacterium tuberculosis. Cell (2010) 140:731–43. doi:10.1016/j. alter cytokine profile and aid in intracellular survival. Front Cell Infect cell.2010.02.012 Microbiol (2017) 7:38. doi:10.3389/fcimb.2017.00038 10. Rohde KH, Veiga DF, Caldwell S, Balazsi G, Russell DG. Linking the tran- 25. Kumar A, Lewin A, Rani PS, Qureshi IA, Devi S, Majid M, et al. Dormancy scriptional profiles and the physiological states of Mycobacterium tuberculosis associated translation inhibitor (DATIN/Rv0079) of Mycobacterium tubercu- during an extended intracellular infection. PLoS Pathog (2012) 8:e1002769. losis interacts with TLR2 and induces proinflammatory cytokine expression. doi:10.1371/journal.ppat.1002769 Cytokine (2013) 64:258–64. doi:10.1016/j.cyto.2013.06.310 11. Bhaskar A, Chawla M, Mehta M, Parikh P, Chandra P, Bhave D, et al. 26. Bashir N, Kounsar F, Mukhopadhyay S, Hasnain SE. Mycobacterium Reengineering redox sensitive GFP to measure mycothiol redox potential tuberculosis conserved hypothetical protein rRv2626c modulates macro- of Mycobacterium tuberculosis during infection. PLoS Pathog (2014) phage effector functions. Immunology (2010) 130(1):34–45. doi:10.1111/j. 10(1):e1003902. doi:10.1371/journal.ppat.1003902 1365-2567.2009.03196.x 12. Geijtenbeek TB, Van Vliet SJ, Koppel EA, Sanchez-Hernandez M, 27. Choudhary RK, Mukhopadhyay S, Chakhaiyar P, Sharma N, Murthy KJ, Vandenbroucke-Grauls CM, Appelmelk B, et al. Mycobacteria target Katoch VM, et al. PPE antigen Rv2430c of Mycobacterium tuberculosis induces DC-SIGN to suppress dendritic cell function. J Exp Med (2003) 197:7–17. a strong B-cell response. Infect Immun (2003) 71(11):6338–43. doi:10.1128/ doi:10.1084/jem.20021229 IAI.71.11.6338-6343.2003 13. Tailleux L, Schwartz O, Herrmann JL, Pivert E, Jackson M, Amara A, et al. 28. Chakhaiyar P, Nagalakshmi Y, Aruna B, Murthy KJ, Katoch VM, Hasnain SE. DC-SIGN is the major Mycobacterium tuberculosis receptor on human den- Regions of high antigenicity within the hypothetical PPE major polymorphic dritic cells. J Exp Med (2003) 197:121–7. doi:10.1084/jem.20021468 tandem repeat open-reading frame, Rv2608, show a differential humoral 14. Ferwerda G, Girardin SE, Kullberg B-J, Le Bourhis L, de Jong DJ, response and a low T cell response in various categories of patients with Langenberg DML, et al. NOD2 and toll-like receptors are nonredundant rec- tuberculosis. J Infect Dis (2004) 190(7):1237–44. doi:10.1086/423938 ognition systems of Mycobacterium tuberculosis. PLoS Pathog (2005) 1(3):e34. 29. Banerjee S, Nandyala AK, Raviprasad P, Ahmed N, Hasnain SE. Iron- doi:10.1371/journal.ppat.0010034 dependent RNA-binding activity of Mycobacterium tuberculosis aconitase. 15. Kang PB, Azad AK, Torrelles JB, Kaufman TM, Beharka A, Tibesar E, et al. J Bacteriol (2007) 189:4046–52. doi:10.1128/JB.00026-07 The human macrophage mannose receptor directs Mycobacterium tubercu- 30. Kruh NA, Troudt J, Izzo A, Prenni J, Dobos KM. Portrait of a pathogen: the losis lipoarabinomannan-mediated phagosome biogenesis. J Exp Med (2005) Mycobacterium tuberculosis proteome in vivo. PLoS One (2010) 5(11):e13938. 202:987–99. doi:10.1084/jem.20051239 doi:10.1371/journal.pone.0013938

Frontiers in Immunology | www.frontiersin.org 3 April 2017 | Volume 8 | Article 466 Kumar et al. Mycobacterial Adaptation and Host Responses

31. Kumar S, Farhana A, Hasnain SE. In-vitro helix opening of M. tuberculosis 37. Wood TK, Knabel SJ, Kwan BW. Bacterial persister cell formation and oriC by DnaA occurs at precise location and is inhibited by IciA like protein. dormancy. Appl Environ Microbiol (2013) 79(23):7116. doi:10.1128/ PLoS One (2009) 4:e4139. doi:10.1371/journal.pone.0004139 AEM.02636-13 32. Ramage HR, Connolly LE, Cox JS. Comprehensive functional analysis 38. Balaban NQ, Gerdes K, Lewis K, McKinney JD. A problem of persistence: of Mycobacterium tuberculosis toxin-antitoxin systems: implications for still more questions than answers? Nat Rev Microbiol (2013) 11:587–91. pathogenesis, stress responses, and evolution. PLoS Genet (2009) 5:e1000767. doi:10.1038/nrmicro3076 doi:10.1371/journal.pgen.1000767 39. Farhat MR, Shapiro BJ, Kieser KJ, Sultana R, Jacobson KR, Victor TC, 33. Kumar A, Majid M, Kunisch R, Rani PS, Qureshi IA, Lewin A, et al. et al. Genomic analysis identifies targets of convergent positive selection in Mycobacterium tuberculosis DosR regulon gene Rv0079 encodes a putative, drug-resistant Mycobacterium tuberculosis. Nat Genet (2013) 45(10):1183–9. ‘dormancy associated translation inhibitor (DATIN)’. PLoS One (2012) doi:10.1038/ng.2747 7:e38709. doi:10.1371/journal.pone.0038709 34. Trauner A, Lougheed KE, Bennett MH, Hingley-Wilson SM, Williams HD. Conflict of Interest Statement: The authors declare that the research was con- The dormancy regulator DosR controls ribosome stability in hypoxic ducted in the absence of any commercial or financial relationships that could be mycobacteria. J Biol Chem (2012) 287:24053–63. doi:10.1074/jbc.M112. construed as a potential conflict of interest. 364851 35. Peddireddy V, Doddam S, Ahmed N. Mycobacterial dormancy systems and Copyright © 2017 Kumar, Rani, Ehtesham and Hasnain. This is an open-access article host responses in tuberculosis. Front Immunol (2017) 8:84. doi:10.3389/ distributed under the terms of the Creative Commons Attribution License (CC BY). fimmu.2017.00084 The use, distribution or reproduction in other forums is permitted, provided the 36. Kwan BW, Valenta JA, Benedik MJ, Wood TK. Arrested protein synthesis original author(s) or licensor are credited and that the original publication in this increases persister-like cell formation. Antimicrob Agents Chemother (2013) journal is cited, in accordance with accepted academic practice. No use, distribution 57:1468–73. doi:10.1128/AAC.02135-12 or reproduction is permitted which does not comply with these terms.

Frontiers in Immunology | www.frontiersin.org 4 April 2017 | Volume 8 | Article 466 ORIGINAL RESEARCH published: 15 February 2017 doi: 10.3389/fcimb.2017.00038

Mycobacterium tuberculosis Peptidyl-Prolyl Isomerases Are Immunogenic, Alter Cytokine Profile and Aid in Intracellular Survival

Saurabh Pandey 1, 2 †, Deeksha Tripathi 3 †, Mohd Khubaib 1, 2 †, Ashutosh Kumar 3, Javaid A. Sheikh 1, Gaddam Sumanlatha 4, Nasreen Z. Ehtesham 1*† and Seyed E. Hasnain 3, 4, 5*†

1 Inflammation Biology and Cell Signaling Laboratory, National Institute of Pathology, New Delhi, India, 2 Department of Biology, Dr. Reddy’s Institute of Life Sciences, University of Hyderabad, Hyderabad, India, 3 Molecular Infection and Functional Biology Laboratory, Kusuma School of Biological Sciences, Indian Institute of Technology, New Delhi, India, 4 Bhagwan Mahavir Medical Research Centre, Hyderabad, India, 5 Jamia Hamdard, Institute of Molecular Medicine, New Delhi, India

Mycobacterium tuberculosis (M. tb) has two peptidyl-prolyl isomerases (Ppiases) PpiA and PpiB, popularly known as cyclophilin A and cyclophilin B. The role of cyclophilins in processes such as signaling, cell surface recognition, chaperoning, and heat shock response has been well-documented. We present evidence that M. tb Ppiases modulate

Edited by: the host immune response. ELISA results revealed significant presence of antibodies Souhaila Al Khodor, to M. tb Ppiases in patient sera as compared to sera from healthy individuals. Sidra Medical and Research Center, Treatment of THP-1 cells with increasing concentrations of rPpiA, induced secretion of Qatar pro-inflammatory cytokines TNF-α and IL-6. Alternatively, treatment with rPpiB inhibited Reviewed by: Jordi Torrelles, secretion of TNF-α and induced secretion of IL-10. Furthermore, heterologous expression Ohio State University, USA of M. tb PpiA and PpiB in Mycobacterium smegmatis increased bacterial survival in Marina Santic, University of Rijeka, Croatia THP-1 cells as compared to those transformed with the vector control. Our results *Correspondence: demonstrate that M. tb Ppiases are immunogenic proteins that can possibly modulate Nasreen Z. Ehtesham host immune response and enhance persistence of the pathogen within the host by [email protected] subverting host cell generated stresses. Seyed E. Hasnain [email protected] Keywords: bacterial survival, cyclophilins, M. tuberculosis, immune response, THP-1 cells †These authors have contributed equally to this work. INTRODUCTION Received: 04 November 2016 Accepted: 31 January 2017 Tuberculosis caused by the intracellular pathogen Mycobacterium tuberculosis (M. tb), Published: 15 February 2017 remains a potential threat regardless of strong efforts to alleviate its toll on humanity. Citation: Intracellular/intraphagosomal survival of this pathogen plays a critical role in the infection cycle Pandey S, Tripathi D, Khubaib M, of the pathogen, a process which majorly relies on array of virulence factors to colonize and Kumar A, Sheikh JA, Sumanlatha G, replicate within the host macrophages (Ehrt and Schnappinger, 2009). Virulence factors which Ehtesham NZ and Hasnain SE (2017) have been known to play a vital role in host pathogen interaction at the molecular level include Mycobacterium tuberculosis stress responders, heat shock proteins (HSPs), foldases, and chaperones (Henderson, 2010). Studies Peptidyl-Prolyl Isomerases Are Immunogenic, Alter Cytokine Profile involving immunological characterization of these effector molecules can bridge huge gaps in our and Aid in Intracellular Survival. understanding of M. tb biology and facilitate better therapeutic and diagnostic interventions. Front. Cell. Infect. Microbiol. 7:38. Protein folding in the cell is assisted by molecular chaperones and foldases. The foldases doi: 10.3389/fcimb.2017.00038 generally include peptidyl-prolyl isomerases and protein disulfide isomerases. In addition to

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 1 February 2017 | Volume 7 | Article 38 Pandey et al. M.tb Ppiases Aid Intracellular Survival peptidyl prolyl isomerase activity, Ppiases have been shown Enzyme Assay of Purified Recombinant to play role in diverse biological processes such as receptor Ppiases signaling, apoptosis, stress response, and RNA-mediated gene The genes encoding M. tb PpiA (Rv0009) and M. tb PpiB (Rv2582) expression (Sykesetal.,1993;Luetal.,1996;Wuetal.,2000;Mark were PCR amplified from genomic DNA of M. tb strain H37Rv, et al., 2001). Human cyclophilin A and also M. tb Ppiases have using forward and reverse primers and cloned in pET28a and been reported to possess chaperone activity (Zhang et al., 2013; pGEX6p-1 vector as described (Pandey et al., 2016). Recombinant Pandey et al., 2016). Many such chaperones and HSPs have also proteins were purified (Banerjee et al., 2007) using Ni-NTA been known to have immune modulatory role during bacterial column for PpiA and glutathione sepharose affinity column for infections. For example, mycobacterial HSP65 induces a strong PpiB. Ppiase activity of both, rPpiA and rPpiB was evaluated cellular and humoral immune response (Peetermans et al., 1994; using a spectrophotometric assay (Pandey et al., 2016). Friedland et al., 2008). M. tb is known to possess two Ppiases (cyclophilins), PpiA Antigenicity Profiling and PpiB. M. tb PpiA is a part of the secretome and is known Antigenic index of PpiA and PpiB was analyzed in silico to interact with host proteins involved in immune defense using protein analysis software (Protean version 4.0, Lasergene mechanism and signal transduction (Henriksson et al., 2004; Navigator; DNA STAR Inc; Madison, Wis; Chakhaiyar et al., Bhaduri et al., 2014) while PpiB has been reported in membrane 2004). fraction and mannosylation enriched culture filtrate (Cole et al., 1998; Gu et al., 2003). Immunological characterization of these Human Subjects enzymes in terms of their possible role in modulating host This Study was approved by Institutional Bioethics Committee, immune response, virulence and intracellular survival, has not and written consent was obtained from all participants. The been investigated so far. PpiB, being an essential protein (Sassetti two categories recruited for the study were: fresh confirmed et al., 2003), makes it an important target for developing new cases of pulmonary TB (n = 43) and healthy control (n = 43). interventions. Subjects reporting to hospital with symptoms of tuberculosis In the present study, we describe the immunogenic potential were selected on the basis of sputum smear positivity. Healthy of M. tb cyclophilins, their involvement in eliciting host immune controls were the volunteers with no symptom or history of response, altering the host cytokine profile and promoting the TB. HIV+ individuals were excluded from the study. Human intracellular survival of the pathogen, significant attributes which blood samples were collected and processed as described earlier highlight the seminal role of these proteins in the infection (Tundup et al., 2008). Briefly, the blood was withdrawn by process of M. tb. venipuncture of median cubital vein by a phlebotomist. Isolated blood was allowed to clot for half an hour at 37◦C. It was then centrifuged at 1500 × g for 15 min to remove the clot. The clear MATERIALS AND METHODS serum was aliquoted and stored at −80◦C until needed. Materials Immune Assays IPTG, imidazole, α-chemotrypsin, reduced Glutathione, N- The antibody levels in human serum were assayed by enzyme succinyl-Ala-Ala-Pro-Phe-p-nitroanilide, trifluoroethanol, LiCl, linked immunosorbent assay (ELISA) using microtiter plates DTT, H O , and polymyxin B were purchased from Sigma. Cell 2 2 (Corning) coated with rPpiA, and rPpiB respectively, as culture reagents were purchased from GIBCO, Thermo Fisher described earlier (Banerjee et al., 2004; Khubaib et al., 2016). Scientific (USA). All restriction enzymes were purchased from Briefly, 96 well plates were coated by specific proteins in PBS (10 New England Biolabs (USA); antibodies from Abcam (UK); and µg/ml) and kept at 4◦C overnight. Plate was washed three times ELISA kits from Peprotech (USA). Mycobacterium growth media with wash buffer and blocked for an hour at room temperature. and supplements were purchased from Becton, Dickinson and After three washes, serum samples in 1:100 dilutions were added Company (USA). All reagents used were analytical grade. The and kept for 2 h. Secondary conjugate antibody (1:10,000) was strains and plasmids used in this study are listed in Table S1. added for an hour. Plate was washed at least five times, TMB substrate was added, and reaction was stopped with 2N H2SO4. Mycobacterial Strains Ppiase induced secretion of cytokines from macrophages was Mycobacterium smegmatis mc2155, initially obtained from also quantified by ELISA as described earlier (Silswal et al., 2005; ATCC, was maintained in our laboratory as glycerol stocks. Nair et al., 2009). Differentiation of THP-1 cells was achieved Running culture was obtained by inoculating in 7H9 broth by phorbol 12-myristate 13-acetate (PMA) treatment (Nair supplemented with 10% OADC and 0.05% Tween 80. Cultures et al., 2009). Differentiated cells were plated and treated with were incubated at 37◦C in a shaker incubator for suspension appropriate concentration of respective proteins. Supernatants cultures. For CFU enumeration, bacteria were plated on 7H10 were harvested at various time points and quantified for Middlebrook agar plates supplemented with 10% OADC and determination of various cytokine levels. Briefly, 96 well ELISA ◦ incubated at 37 C. Genomic DNA of M. tb H37Rv used in plates were coated with capture antibody in coating buffer the cloning was kind gift from Dr. Manjula Sritharan and Dr. (bicarbonate/phosphate buffer) kept at 4◦C overnight. After an Sharmishtha Banerjee, University of Hyderabad, Hyderabad, hour of blocking with assay diluent, supernatants along with India. standards were added for 2 h. Enzyme conjugates were then

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 2 February 2017 | Volume 7 | Article 38 Pandey et al. M.tb Ppiases Aid Intracellular Survival

added for 1 h. TMB substrate was added and 2N H2SO4 was M. smegmatis strains were designated as Ms_ppiA, Ms_ppiB, and added to stop the reaction. Absorbance was measured at 450 nm Ms_VC (Table S1). and curve was plotted along with standards to determine the cytokine levels in test samples. All the usual steps of intermittent In vitro Growth and Stress Assay washings were included as per manufacturer’s instructions. Log phase cultures of M. smegmatis strains (Ms_WT, Ms_VC, Ms_ppiA, Ms_ppiB) were inoculated in Middlebrook 7H9 broth with 10% Oleic Albumin Dextrose Catalase (OADC) in the Cloning, Expression of M. tb Ppiases in presence/absence of kanamycin (25 µg/ml). The cultures were M. smegmatis grown at 37◦C at 200 rpm. The cell density was measured Cloning and expression of M. tb Ppiases in M. smegmatis was periodically at 600nm (OD600) using spectrophotometer. For carried out as described (Farhana et al., 2008; Tripathi et al., 2015) hydrogen peroxide stress, log phase cultures (OD600 of 0.8–1.0) using E. coli-mycobacterium shuttle vector pST2K and specific of M. smegmatis strains were diluted 1:100 into Middlebrook 7H9 oligonucleotide primers (Tables S1, S2). The resulting constructs, broth and grown for approximately 12 h until the OD600 reached pST_ppiA, pST_ppiB and pST2K vector were then electroporated 0.4. Re-inoculated cells were then treated with the 7 mM H2O2. in wild type M. smegmatis (Ms_WT). Transformed recombinant At0,1,2,and3h,100 µl samples were serially diluted and plated

FIGURE 1 | Mycobacterial Ppiases elicit B cell response in tuberculosis patients. (A) In silico analysis of PpiA and PpiB proteins (182 and 308 amino acids, respectively) of M. tb, using protein analysis software (Lasergene Navigator; DNA STAR) displayed major antigenic stretches with a peak value ≥1.7. (B) Reactivity to rPpiA and rPpiB was studied then in two sets (patients and healthy groups) by ELISA. Statistical analysis of immunoactivity revealed that the patient group mounted significantly higher (P < 0.001) immune response against rPpiases compared to healthy controls. The horizontal line indicates the mean of the absorbance values.

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 3 February 2017 | Volume 7 | Article 38 Pandey et al. M.tb Ppiases Aid Intracellular Survival

FIGURE 2 | PpiA stimulates secretion of pro-inflammatory cytokines by THP-1 cells. Concentration dependent increase in the release of pro-inflammatory cytokines (A, TNF-α; and B, IL-6) as a function of treatment of THP-1 cells with rPpiA (2.5, 5.0, 10, and 20 µg) for 24 h. Data represent the mean ± SD of three replicates. LPS was used as a positive control. (***P < 0.001). on Middlebrook 7H10 plates to determine the colony forming humoral immune response against the M. tb PpiA and PpiB in units (Li et al., 2014). patients and healthy individuals. Statistical analysis revealed that TB patient group mounted a significantly higher (P < 0.001) Uptake and Intracellular Growth of antibody response against rPpiA and rPpiB compared to healthy M. smegmatis Expressing M. tb Ppiases in group (Figure 1B). These results indicate that M. tb Ppiases Human THP-1 Macrophages elicit B-cell response and have possible immunomodulatory THP-1 monocytic cell line was cultured and differentiated with effect. PMA in RPMI 1640 medium as reported earlier (Nair et al., 2009; Tripathi et al., 2015). The THP-1 monolayers were infected with rPpiases Alter Cytokine Profile in THP-1 log-phase bacteria (Ms_WT, Ms_VC, Ms_ppiA, and Ms_ppiB), Macrophages for 4h at MOI of 50. The cells after infection were treated with Having shown the ability of M. tb Ppiases to elicit higher medium containing 20 µg/ml gentamicin for 30 min, and washed B-cell response, we investigated the ability of these proteins twice with RPMI medium. The plates were incubated at 37◦C in modulating the secretion of various cytokines. Human after adding fresh complete medium. The cells were dislodged monocytes (THP-1) differentiated with PMA were treated with gently at different time points and centrifuged at 2000 rpm for increasing concentrations of polymyxin B treated rPpiA and 3 min, washed twice with fresh RPMI 1640 medium and lysed in rPpiB. Soup was collected after 24 h to estimate levels of different sterile water. The lysate was then diluted in Middlebrook 7H9 cytokines by ELISA. Significant increase in the levels of TNF-α broth and plated on Middlebrook 7H10-OADC agar plates. The and IL-6 (Figure 2) was observed, when treated with increasing plates were incubated at 37◦C and colonies were counted after 4 concentration of rPpiA (2.5, 5, 10, 20 µg). However, no such days. For each data point, the mean of triplicate was used. significant increase in the levels of IL-10 could be noticed (Supplementary Figure 2). Interestingly, decrease in the level RESULTS of TNF-α and IL-6 along with increase in the levels of IL- 10 was observed (Figures 3A–C) when treated with increasing Mycobacterium tuberculosis Ppiases concentration of rPpiB (2.5, 5, 10, 20 µg). These results Display Antigenic Stretches and Also Elicit demonstrate that mycobacterial Ppiases alter cytokine secretion Significantly Higher B-Cell Response in TB in human monocytic-macrophage cell lines. These observations are in agreement with previous reports involving mycobacterial Patients chaperones and HSPs in immune modulation (Peetermans et al., His tagged M. tb rPpiA and GST tagged rPpiB were purified using 1994; Naffin-Olivos et al., 2014). Ni-NTA column and glutathione sepharose affinity column, respectively (Supplementary Figure 1). In silico antigenicity M. smegmatis Expressing M. tb Ppiases profiling of M. tb PpiA and PpiB protein (Figure 1A) using protein analysis software (Protean version 4.0, Lasergene Show Increased Survival under Hydrogen Navigator; DNA STAR Inc; Madison, Wis) displayed major Peroxide Stress antigenic stretches with a peak value ≥1.7 (Chakhaiyar et al., M. smegmatis strains (Ms_WT, Ms_VC, Ms_ppiA, and Ms_ppiB) 2004). To investigate if M. tb Ppiases could indeed elicit humoral were grown in Middlebrook 7H9 broth with 10% OADC and immune response, experiments were designed to compare growth profile was analyzed by measuring optical density at 600

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 4 February 2017 | Volume 7 | Article 38 Pandey et al. M.tb Ppiases Aid Intracellular Survival

FIGURE 4 | M. smegmatis carrying PpiA/PpiB do not alter bacterial growth in vitro. Bacterial growth (OD600) of wild type M. smegmatis (Ms_WT), M. smegmatis transformed with vector alone (Ms_VC), with vector carrying M. tuberculosis ppiA gene (Ms_ppiA) or ppiB gene (Ms_ppiB) was plotted. Data represent mean ± SD of values obtained from three independent cultures.

FIGURE 5 | M. smegmatis growth profile under conditions of oxidative stress. Secondary culture of wild type M. smegmatis mc2155, vector transformed M. smegmatis, or those transformed with vector plasmid carrying M. tuberculosis ppiA gene (Ms_ppiA), or ppiB gene (Ms_ppiB) were grown until O.D. = 0.4 and were treated with 7 mM H2O2 for 3 h and log c.f.u. was calculated at different time points. The data represent the mean ± SEM of triplicate wells and are representative of three individual experiments. (***P < 0.001).

rPpiases does not affect in vitro growth of M. smegmatis under normal conditions. For investigating survival under hydrogen FIGURE 3 | M. tb PpiB decreases secretion of TNF-α, IL-6 and stimulates secretion of IL-10 cytokine by THP-1 cells. Dose dependent peroxide stress, which mimics one of the stresses faced by decrease in the release of pro-inflammatory cytokines (A, TNF-α; B, IL-6) and intracellular mycobacteria, log phase cultures (OD600 of 0.4) concomitant increase in the release of anti-inflammatory cytokine (C, IL-10), of recombinant M. smegmatis strains were treated with the consequent to treatment by rPpiB (2.5, 5, 10, and 20 µg) for 24h. Data 7 mM concentration of H2O2 for a period of 3 h. Colony represent the mean ± SD of three replicates. (***P < 0.001). LPS was used as positive control and mean levels of TNF-α and IL-6 observed in LPS treated forming unit counts indicated a significant difference in the cells were 1,753 ± 135 and 2070 ± 272, respectively. survival of Ms_ppiA and Ms_ppiB as compared to the Ms_WT and Ms_VC cells (Figure 5). These results clearly indicate that M. tb cyclophilins play a critical role in stress adaptation, nm. No significant difference was observed in the growth profile pointing to their role in virulence, as reported earlier, for of recombinant M. smegmatis strains, as compared to the wild another intracellular pathogen Brucella abortus (Roset et al., type M. smegmatis (Figure 4), indicating that overexpression of 2013).

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 5 February 2017 | Volume 7 | Article 38 Pandey et al. M.tb Ppiases Aid Intracellular Survival

B-cell response (Mehlert and Young, 1989; Young and Garbe, 1991). These findings clearly indicate the immunodominant nature of these cyclophilins expressed during infection and also highlight their likely diagnostic potential. In M. tb infection cycle, macrophages act as the first line of defense and the pathogen hires a plethora of strategies to counteract the host immune response. Skewing of the balance in secretion of pro-inflammatory and anti-inflammatory cytokines results in intracellular bacterial clearance or survival, respectively (Newman et al., 1991; Byrd, 1997). Chronic inflammatory pathology is the hallmark of tuberculosis, which indicates that overproduction of pro-inflammatory cytokines lies at the heart of the infection. TNF-α is crucial for the formation and maintenance of granuloma that are main effector sites of FIGURE 6 | M. smegmatis expressing PpiA and PpiB show increased antimicrobial activity against the pathogen (Kindler et al., 1989). survival in THP-1 cells. THP cells were infected with wild type M. smegmatis Estimation of proinflammatory cytokines secreted by THP-1 mc2155, vector control transformed M. smegmatis, or those transformed with M. tuberculosis ppiA gene (Ms_ppiA), and ppiB (Ms_ppiB) and plated after 24, cells on treatment with increased doses of rPpiA revealed that 48, and 72 h and log c.f.u./monolayer was calculated at different time points. this Ppiase is a potent stimulator of TNF-α and IL-6 and thus The data represented are mean ± SEM of triplicate wells and are may play an important role in host inflammatory pathology. representative of three individual experiments. (***P < 0.001). Mycobacterial HSPs Hsp65 and chaperones have been earlier shown to act as strong immune modulators and are potent M. smegmatis Expressing M.tb Ppiases stimulators of pro-inflammatory cytokines (Peetermans et al., 1994; Naffin-Olivos et al., 2014). Show Increased Survival in Human THP-1 Our results implicate M. tb PpiA as a stimulator of pro- Cells inflammatory cytokines thereby pointing to its vital role in the To directly demonstrate the impact of M. tb PpiA and PpiB on inflammatory pathology of tuberculosis (Figure 7). Conversely, survivability of mycobacterium in host cells, recombinant M. assessment of pro-inflammatory cytokines secreted by THP-1 smegmatis strains were assayed for in vitro growth in THP-1 cells cells upon treatment with increasing concentrations of rPpiB and compared with wild type M. smegmatis. As could be seen, revealed that the protein is an inhibitor of TNF-α and IL-6 along recombinant Ms_ppiA and Ms_ppiB survived longer, upto 72 h with stimulating secretion of IL-10. That M. tb PpiB is able to (p < 0.001), in THP-1 cells than Ms_WT or Ms_VC (Figure 6). subvert the innate immune response points to its role in aiding An increase of ∼1 log was observed after 24 h in the control the establishment of infection, which fits well with previous group, after which the colony forming unit (c.f.u.) declined. In findings that PpiB is essential for the survival of the pathogen contrast, Ms_ppiA and Ms_ppiB strains continued to replicate (Sassetti et al., 2003). and survive up to 72 h within the macrophages. These results The kinetics of production along with balance between establish the critical role of Ppiases in intracellular survival. pro-inflammatory and anti-inflammatory cytokines released by macrophages upon exposure to mycobacterial antigens regulate DISCUSSION the T cell responses (Dheenadhayalan et al., 2006; Kim et al., 2011). Production of pro-inflammatory cytokines is known to Besides their biological function as helpers in protein folding, contribute to the host response against mycobacteria (Ladel bacterial molecular chaperones have a distinctive role in et al., 1997a; Tsao et al., 1999; Sasindran and Torrelles, 2011). virulence and stress tolerance. They are also termed as Our apparently paradoxical observation that PpiA increases moonlighting proteins (Vanghele and Ganea, 2010). Hsp70 intracellular survival despite the pro-inflammatory role suggests chaperone is present on bacterial surface and functions also an alternative role of PpiA mediated TNF-α and IL-6. Since PpiA as a plasminogen receptor in pathogens like M. tb, Neisseria is known to be expressed late in the infection cycle (Pathakumari meningitides, and Listeria monocytogenes (Knaust et al., 2007; et al., 2015), the pro-inflammatory cytokines induced by PpiA Xolalpa et al., 2007). We investigated whether mycobacterial is likely associated with immunopathological responses in cyclophilins also have roles beyond protein folding to modulate tuberculosis leading to necrosis and cachexy that aid in disease the host immune response and assist in pathogenesis. Since progression. Production of TNF-α has earlier been shown to be B cells can exert influence on T cells, they are considered as directly related to virulence correlating with intracellular viability important determinants in the outcome of infection with M. tb. (Newman et al., 1991; Engele et al., 2002) and is also known to In-silico analysis of M. tb PpiA and PpiB proteins displayed major promote growth of virulent M. tb in monocytes (Byrd, 1997). antigenic stretches with a peak value ≥1.7. This was vindicated by Thus, induction of PpiA mediated TNF-α by macrophages at the observation of significantly increased presence of antibodies the site of infection permits the multiplication of intracellular to M.tb Ppiases in sera from TB patients as compared to healthy bacteria and may therefore present an evasion mechanism individuals. This is consistent with earlier findings that members employed by M. tb. This pro-inflammatory response induced of stress family of proteins such as Hsp70 and Hsp10 elicit strong by high bacterial load is also known to induce apoptosis that

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 6 February 2017 | Volume 7 | Article 38 Pandey et al. M.tb Ppiases Aid Intracellular Survival

FIGURE 7 | Model of PpiA and PpiB action on M. tb pathogenesis. PpiB aids mycobacterial growth by production of inhibitory cytokine IL-10. PpiA induces secretion of pro-inflammatory cytokines and may be associated with immunopathological responses in tuberculosis, which aids in bacterial dissemination and disease progression. is associated with mycobacterial survival (Santucci et al., 2000). expression of this protein, late in infection cycle (Pathakumari Earlier studies have also shown that mycobacterial components et al., 2015). Moreover PpiA is also known to be upregulated in increase the production of pro-inflammatory cytokine TNF-α intraphagosomal niche during infections (Mattow et al., 2006). in macrophages/monocytes and this elevated level of TNF-α is On the contrary PpiB being essential for the in vitro growth of regarded as reason for mycobacterial persistence and virulence M. tb, suggests its expression early during infection. within human macrophages (Dheenadhayalan et al., 2006; Kim Intracellular infectious agents that have co-evolved in et al., 2011). IL-6, although important for control of tuberculosis long standing association with the host have acquired in mice model (Ladel et al., 1997b), in humans is considered mechanisms to persist within the host cell. The persistence as a correlate of disease progression because of its role in of pathogenic mycobacteria within the hostile environment of inflammation and pathology (Tsao et al., 1999; Ilonidis et al., host macrophages is in part due to the bacterial ability to adapt 2006). Conversely, PpiB employs the classical approach to aid to the stress conditions encountered. Our data suggest that mycobacterial growth by production of inhibitory cytokine IL- M. tb Ppiases might play a role in the intracellular survival by 10 (Figure 7). The different strategies employed by the same subverting the host cell defenses, such as oxidative stress as well as classes of protein is possibly a function of the difference in by immunomodulation. M. tb Hip1 has been shown to modulate their expression kinetics: PpiB being expressed early during macrophage responses through proteolysis of GroEL2 (Naffin- infection employs IL-10 to dampen the inflammatory host Olivos et al., 2014) and this augers well with our earlier report of response, whereas PpiA, expressed late in infection (Pathakumari chaperone activity of M.tb Ppiases (Pandey et al., 2016), though et al., 2015) aids in immunopathology. IL-10 secreted by the exact mechanism remains to be elucidated. In conclusion, host cell after mycobacterial infection is also known to our study demonstrates immunomodulatory potential of M. tb increases the intracellular bacterial survivability by blocking Ppiases and in the process unveils previously unknown functions phagosomal maturation (O’Leary et al., 2011). Although Ppiases of peptidyl-prolyl isomerases. The novel immunological features are constitutively expressed for their action as a peptidyl prolyl attributed herein to PpiB, an essential protein of M. tb, highlights isomerase to catalyze the cis-trans isomerization of proline imidic the importance of this protein/enzyme as an important target for peptide bonds in oligopeptides, the convergent evolution has the development of more efficacious therapeutic interventions adapted the mycobacterial proteins for moonlighting functions. against TB. Thus, different immunological response could be based upon the different expression patterns at different stages of infection ETHICS STATEMENT cycle. Although currently there are no gene expression data to claim this but an earlier study wherein PpiA was shown to induce This study was carried out in accordance with the IFN-γ in LTBI as compared to that of PTB suggested enhanced recommendations of Institutional Bioethics Committee,

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 7 February 2017 | Volume 7 | Article 38 Pandey et al. M.tb Ppiases Aid Intracellular Survival

Bhagwan Mahavir Medical Research Centre, Hyderabad, India from the Department of Biotechnology, Ministry of Science and with written informed consent from all subjects. All subjects gave Technology (DBT), Government of India. written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the Institutional ACKNOWLEDGMENTS Bioethics Committee. SP and MK thank the CSIR, Government of India (GoI) for AUTHOR CONTRIBUTIONS the award of Research Fellowship. DT is a recipient of Research Associateship from Department of Biotechnology, GoI, India. NE and SH conceptualized and designed the research; SH is a JC Bose National Fellow of the Department of Science SP, DT, and MK performed research; SP, DT, AK, JS, and Technology, Ministry of Science and Technology, GoI. SH SH, and NE carried out data analysis; GS recruited and is a Robert Koch Fellow of the Robert Koch Institute, Berlin, classified human subjects; SP, DT, JS, SH, and NE wrote the Germany. manuscript. SUPPLEMENTARY MATERIAL FUNDING The Supplementary Material for this article can be found This work was partially funded by a Centre of Excellence online at: http://journal.frontiersin.org/article/10.3389/fcimb. research grant (BT/PR12817/COE/34/23/2015) to SH and NE 2017.00038/full#supplementary-material

REFERENCES cytokines from human monocytic cells. Clin. Exp. Immunol. 91, 58–62. doi: 10.1111/j.1365-2249.1993.tb03354.x Banerjee, S., Nandyala, A. K., Raviprasad, P., Ahmed, N., and Hasnain, S. E. (2007). Gu, S., Chen, J., Dobos, K. M., Bradbury, E. M., Belisle, J. T., and Chen, X. Iron-dependent RNA-binding activity of Mycobacterium tuberculosis aconitase. (2003). Comprehensive proteomic profiling of the membrane constituents J. Bacteriol. 189, 4046–4052. doi: 10.1128/JB.00026-07 of a Mycobacterium tuberculosis strain. Mol. Cell. Proteomics 2, 1284–1296. Banerjee, S., Nandyala, A., Podili, R., Katoch, V. M., Murthy, K. J. R., and Hasnain, doi: 10.1074/mcp.M300060-MCP200 S. E. (2004). Mycobacterium tuberculosis (Mtb) isocitrate dehydrogenases show Henderson, B. (2010). “Heat shock proteins are mediators of bacterial-host strong B cell response and distinguish vaccinated controls from TB patients. interactions,” in Prokaryotic and Eukaryotic Heat Shock Proteins in Infectious Proc. Natl. Acad. Sci. U.S.A. 101, 12652–12657. doi: 10.1073/pnas.0404347101 Disease, eds A. G. Pockley, S. K. Calderwood, and M. G. Santoro (Dordrecht: Bhaduri, A., Misra, R., Maji, A., Bhetaria, P. J., Mishra, S., Arora, G., et al. Springer), 185–209. (2014). Mycobacterium tuberculosis cyclophilin A uses novel signal sequence for Henriksson, L. M., Johansson, P., Unge, T., and Mowbray, S. L. secretion and mimics eukaryotic cyclophilins for interaction with host protein (2004). X-ray structure of peptidyl-prolyl cis-trans isomerase a repertoire. PLoS ONE 9:e88090. doi: 10.1371/journal.pone.0088090 from Mycobacterium tuberculosis. Eur. J. Biochem. 271, 4107–4113. Byrd, T. F. (1997). Tumor necrosis factor alpha (TNFα) promotes growth doi: 10.1111/j.1432-1033.2004.04348.x of virulent Mycobacterium tuberculosis in human monocytes iron- Ilonidis, G., Parapanisiou, E., Anogeianaki A., Giavazis I., Theofilogiannakos E. mediated growth suppression is correlated with decreased release of K., Tsekoura P., et al. (2006). Interleukin-1beta (IL-1 beta), interleukin 6 (IL-6) TNFα from iron-treated infected monocytes. J. Clin. Invest. 99, 2518–2529. and tumor necrosis factor (TNF) in plasma and pleural fluid of pneumonia, doi: 10.1172/JCI119436 lung cancer and tuberculous pleuritis. J. Biol. Regul. Homeost. Agents 20, Chakhaiyar, P., Nagalakshmi, Y., Aruna, B., Murthy, K. J. R., Katoch, V. M., and 41–46. Hasnain, S. E. (2004). Regions of high antigenicity within the hypothetical Khubaib, M., Sheikh, J. A., Pandey, S., Srikanth, B., Bhuwan, M., Khan, N., et al. PPE major polymorphic tandem repeat open-reading frame, Rv2608, show a (2016). Mycobacterium tuberculosis Co-operonic PE32/PPE65 proteins alter differential humoral response and a low T cell response in various categories of host immune responses by Hampering Th1 response. Front. Microbiol. 7:719. patients with tuberculosis. J. Infect. Dis. 190, 1237–1244. doi: 10.1086/423938 doi: 10.3389/fmicb.2016.00719 Cole, S. T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., et al. (1998). Kim, K.-H., Yang, C.-S., Shin, A.-R., Jeon, S.-R., Park, J.-K., Kim, H.-J., et al. (2011). Deciphering the biology of Mycobacterium tuberculosis from the complete Mycobacterial heparin-binding hemagglutinin antigen activates Inflammatory genome sequence. Nature 393, 537–544. doi: 10.1038/31159 responses through PI3-K/Akt, NF-κB, and MAPK pathways. Immune Netw. 11, Dheenadhayalan, V., Delogu, G., and Brennan, M. J. (2006). Expression of 123–133. doi: 10.4110/in.2011.11.2.123 the PE_PGRS 33 protein in Mycobacterium smegmatis triggers necrosis in Kindler, V., Sappino, A. P., Grau, G. E., Piguet, P. F., and Vassalli, P. macrophages and enhanced mycobacterial survival. Microbes Infect. 8, 262–272. (1989). The inducing role of tumor necrosis factor in the development doi: 10.1016/j.micinf.2005.06.021 of bactericidal granulomas during BCG infection. Cell 56, 731–740. Ehrt, S., and Schnappinger, D. (2009). Mycobacterial survival strategies in the doi: 10.1016/0092-8674(89)90676-4 phagosome: defence against host stresses. Cell Microbiol. 11, 1170–1178. Knaust, A., Weber, M. V., Hammerschmidt, S., Bergmann, S., Frosch, doi: 10.1111/j.1462-5822.2009.01335.x M., and Kurzai, O. (2007). Cytosolic proteins contribute to surface Engele, M., Stössel, E., Castiglione, K., Schwerdtner, N., Wagner, M., Bolcskei, plasminogen recruitment of Neisseria meningitidis. J. Bacteriol. 189, 3246–3255. P., et al. (2002). Induction of TNF in human alveolar macrophages as doi: 10.1128/JB.01966-06 a potential evasion mechanism of virulent Mycobacterium tuberculosis. J. Ladel, C. H., Blum, C., Dreher, A., Reifenberg, K., Kopf, M., and Kaufmann, S. Immunol. 168:1328. doi: 10.4049/jimmunol.168.3.1328 H. (1997a). Lethal tuberculosis in interleukin-6-deficient mutant mice. Infect. Farhana, A., Kumar, S., Rathore, S. S., Ghosh, P. C., Ehtesham, N. Z., Tyagi, A. Immun. 65, 4843–4849. K., et al. (2008). Mechanistic Insights into a novel exporter-importer System Ladel, C. H., Szalay, G., Riedel, D., and Kaufmann, S. H. (1997b). Interleukin-12 of Mycobacterium tuberculosis unravel its role in trafficking of iron. PLoS ONE secretion by Mycobacterium tuberculosis-infected macrophages. Infect. Immun. 3:e2087. doi: 10.1371/journal.pone.0002087 65, 1936–1938. Friedland, J. S., Shattock, R., Remick, D. G., and Griffin, G. E. (2008). Li, W., Zhao, Q., Deng, W., Chen, T., Liu, M., and Xie, J. (2014). Mycobacterium Mycobacterial 65-kD heat shock protein induces release of proinflammatory tuberculosis Rv3402c enhances mycobacterial survival within macrophages

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 8 February 2017 | Volume 7 | Article 38 Pandey et al. M.tb Ppiases Aid Intracellular Survival

and modulates the host pro-inflammatory cytokines production via NF-kappa Sassetti, C. M., Boyd, D. H., and Rubin, E. J. (2003). Genes required for B/ERK/p38 signaling. PLoS ONE 9:e94418. doi: 10.1371/journal.pone.0094418 mycobacterial growth defined by high density mutagenesis. Mol. Microbiol. 48, Lu, K. P., Hanes, S. D., and Hunter, T. (1996). A human peptidyl-prolyl isomerase 77–84. doi: 10.1046/j.1365-2958.2003.03425.x essential for regulation of mitosis. Nature 380, 544–547. doi: 10.1038/380544a0 Silswal, N., Singh, A. K., Aruna, B., Mukhopadhyay, S., Ghosh, S., and Ehtesham, Mark, P. J., Ward, B. K., Kumar, P., Lahooti, H., Minchin, R. F., and Ratajczak, N. Z. (2005). Human resistin stimulates the pro-inflammatory cytokines T. (2001). Human cyclophilin 40 is a heat shock protein that exhibits altered TNF-α and IL-12 in macrophages by NF-kappaB-dependent pathway. intracellular localization following heat shock. Cell Stress Chaperones 6, 59–70. Biochem. Biophys. Res. Commun. 334, 1092–1101. doi: 10.1016/j.bbrc.2005. Mattow, J., Siejak, F., Hagens, K., Becher, D., Albrecht, D., Krah, A., et al. (2006). 06.202 Proteins unique to intraphagosomally grown Mycobacterium tuberculosis. Sykes, K., Gething, M. J., and Sambrook, J. (1993). Proline isomerases Proteomics 6, 2485–2494. doi: 10.1002/pmic.200500547 function during heat shock. Proc. Natl. Acad. Sci. U.S.A. 90, 5853–5857. Mehlert, A., and Young, D. B. (1989). Biochemical and antigenic doi: 10.1073/pnas.90.12.5853 characterization of the Mycobacterium tuberculosis 71kD antigen, a Tripathi, D., Kant, S., Garg, R., and Bhatnagar, R. (2015). Low expression level of member of the 70kD heat-shock protein family. Mol. Microbiol. 3, 125–130. glnA1 accounts for absence of cell wall associated poly-l-glutamate/glutamine doi: 10.1111/j.1365-2958.1989.tb01801.x in Mycobacterium smegmatis. Biochem. Biophys. Res. Commun. 458, 240–245. Naffin-Olivos, J. L., Georgieva, M., Goldfarb, N., Madan-Lala, R., Dong, L., Bizzell, doi: 10.1016/j.bbrc.2015.01.079 E., et al. (2014). Mycobacterium tuberculosis Hip1 modulates macrophage Tsao, T. C., Hong, J., Huang, C., Yang, P., Liao, S. K., and Chang, K. S. responses through proteolysis of GroEL2. PLoS Pathog. 10:e1004132. (1999). Increased TNF-α, IL-1 β and IL-6 levels in the bronchoalveolar lavage doi: 10.1371/journal.ppat.1004132 fluid with the upregulation of their mRNA in macrophages lavaged from Nair, S., Ramaswamy, P. A., Ghosh, S., Joshi, D. C., Pathak, N., Siddiqui, I., patients with active pulmonary tuberculosis. Tuber. Lung Dis. 79, 279–285. et al. (2009). The PPE18 of Mycobacterium tuberculosis Interacts with TLR2 doi: 10.1054/tuld.1999.0215 and Activates IL-10 Induction in Macrophage. J. Immunol. 183, 6269–6281. Tundup, S., Pathak, N., Ramanadham, M., Mukhopadhyay, S., Murthy, K. J. doi: 10.4049/jimmunol.0901367 R., Ehtesham, N. Z., et al. (2008). The co-operonic PE25/PPE41 protein Newman, G. W., Gan, H. X., McCarthy, P. L., and Remold, H. G. (1991). Survival complex of Mycobacterium tuberculosis elicits increased humoral and cell of human macrophages infected with Mycobacterium avium intracellulare mediated immune response. PLoS ONE 3:e3586. doi: 10.1371/journal.pone.00 correlates with increased production of tumor necrosis factor-alpha and IL-6. 03586 J. Immunol. 147, 3942–3948. Vanghele, M., and Ganea, E. (2010). The role of bacterial molecular chaperones in O’Leary, S., O’Sullivan, M. P., and Keane, J. (2011). IL-10 blocks phagosome pathogen survival within the host. Rom. J. Biochem. 100, 87–100. maturation in Mycobacterium tuberculosis-infected human macrophages. Am. Wu, X., Wilcox, C. B., Devasahayam, G., Hackett, R. L., Arévalo-Rodríguez, M., J. Respir. Cell Mol. Biol. 45, 172–180. doi: 10.1165/rcmb.2010-0319OC Cardenas, M. E., et al. (2000). The Ess1 prolyl isomerase is linked to chromatin Pandey, S., Sharma, A., Tripathi, D., Kumar, A., Khubaib, M., Bhuwan, M., remodeling complexes and the general transcription machinery. EMBO J. 19, et al. (2016). Mycobacterium tuberculosis peptidyl-prolyl isomerases also 3727–3738. doi: 10.1093/emboj/19.14.3727 exhibit chaperone like activity in-vitro and in-vivo. PLoS ONE 11:e0150288. Xolalpa, W., Vallecillo, A. J., Lara, M., Mendoza-Hernandez, G., Comini, M., doi: 10.1371/journal.pone.0150288 Spallek, R., et al. (2007). Identification of novel bacterial plasminogen-binding Pathakumari, B., Anbarasu, D., Parthasarathy, R. T., and Raja, A. proteins in the human pathogen Mycobacterium tuberculosis. Proteomics 7, (2015). PpiA antigen specific immune response is a potential 3332–3341. doi: 10.1002/pmic.200600876 biomarker for latent tuberculosis infection. Tuberculosis 95, 736–743. Young, D. B., and Garbe, T. R. (1991). Heat shock proteins and antigens of doi: 10.1016/j.tube.2015.07.006 Mycobacterium tuberculosis. Infect. Immun. 59, 3086–3093. Peetermans, W. E., Raats, C. J., Langermans, J. A., and Van Furth, R. (1994). Zhang, X. C., Wang, W. D., Wang, J. S., and Pan, J. C. (2013). PPIase independent Mycobacterial heat-shock protein 65 induces proinflammatory cytokines but chaperone-like function of recombinant human Cyclophilin a during arginine does not activate human mononuclear phagocytes. Scand. J. Immunol. 39, kinase refolding. FEBS Lett. 587, 666–672. doi: 10.1016/j.febslet.2013.01.028 613–617. doi: 10.1111/j.1365-3083.1994.tb03421.x Roset, M. S., García Fernández, L., DelVecchio, V. G., and Briones, G. Conflict of Interest Statement: The authors declare that the research was (2013). Intracellularly induced cyclophilins play an important role in stress conducted in the absence of any commercial or financial relationships that could adaptation and virulence of Brucella abortus. Infect. Immun. 81, 521–530. be construed as a potential conflict of interest. doi: 10.1128/IAI.01125-12 Santucci, M. B., Amicosante, M., Cicconi, R., Montesano, C., Casarini, M., Copyright © 2017 Pandey, Tripathi, Khubaib, Kumar, Sheikh, Sumanlatha, Giosuè, S., et al. (2000). Mycobacterium tuberculosis-induced apoptosis in Ehtesham and Hasnain. This is an open-access article distributed under the terms monocytes/macrophages: early membrane modifications and intracellular of the Creative Commons Attribution License (CC BY). The use, distribution or mycobacterial viability. J. Infect. Dis. 181, 1506–1509. doi: 10.1086/315371 reproduction in other forums is permitted, provided the original author(s) or licensor Sasindran, S. J., and Torrelles, J. B. (2011). Mycobacterium tuberculosis infection are credited and that the original publication in this journal is cited, in accordance and inflammation: what is beneficial for the host and for the bacterium? Front. with accepted academic practice. No use, distribution or reproduction is permitted Microbiol. 2:2. doi: 10.3389/fmicb.2011.00002 which does not comply with these terms.

Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 9 February 2017 | Volume 7 | Article 38

201 6

www.nature.com/scientificreports

OPEN Cytotoxic and apoptotic effects of heat killed Mycobacterium indicus pranii (MIP) on various human received: 13 August 2015 accepted: 18 December 2015 cancer cell lines Published: 28 January 2016 Menaga Subramaniam1, Lionel LA In2, Ashutosh Kumar3, Niyaz Ahmed3 & Noor Hasima Nagoor1,4

Mycobacterium indicus pranii (MIP) is a non-pathogenic mycobacterium, which has been tested on several cancer types like lung and bladder where tumour regression and complete recovery was observed. In discovering the potential cytotoxic elements, a preliminary test was carried out using four different fractions consisting of live bacteria, culture supernatant, heat killed bacteria and heat killed culture supernatant of MIP against two human cancer cells A549 and CaSki by 3-(4,5-dimethyl thiazol)- 2,5-diphenyl tetrazolium bromide (MTT) assay. Apoptosis was investigated in MCF-7 and ORL-115 cancer cells by poly-(ADP-ribose) polymerase (PARP) and DNA fragmentation assays. Among four MIP fractions, only heat killed MIP fraction (HKB) showed significant cytotoxicity in various cancer cells with 6 inhibitory concentration, IC50 in the range 5.6–35.0 μl/(1.0 × 10 MIP cells/ml), while cytotoxicity effects were not observed in the remaining fractions. HKB did not show cytotoxic effects on non-cancerous cells contrary to cancerous cells, suggesting its safe usage and ability to differentially recognize between these cells. Evaluation on PARP assay further suggested that cytotoxicity in cancer cells were potentially induced via caspase-mediated apoptosis. The cytotoxic and apoptotic effects of MIP HKB have indicated that this fraction can be a good candidate to further identify effective anti-cancer agents.

The use of bacteria in cancer treatment is a well-known approach which was championed by W. Coley and German physicians W. Busch and F. Fehleisen who reported recovery of neck and other cancers following an infection with Streptococcus pyogenes1. Following these discoveries, several other bacterial species have been found to elicit significant anti-tumour activity in both in vitro and in vivo systems such as Lactobacillus spe- cies on bladder cancer2 and attenuated Salmonella species in murine tumour models3. Mycobacteria may be yet another promising species as it has shown a long successful history in treating cancer, for instance, the bacil- lus Calmette-Gue’rin (BCG) vaccine derived from Mycobacterium bovis was reported to be effective in treating human bladder cancer4. Bacteria based anti-tumour therapy possess several advantages over chemical based drug. Firstly, some bac- teria are able to selectively replicate and accumulate within tumour due to hypoxia environment and inhibits tumour growth. Next, motile bacteria are able to spread throughout the tumour and help in targeting systemic diseases. They can readily express multiple therapeutic transgenes such as cytokines and pro-drug converting enzymes to eradicate tumour mass5. Mycobacterium indicus pranii (MIP) or conventionally known as Mycobacterium w (M.w) is a non-pathogenic, cultivable Mycobacterium species; which is now used widely as a vaccine for a number of diseases6. This vaccine works by boosting up the patient’s immunity through the induction of CD4+ T helper 1 (Th-1) cells response to release cytokines IL-2, IL-12, IL-15 and IFN-γ in order to promote cell-mediated immunity, and has been

1Institute of Biological Science (Genetics and Molecular Biology), Faculty of Science, University of Malaya, 50603, Kuala Lumpur, Malaysia. 2Department of Biotechnology, Faculty of Applied Sciences, UCSI University, 56000, Kuala Lumpur, Malaysia. 3Pathogen Biology Laboratory, Department of Biotechnology, School of Life Sciences, University of Hyderabad, Professor C.R. Rao Road, Hyderabad, Andhra Pradesh 500046, India. 4Centre for Research in Biotechnology for Agriculture (CEBAR), University of Malaya, 50603 Kuala Lumpur, Malaysia. Correspondence and requests for materials should be addressed to N.H.N. (email: [email protected])

Scientific Reports | 6:19833 | DOI: 10.1038/srep19833 1 www.nature.com/scientificreports/

Figure 1. Cytotoxicity assay using MIP fractions at 24 hrs in human cervical carcinoma cell line (CaSki) and human lung carcinoma cell line (A549). MIP fractions: MIP live bacteria (LB), MIP culture supernatant (CS), MIP heat killed bacteria (HKB) and MIP heat killed culture supernatant (HKS). All MTT data were represented as mean ± SD of three independent experiments.

reported to be safe for human use in the treatment of leprosy6, tuberculosis7, HIV infection8 and lung cancer9 diseases. Apart from inducing the immune system, certain mycobacteria species were also reported to induce a direct cytotoxic effect on cancer cells. As reported by Saitoh and Morales, BCG and BCG components4 could induce cancer cell apoptosis, while M. phlei or mycobacterial cell wall and DNA components possess certain anti-tumour activity10. These findings provides an insight on exploiting MIP and its cellular components as a potential anti-cancer agent against various human cancer cell lines. As such to date, only certain types of cancers were reported to show cytotoxic effects upon MIP treatment in in vivo11,12. Therefore, there is a great interest to discover MIP cytotoxicity on various human cancer cell lines to broaden its utility. In this study, we evaluated in vitro cytotoxicity effect of four different MIP fractions consist- ing of live bacteria, culture supernatant, heat killed bacteria and heat killed culture supernatant against various human cancer type namely breast, cervical, oral, lung, bladder, liver and prostate. Results Cytotoxic screening for active MIP fractions. MIP was separated into four fractions: live bacteria (LB), culture supernatant (CS), heat killed bacteria (HKB) and heat killed culture supernatant (HKS). In identifying the active MIP fraction with cytotoxic effects, all four fractions were treated in two different cancer cell lines; cervical (CaSki) and lung (A549). The cell viability upon 24 hrs treatment was measured using MTT assay based on the mitochondrial activity in viable cells. In both cancer cells, only MIP HKB fraction showed cytotoxic effects where cell viability reduced to 24% in CaSki and 26% in A549, while the remaining fractions did not show any killing effects (Fig. 1). Thus, MIP HKB fraction was used here after to assess its cytotoxic consistency on various other cancer cell lines.

In vitro cytotoxic effects of MIP HKB. Cytotoxic applicability of the MIP HKB fraction was analyzed on various human cancer cell types (breast, cervical, lung adenocarcinoma, prostate, liver, bladder and oral) and non-cancerous cells (immortalized human cell lines from keratinocyte (HaCaT), nasopharyngeal epithelial (NP- 69) and breast epithelial (MCF-10A)) as controls at 24 hrs treatment using MTT cell viability assay. Results indi- cated that cells treated with MIP HKB induced cytotoxicity in a dose dependent manner, similar to that of CaSki and A549 cells (Fig. 2). Table 1 shows IC50 values of MIP heat killed bacteria on various human cancer cell lines. 6 Highest cytotoxicity was observed in liver cancer cell, HepG2 with an IC50 of 5.6 μ l/(1.0 × 10 MIP cells/ml) at 24 hrs. Oral cancer cells showed the second highest cytotoxicity (ORL-48, ORL-115 and ORL-136), with IC50 values of 13.6 μ l/(1.0 × 106 MIP cells/ml), 7.8 μ l/(1.0 × 106 MIP cells/ml) and 5.9 μ l/(1.0 × 106 MIP cells/ml), respec- 6 tively followed by lung and breast cancers. HaCaT, MCF-10A and NP-69 with IC50 values of 23.5 μ l/(10 cells/ml), 25.7 μ l/(106 cells/ml) and 32.9 μ l/(106 cells/ml) respectively, implies that concentrations higher than these values

Scientific Reports | 6:19833 | DOI: 10.1038/srep19833 2 www.nature.com/scientificreports/

Figure 2. Cytotoxicity of MIP heat killed bacteria at 24 hrs in various human cancer cell lines by MTT assay. Bladder cancer cell lines (RT-112 and EJ-28); breast cancer cell lines (MDA-MB-231 and MCF-7); liver carcinoma cell line (HepG2); prostate cancer cell lines (PC-3 and DU-145); cervical carcinoma cell lines (CaSki, and HeLa S3); lung carcinoma cell lines (A549 and SK-LU-1); oral cancer cell lines (ORL-48, ORL-115 and ORL-136); immortalized human keratinocyte cell line (HaCaT); normal human nasopharyngeal epithelial cell line (NP-69); normal human breast epithelial (MCF-10A). Data is shown as mean ± S.D. of three independent replicates.

6 IC50 [μl/(1.0 × 10 Cell type Cell line MIP cells/ml)] Immortalized human keratinocytes HaCat 23.5 ± 5.4 Immortalized human nasopharyngeal epithelial NP-69 32.9 ± 1.0 Immortalized human breast epithelial MCF-10A 25.7 ± 0.6 MDA-MB-231 15.4 ± 0.1 Breast cancer MCF-7 12.0 ± 0.7 CaSki 15.9 ± 1.8 Cervical cancer HeLa S3 21.1 ± 2.2 A549 14.3 ± 1.3 Lung cancer SK-LU-1 7.8 ± 2.8 PC-3 34.5 ± 1.6 Prostate cancer DU-145 18.4 ± 1.7 Liver cancer HepG2 5.6 ± 0.2 EJ-28 51.9 ± 2.0 Bladder cancer RT-112 35.5 ± 3.2 ORL-48 13.6 ± 1.0 Oral cancer ORL-115 7.8 ± 1.0 ORL-136 5.9 ± 0.5

Table 1. IC50 values of MIP HKB treated fraction on various human cancer and non-cancerous cell lines.

6 are toxic to non-cancerous cells. Also IC50 values higher than 23 μ l/(1.0 × 10 MIP cells/ml) in several cancer cells (PC-3, 34.5 μ l/(1.0 × 106 MIP cells/ml); EJ-28, 51.9 μ l/(1.0 × 106 MIP cells/ml); RT-112, 35.5 μ l/(1.0 × 106 MIP cells/ml) indicates heat killed MIP treatment is less effective in these cancer cells.

DNA fragmentation and PARP assay. To further analyze the mode of cell death upon MIP HKB treat- ment, MCF-7 and ORL-115 cells were selected as model cell lines owing to it having an IC50 value below the HaCat cell line threshold. The morphological changes in both cells shows MIP induced apoptotic cell death (data not shown). The PARP cleavage assay was carried out to validate the apoptosis mediated cell death in both cell lines. Results showed cleavage of the inhibitory fragment from the 116 kDa full length PARP into an 89 kDa fragment (Fig. 3). Cells were treated with PBS and MIP HKB, 12 μ l/(1.0 × 106 MIP cells/ml) for MCF-7 while 7.8 μ l/(1.0 × 106 MIP cells/ml) for ORL-115 in a time dependent manner at 6 and 12 hrs to observe the initiation and progression of apoptosis. The housekeeping gene, GAPDH was used as a protein normalization and loading control. DNA fragmentation assay was carried out to confirm and observe the occurrence of late apoptosis in MCF-7 and ORL-115 cells at 6, 12 and 24 hrs. A 150 bp to 200 bp laddering of DNA at 12 hrs upon MIP exposure in MCF-7 indicates a strong hallmark of late apoptotic events (Fig. 4). Ladder formation was absent in both untreated and PBS treated cells, which showed that the appearance of apoptotic DNA fragments were due to the cytotoxic effect of MIP HKB treatment.

Scientific Reports | 6:19833 | DOI: 10.1038/srep19833 3 www.nature.com/scientificreports/

Figure 3. Effects of MIP HKB on PARP cleavage at 6 and 12 hr. i. MCF-7 cell line; ii. ORL-115 cell line. (a) Cells were treated with MIP HKB at 6 and 12 hrs and PARP was measured by the Western blot analysis. (b) GAPDH was used as a loading control. Lane M: Biotinylated protein ladder; Lane 1: Untreated cells; Lane 2: PBS treated cells; Lane 3: HKB treated cells.

Figure 4. DNA gel electrophoresis of inter-nucleosome DNA fragmentation in 1.5% (w/v) agarose gel at 6, 12 and 24 hrs treatment in MCF-7 and ORL-115 cell lines. Lane P: positive control with 1’-(S)-1’- acetoxychavicol acetate (ACA) treated MCF-7 cells19; Lane N: negative control (untreated MCF-7 cells); Lane M: DNA molecular weight marker; Lane 1: Untreated cells; Lane 2: PBS treated cells for 24 hrs; Lane 3: HKB treated at 6 hrs; Lane 4: HKB treated at 12 hrs; Lane 5: HKB treated at 24 hrs. DNA laddering was demonstrated in cells treated with MIP HKB in Lane 5.

Discussion In cancer treatment, MIP is used as an adjuvant to radiation therapy in patients with bladder cancer12 and to chemotherapy plus radiotherapy in non-small cell lung cancers10. Four types of fractions can be obtained from MIP: LB, HKB, CS and HKS, with the most widely used fractions being the HKB fraction13,14 and CS fraction15. While past studies have cited autoclaving for 20 mins at 15 lb/in2 as the most common heat killing method, this method may denature important and biologically active proteins, which led us to heat-kill MIP at 60 °C, which was also found to be sufficient in killing MIP cultures. When all MIP fractions were cultivated in 7H10 agar, no growth was observed after a week of incubation, with the exception of LB fraction, thus confirming the com- plete killing of MIP at 60 °C. This method is recommended because even though MIP cultures were completely heat-killed, other intracellular and extracellular proteins/precursors potentially responsible for its cytotoxicity would likely remain intact. MIP HKB demonstrated therapeutic cytotoxicity against most of the tested human cancer cells, and was less potent towards non-cancerous human cells based on its high IC50 value. The difference in MIP selectivity between non-cancerous cells and cancer cells may be due to differences in growth rate, which results from the presence of distinct cell surface receptors, differences in the uptake of certain drugs and the method used for assessment of toxicity16. This selective cytotoxic effect is an important criteria to ensure the drug’s safety and efficacy in patients with minimal side effects. Experimental results imply MIP HKB selectivity in MDA-MB-231, MCF-7, CaSki, A549, SK-LU-1, DU-145, HepG2, ORL-48, ORL-115 and ORL-136 with IC50 values between 5.6 to 21 μ l/ (1.0 × 106 MIP cells/ml), all of which are lower than non-cancerous cells. According to a previous study on apop- totic cell death in in vitro by Pandey et al., 201115, 60–70 μ l of MIP is required to induce cell death in 40–45% mouse peritoneal macrophages while in this study, 60–70 μ l/(1.0 × 106 MIP cells/ml) of MIP HKB induced 75% cell death. This clearly shows a tremendous reduction in MIP dose when a 60 °C heat kill technique was applied compared to autoclave heat kill technique. This study also identified that cancer cell death was induced via apoptosis in MCF-7 and ORL-115 cells as con- firmed through PARP and DNA fragmentation assays. Apoptosis is a cell suicide mechanism to remove redun- dant, damaged, or infected cells through a group of caspases activation. These caspases are grouped into initiator (caspases-2, - 8, - 9, and - 10) and effector (caspases-3, -6 and - 7) caspases. Effector caspases are responsible for dismantling of necessary cell components, which results in morphological and biochemical changes that char- acterize apoptotic cell death as cytoskeletal rearrangement, cell membrane blebbing, nuclear condensation and

Scientific Reports | 6:19833 | DOI: 10.1038/srep19833 4 www.nature.com/scientificreports/

DNA fragmentation. The DNA fragmentation observed in MCF-7 cells, a caspase-3 deficient cell line, was most probably due to other effector caspases, such as, caspase 7 activation17. Materials and Methods Materials. Dulbecco modified Eagle medium (DMEM) supplemented with 4.5 g glucose/L and 300 mg/L L − glutamine was purchased from Hyclone Laboratories Inc, Logan, Utah. Roswell Park Memorial Institute 1640 (RPMI-1640) was purchased from Thermo Scientific Hyclone, USA. Fetal bovine serum (FBS) was purchased from Lonza Inc. (Allendale, New Jersey, USA). Keratinocyte serum-free medium (KSFM) and TRIzol reagent were obtained from Invitrogen, Grand Island, New York. Minimum essential medium alpha (MEM-α ) was pur- chased from Life Technologies, USA. Cisplatin and 3–(4, 5-dimethylthiazol-2-gl)-2, 5-diphenyl-tetrazoliumbro- mide (MTT) reagents were obtained from EMD Chemicals Inc. Middlebrook 7H10 agar and 7H9 broth were obtained from Sigma-Aldrich, Germany. Mycobacterium indicus pranii was provided by Prof. Dr. Niyaz Ahmed, Department of Biotechnology & Bioinformatics, School of Life Sciences, University Hyderabad, India.

Cultivation of cancer cells. A total of fourteen human cancer and three non-cancerous cell lines were used in this study: breast adenocarcinoma cell lines (MCF-7 and MDA-MB-231), hepatocyte carcinoma cell line (HepG2), cervical cancer cell lines (CaSki and HeLa S3), prostate carcinoma cell lines (PC-3 and DU-145), lung adenocarcinoma cell lines (A549 and SK-LU-1), oral cancer cell lines (ORL-48, ORL-115 and ORL-136) and bladder cancer cell lines (RT-112 and EJ-28). Immortalized human cell lines from nasopharyngeal epithelial (NP- 69), breast epithelial (MCF-10A) and keratinocyte (HaCaT) were used as representatives of non-cancerous cells. All cell lines were obtained from ATCC except human oral cancer cell lines which were obtained from Cancer Research Initiative Foundation (CARIF, Malaysia) and NP-69 is a gift from Prof GSW Tsao, The University of Hong Kong18. HaCaT, HepG2, CaSki and HeLa S3 were cultured in DMEM, while MCF-7, PC-3, MDA-MB-231, A549, RT-112, EJ-28 and DU-145 cells were cultured in RPMI 1640, supplemented with 10.0% (v/v) FBS. SK-LU-1 cells were cultured in MEM-α supplemented with 10% (v/v) heat inactivated FBS. NP-69 cells were cultured in keratinocyte serum-free medium (KSFM) (Gibco, USA) supplemented with 1 × 2.5 μ g human recom- binant epidermal growth factor (rEGF) (Gibco, USA) and 1 × 2.5 mg bovine pituitary extract (Gibco, USA). The MCF-10A was grown in serum-free mammary epithelial basal media (MEBM, Lonza, USA) supplemented with cholera toxin (100 ng/ml). All cells were grown as monolayers and were maintained in a humidified CO2 incu- 4 bator at 37 °C in 5.0% CO2 and 95.0% air. Each cell line was seeded at a density of 1.0 × 10 cells/well in 96-well plates (100 μ l/well) and left overnight in the incubator prior to commencement of treatment.

Bacterial cultures. MIP was cultured in Middlebrook (MB) 7H9 broth supplemented with 5ml glycerol, 0.2% Tween-80, 10% albumin-dextrose complex enrichment (ADC) and incubated at 37 oC, 100 rpm agitation until 1.5 OD600.

9 Preparation of MIP fractions. Original MIP suspension containing 6.0 × 10 MIP cells/ml at 1.5 OD600 as determined using CFU plate count assays was harvested to prepare MIP fractions: live bacteria (LB), culture supernatant (CS), heat killed bacteria (HKB) and heat killed culture supernatant (HKS). The suspension was centrifuged at 3500 rpm for 15 mins and supernatant (CS fraction) and MIP pellets were separated. Pellets were suspended in 0.9% (w/v) sodium chloride and 0.01% (w/v) thimerosal and centrifuged at 3500 rpm for 10 mins. The supernatant was then discarded. The pellet containing MIP cells was washed and resuspended in the original volume in sterile PBS, creating the LB fraction. HKB and HKS fractions were then prepared by heat inactivating the CS and LB fractions at 60 °C for 20 mins in a water bath. Various cell lines were treated with different volumes [10.0 to 100.0 μ l/(1.0 × 106 MIP cells/ml)] of MIP fractions.

MTT cell viability assay. MTT assay was carried out to measure cytotoxic effects of MIP fractions on vari- ous cancer cell lines. 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) is a substrate which is reduced by dehydrogenase enzymes present in the mitochondria of viable cells. In MTT assay, the intensity of the purple formazan product was measured and used to quantify viable cells in culture. A total of 100.0 μ L of cells were plated per well (1.0 × 104 cells/well), incubated overnight, and treated with MIP fractions at various concen- trations then incubated for 24 hrs. 20 μ L of MTT reagent (5.0 mg/mL) was added to each well. The plate was left on a shaker for 10 mins and incubated in the dark at 37 °C. After 1 hr of incubation, the spent medium containing excess dye was aspirated and 200 μ L of DMSO added to dissolve the purple formazan precipitates. Results were obtained using micro-titer plate reader (Tecan Sunrise, Switzerland), to detect absorbance at a test wavelength of 570 nm, and a reference wavelength of 650 nm. From absorbance data obtained, a graph was plotted employing the following equation: Viability (%) = [100% – cytotoxicity (%)]; where cytotoxicity (%) = [(absorbance value of solvent - absorbance value of MIP fraction)/absorbance value of untreated cells] × 100%. IC50 values for MIP HKB fraction were determined from the graph at 50% cell viability.

PARP cleavage assay. The occurrence of apoptosis was assessed based on the proteolytic cleavage of PARP 6 by caspase-3. Briefly, 2.0 × 10 cells/mL were treated with MIP HKB (IC50) and total proteins were extracted using the NE-PERW nuclear and cytoplasmic extraction kit according to manufacturer’s protocol. Fractionation was done using SDS-PAGE and electro-transferred onto nitrocellulose membranes. All membranes were blocked with 5% w/v BSA, 1 × TBS, 0.1% Tween-20 at room temperature with gentle shaking for 90 mins, and incubated with primary antibodies: GAPDH (1:1000) and PARP (1:1000) overnight at 4 °C, followed by detection using HRP-conjugated secondary antibodies (Cell Signaling, USA), and Super Signal West Pico chemiluminescent substrate. Images were captured using the Fusion FX7 imaging system (Vilber Lourmat, France). Apoptosis was represented by cleavage of 116 kDa full length PARP into an 89 kDa product.

Scientific Reports | 6:19833 | DOI: 10.1038/srep19833 5 www.nature.com/scientificreports/

DNA fragmentation assay. Cells were treated with 1x PBS and MIP HKB at 6, 12 and 24 hrs before har- vesting, and total DNA was extracted from both untreated and treated cells using the Suicide TrackTM DNA Ladder isolation kit according to the manufacturer’s protocol. MCF-7 cells treated with the apoptosis inducing agent, 1’-(S)-1’S-1’-acetoxychavicol acetate (ACA) served as a positive control. Extracted DNA was analyzed on a 1.5% (w/v) agarose gel electrophoresis and stained with ethidium bromide. Fragmentation of DNA was observed under UV illumination and visualized using a gel documentation system (Alpha Inotech, USA). Conclusion Currently Mycobacterium indicus pranii (MIP) has only been tested on lung and bladder cancers with tumour regression and complete recovery observed. Its effects on other cancer cell types have yet to be determined. The MIP HKB fraction was identified as the most potent cytotoxic fraction compare to LB, CS and HKS in terms of its low IC50 values and induction of apoptotic cell death in breast and oral cancer cells. Therefore the cytotoxic and apoptotic effects of MIP HKB in these two human cancer cells indicate that it can be a good candidate for further pharmacological studies to identify effective biologically active anti-cancer agents. In summary, this study has proven that MIP HKB, killed at 60 °C can inhibit the growth of various human cancer cell lines through activation of apoptosis. References 1. Nauts Coley, H. The beneficial effects of bacterial infections on host resistance to cancer. End result in 449 cases. Cancer Research Institute, Inc., New York (1980). 2. Seow, S. W. et al. Lactobacillus species is more cytotoxic to human bladder cancer cells than Mycobacterium Bovis (bacillus Calmette- Guerin). J Urol 168(5), 2236–2239 (2002). 3. Luo, X, Li, Z. & Lin, S. Anti-tumour effect of VNP20009, an attenuated Salmonella, in murine tumour models. Oncol Res, 12,501–508 (2002). 4. Morales, A., Eidinger, D. & Bruce, A. W. Intracavitary BCG in the treatment of superficial bladder tumours. J Urol 116,180–3 (1976). 5. Nauts, H., Fowler,G. & Bogatko, F. A review of the influence of bacterial infection and of bacterial products (Coley’s toxins) on malignant tumours in man. Acta Medica Scandinavica 276,1–103 (1953). 6. Zaheer, S. A. et al. Combined multidrug and Mycobacterium w vaccine therapy in patients with multi bacillary leprosy. J Infect Dis 167, 401–410 (1993). 7. Nyasulu, P. S. The role of adjunctive Mycobacterium w immunotherapy for tuberculosis. J. Exp. Clin. Med. 2, 124–129 (2010). 8. Kharkar, R. Immune recovery in HIV with Mycobacterium. J Indian Med Assoc 100,578–9 (2002). 9. Sur, P. K. & Dastidar, A. G. Role of Mycobacterium was adjuvant treatment of lung cancer (non-small cell lung cancer). J Indian Med Assoc; 101, 118–20 (2003). 10. Filion, M. C. Mycobacterium phlei cell wall complex directly induces apoptosis in human bladder cancer cells. Br J Cancer 79, 229–35 (1999). 11. Chaudhuri, P. & Mukhopadhyay, S. Bladder preserving approach for muscle invasive bladder cancer-role of Mycobacterium w. J Indian Med Assoc. 101, 559–60 (2003). 12. Ahmad, F. et al. Activation of Anti-Tumour Immune Response and Reduction of Regulatory T Cells with Mycobacterium indicus pranii (MIP) Therapy in Tumour Bearing Mice. PLoSONE 6(9), e25424 (2011). 13. Rakshit, S. et al. Immunotherapeutic efficacy of Mycobacterium indicus pranii in eliciting anti-tumour T cell responses: critical roles of IFNγ . Int J Cancer. 130, 865–875 (2011). 14. Gupta, A. et al. Efficacy of Mycobacterium indicus pranii Immunotherapy as an Adjunct to Chemotherapy for Tuberculosis and Underlying Immune Responses in the Lung. PLoSONE 7(7), e39215 (2012). 15. Pandey, R. K., Bhatt, K. H., Dahiya, Y. & Sodhi, A. Mycobacterium indicus pranii Supernatant Induces Apoptotic Cell Death in Mouse Peritoneal Macrophages in vitro. PLoSONE 6(2), e17093 (2011). 16. Escribano, J. et al. The cytotoxic effect of gluco-conjugate extracted from corms of saffron plantCrocus ( sativus) on human cell lines in culture. Planta Med. 66, 157–62 (2000). 17. Margaret, M. et al. Caspase-3 is not essential for DNA fragmentation in MCF-7 cells during apoptosis induced by the pyrrolo-1, 5-benzoxazepine, PBOX-6. FEBS Letters 515 66–70 (2002). 18. Sai, W. T. et al. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research Volume 1590, Issues 1–3, Pages 150–158 (2002). 19. Awang, K. et al. The Apoptotic Effect of 1’S-1’-Acetoxychavicol Acetate from Alpinia Conchigera on Human Cancer Cells. Molecules. 15, 8048–8059 (2010). Acknowledgements This study was supported by a grant from the Ministry of Science, Technology and Innovation (MOSTI) through the Science Fund (PROJECT NO. 02-01-03-SF0863). Author Contributions M.S. carried out all experimental assays and drafted the manuscript. All authors participated in the discussion of experimental results, revision of manuscripts and approval of final copy. N.H.N. conceived the entire study. Additional Information Competing financial interests: The authors declare no competing financial interests. How to cite this article: Subramaniam, M. et al. Cytotoxic and apoptotic effects of heat killed Mycobacterium indicus pranii (MIP) on various human cancer cell lines. Sci. Rep. 6, 19833; doi: 10.1038/srep19833 (2016). This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Scientific Reports | 6:19833 | DOI: 10.1038/srep19833 6 RESEARCH ARTICLE Mycobacterium tuberculosis Peptidyl-Prolyl Isomerases Also Exhibit Chaperone like Activity In-Vitro and In-Vivo

Saurabh Pandey1,2, Ashish Sharma3, Deeksha Tripathi3, Ashutosh Kumar3, Mohd Khubaib1,2, Manish Bhuwan1, Tapan Kumar Chaudhuri4, Seyed Ehtesham Hasnain2,3*, Nasreen Zafar Ehtesham1*

1 Inflammation Biology and Cell Signaling Laboratory, National Institute of Pathology, New Delhi, India, 2 Dr. Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Professor CR Rao Road, Hyderabad, India, 3 Molecular Infection and Functional Biology Laboratory, Kusuma School of Biological Sciences, Indian Institute of Technology, New Delhi, India, 4 Kusuma School of Biological Sciences, Indian Institute of Technology, New Delhi, India

* [email protected] (NZE); [email protected]; [email protected] (SEH)

Abstract OPEN ACCESS Peptidyl-prolyl cis-trans isomerases (Ppiases), also known as cyclophilins, are ubiquitously Citation: Pandey S, Sharma A, Tripathi D, Kumar A, Khubaib M, Bhuwan M, et al. (2016) Mycobacterium expressed enzymes that assist in protein folding by isomerization of peptide bonds preced- tuberculosis Peptidyl-Prolyl Isomerases Also Exhibit ing prolyl residues. Mycobacterium tuberculosis (M.tb) is known to possess two Ppiases, In-Vitro In-Vivo Chaperone like Activity and . PLoS PpiA and PpiB. However, our understanding about the biological significance of mycobacte- ONE 11(3): e0150288. doi:10.1371/journal. pone.0150288 rial Ppiases with respect to their pleiotropic roles in responding to stress conditions inside the macrophages is restricted. This study describes chaperone-like activity of mycobacterial Editor: Joyoti Basu, Bose Institute, INDIA Ppiases. We show that recombinant rPpiA and rPpiB can bind to non-native proteins in vitro Received: December 11, 2015 and can prevent their aggregation. Purified rPpiA and rPpiB exist in oligomeric form as evi- Accepted: February 11, 2016 dent from gel filtration chromatography.E. coli cells overexpressing PpiA and PpiB of M.tb Published: March 16, 2016 could survive thermal stress as compared to plasmid vector control. HEK293T cells tran- siently expressing M.tb PpiA and PpiB proteins show increased survival as compared to Copyright: © 2016 Pandey et al. This is an open access article distributed under the terms of the control cells in response to oxidative stress and hypoxic conditions generated after treat-

Creative Commons Attribution License, which permits ment with H2O2 and CoCl2 thereby pointing to their likely role in adaption under host gener- unrestricted use, distribution, and reproduction in any ated oxidative stress and conditions of hypoxia. The chaperone-like function of these M. medium, provided the original author and source are tuberculosis credited. cyclophilins may possibly function as a stress responder and consequently contribute to virulence. Data Availability Statement: All relevant data are within the paper and its Supporting Information files.

Funding: SP, MK, AS and MB thank the UGC/CSIR/ ICMR, Government of India (GoI) for the award of Research Fellowship. DT is a recipient of Research Associateship from Department of Biotechnology, Introduction GoI, India. This research was supported by Centre of The disease tuberculosis (TB) caused by Mycobacterium tuberculosis (M.tb), is the second larg- Excellence Grant to SEH and NZE from the est killer after HIV-AIDS [1]. For successful colonization of human host, M.tb forms a niche by Department of Biotechnology, Ministry of Science and M tb Technology, GoI. The funders had no role in study establishing molecular interaction networks within the host system. . has evolved mecha- design, data collection and analysis, decision to nisms to survive in macrophages that represent one of the most stressful environments for bac- publish, or preparation of the manuscript. teria. Successful colonization of the intraphagosomal niche by the pathogen depends on

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 1/14 M.tb Ppiases Have Chaperone Activity

Competing Interests: The authors have declared molecular interaction network within the host [2,3]. The effector molecules which play crucial that no competing interests exist. role in host pathogen interaction at the molecular level are mostly stress responders, HSPs, chaperones and other protein modifying enzymes [4]. The pathogen subverts host defenses by quenching ROS and RNS [5–7] disrupting the membrane repair [8], phagolysosomal fusion [9], suppression of autophagy [10] and by escaping immune challenges [11,12] and immune quorum sensing [13]. Bacterial chaperones play a vital role in protein folding and secretion, thereby indirectly contributing to the virulence and survival of the pathogen inside the host [14]. Prolyl isomerases, also known as cyclophilin, are expressed ubiquitously from bacteria to human and as of now 17 cyclophilin proteins in humans, 29 in Arabidopsis and 8 in Saccharo- myces have been reported so far [15,16]. Isoform diversity, various subcellular localization and differences during evolution are indicative of their functional importance and acquisition of new roles. Prolyl isomerases lie in three structurally and sequentially unrelated classes; cyclo- philins, FKBPs and Parvullins [17]. Moonlighting functions of prolyl isomerases include viru- lence character [18], stress response [19,20], cell cycle regulation [21,22], chromatin remodeling [23], transcription factor regulation [24], RNA-mediated gene expression [25,26] etc. Infection biology is not only influenced by pathogen encoded Ppiases, but host Ppiases also play a crucial role in development of the disease. For example, human CypA and Cyp B bind to the capsid protein of HIV and facilitate internalization of the virion particles in CD4 cells. Additionally, human cyclophilin A (PpiA) is recruited with nascent HIV-1 virions as well as incoming HIV-1 capsids where it is involved in isomerization of an exposed proline [27]. M.tb has two cyclophilins, PpiA and PpiB coded by ppiA (Rv0009) and ppiB (Rv2582) respectively, which are located apart in the genome. M.tb PpiA is known to be structurally and phylogenetically related to eukaryotic cyclophilins. It has been previously reported that it is a secretory protein and interacts with several host proteins such as those involved in iron regula- tion, immune defense mechanism and signal transduction [28,29]. Presence of PpiB has been reported in proteomes of membrane fraction [30] and mannosylation enriched culture filtrate [31], which are indicative of its surface expression. PpiB is reported to be essential for the sur- vival of the pathogen [32]. Functional characterization of the enzymes reflecting their possible role as a stress responder in the pathogen, and thus contributing to its virulence, has not been investigated so far. We describe the functional characterization of M.tb Ppiases (PpiA and PpiB) and demonstrate that they display chaperone-like activity. We show that recombinant M.tb Ppiases (PpiA and PpiB) expressed in E.coli could bind to heat labile MalZ protein in- vitro and can prevent its aggregation. E.coli transformants expressing M.tb Ppiases exhibited increased survival under heat shock, as compared to vector control. That these cyclophilins enabled the survival of HEK293T cells under conditions of hypoxic and oxidative stress pointed to the potential role of M.tb Ppiases in vivo to absorb cellular stress.

Materials and Methods Materials IPTG, imidazole, N-succinyl- Ala-Ala-Pro-Phe-p-nitroanilide, trifluoroethanol, LiCl, α-che- motrypsin, 8-anilino-1-naphthalene-sulfonic acid (ANS), reduced Glutathione, DTT and MTT were obtained from Sigma. All cell culture reagents were obtained from GIBCO. All enzymes were purchased from NEB (USA); ELISA kit from Peprotech and toxicity removal kit from Norgen. All reagents used were analytical grade. The plasmids and strains used in this study are listed (S1 Supporting Information).

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 2/14 M.tb Ppiases Have Chaperone Activity

Enzyme assay of purified recombinant Ppiases The ORF encoding M.tb ppiA (Rv0009) and M.tb ppiB (Rv2582) were PCR amplified from M.tb ppiA H37Rv genomic DNA by using forward and reverse primers. was cloned in pET28a vector using BamHI and HindIII restriction sites and ppiB in pGEX6p1 vector using BamH1 and Xho1 restriction sites. Recombinant proteins were purified using Ni-NTA column for PpiA and gluta- thione sepharose affinity column for PpiB as described earlier [33].Endotoxin removal was achieved by passing the recombinant protein through polymyxin B resins as described [34]. PPIase activity of both, rPpiA and rPpiB was evaluated using a spectrophotometric assay [35]. 8mM oligo peptide N-succinyl-Ala-Ala-Pro-Phe-p-nitroanilide substrate solution was prepared in trifluoroethanol containing 0.45 M LiCl in cold. Coupled enzyme α-Chymotrypsin was pre- pared at a concentration of 60mg/ml in cold solubilizing buffer (33 μl of 1 mM HCl and 2mM μ μ μ α CaCl2 solution). Reaction mixture included 910 l 0.1 M TrisCl, pH 8.0, 50 l600 M -Chymo- trypsin and 30μl rPpiA which were incubated for 2 minutes at 15°C. Subsequent addition of 10μl of peptide solution resulting in a final concentration of 80μM initiated the reaction. The enzyme catalyzed cis–trans isomerization of Ala-Pro bond, coupled with cleavage of the trans peptide by α-chymotrypsin was observed as increase in absorbance at 390nm at 15°C. Measurements were recorded every 0.5 sec till 3 minutes and the final absorbance was measured from each curve. The absorbance at each time point was subtracted from that value.

ANS Fluorescence of rPpiases Fluorescence of ANS in the presence of rPpiA and rPpiB was measured by exciting at 390 nm and following the emission between 450 and 550 nm [36]. 0.5mg/ml of the recombinant pro- teins, rPpiA and rPpiB, respectively was incubated with 50 μM ANS for 30 min at room tem- perature, and fluorescence of protein-bound dye was recorded. Fluorescence emission spectrum of ANS alone was used as control. The spectra were corrected with appropriate buffer and protein blanks. The emission and excitation slit widths were set at 10 and 10nm, respectively.

Aggregation Assay Gel filtration chromatography was carried out to know the oligomeric states of rPpiA and rPpiB in solution and the protein profile was compared with protein molecular size standards. Further to know the presence of hydrophobic surfaces which are associated with chaperon activity, 3D model of PpiB was constructed by submitting amino acid sequence to SWISS-MO- DEL[37]. Crystal structure of PpiA was downloaded from protein data bank (PDB ID:1W74). PyMOL program was used to carry out Molecular visualization and general analysis [38].Chap- erone activity of rPpiases was investigated in terms of its ability to prevent aggregation of MalZ. MalZ loses its native conformation and undergoes aggregation during incubation at 47°C. MalZ and GroEL were purified for the assay as reported earlier [39]. Assay for MalZ aggregation was performed in presence and absence of rPpiA or rPpiB at 47°C. Light scattering was measured by recording the absorbance as described earlier [40]. The samples used for the assay involved (0.4uM) MalZ alone, rPpiA alone, rPpiA with lysozyme (negative control) and purified GroEL (positive control) and with increasing molar ratios of rPpiA (10, 20, 40) and rPpiB (5, 10, 20).

Residual activity of denatured Nde1 Nde1 (10 U) was incubated at 60°C for 20 min in the absence or presence of rPpiases (rPpiA and rPpiB). BSA (20 μg) was used as a control [36] Assessment of the residual enzyme activity

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 3/14 M.tb Ppiases Have Chaperone Activity

was assayed by digesting 150ng of circular pUC18 at 37°C for 1 h. The digestion mixture was electrophoresed on 1% agarose gel, stained with ethidium bromide and visualized under UV light in a Gel doc system (Bio-Rad).

Growth rescue of E. coli from thermal shock Rescue of E. coli cells from thermal shock was performed using the method reported previously [36]. Fold survival with and without heat shock was calculated and the value was normalized taking the vector control (pET28A and pGEX6p1) as one fold.

Cloning of ppiA and ppiB for expression in HEK 293T cells Standard procedures were followed for cloning. The genes coding for ppiA and ppiB were M tuberculosis ppiA amplified using . H37Rv genomic DNA as template. For PCR amplification forward primer with XhoI site (GCCCTCTAGACTCGAGATGGCAGACTGTGATTCC) and reverse primer with HindIII site (GTTTAAACTTAAGCTTGGAGATGGTGATCGACTCGA) and similarly for ppiB forward primer with XhoI site (CCCTCTAGACTCGAGATGGGCCA CTTGACACCG) and reverse primer with HindIII site (GTTTAAACTTAAGCTTATCCA GCAGCACCGACGTGA) were used. The PCR was carried out as described [41] and PCR products and pcDNA3.1mychis (-) vector were digested and ligation reactions were set up. Recombinant plasmid constructs were transfected into the HEK293T cells (obtained from NCCS, Pune, India) by lipofection.

Survival assays under various stresses by MTT assay

It has been reported that H2O2 and CoCl2 induce oxidative stress and hypoxia, respectively in mammalian cells [42–45].Assessment of the hypoxia stress and oxidative stress on the prolifer- ation of HEK293T cells was carried out by MTT assay as described [46]. Cells were transfected with pcDNA_ppiA, pcDNA_ppiB and vector alone. 3 million cells were taken in 35 mm dish in each category. For hypoxia, 5000 cells/well from each category were seeded in a 96-well plate

cultured for 12 hours at 37°C in 5% CO2 in 150ul complete RPMI1640. Dose dependent treat- μ ment with CoCl2 (50, 100, 150, 200 M) was performed for 24 hours [47]. Untransfected cells, only media control and empty wells were used as controls. At the end of the treatment, MTT 25μl (from 5mg/ml in PBS), was added and then incubated for 4 hours. Acidic isopropanol (4%HCl and 0.1% NonidetP-40) was added in each well after removal of the supernatant. After shaking the plate for 10 min, cell viability was assessed by measuring the absorbance at 590 nm with 620 reference filter. Similarly, resistance to oxidative stress was determined by treating the – μ cells expressing PpiA and PpiB, respectively with 10 40 MH2O2 in increasing concentrations. All assays were carried out in triplicate.

Results Recombinant Ppiases are enzymatically active His tagged M.tb rPpiA and GST tagged rPpiB was purified using Ni-NTA column and glutathi- one sepharose affinity column, respectively. PpiB was less stable in the absence of GST due to its structural complexity. We tried to remove GST, but in the absence of GST tag PpiB got pre- cipitated. rPpiA displayed the expected 25kDa molecular size while rPpiB protein band was observed at 69kDA molecular size after SDS gel. Enzymatic activity of the purified proteins was measured in a spectrophotometric-coupled assay using the chromogenic peptide N-succinyl- Ala-Ala-Pro-Phe-p-nitro-anilide and α chymotrypsin at 15°C(Fig 1). Increase in the rate of

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 4/14 M.tb Ppiases Have Chaperone Activity

Fig 1. Purified rPpiases of M.tb are enzymatically active. A. Isomerization activity of rPpiA and rPpiB at a concentration of 50nM was measured in a coupled assay using the chromogenic peptide N-succinyl-Ala-Ala- Pro-Phe-p-nitroanilide and α chymotrypsin compared with the spontaneous background rate of cis-trans isomerization in the absence of the recombinant enzymes. B. Molecular weight of purified histidine-tagged rPpiA and GST tagged rPpiB as checked on 10% SDS polyacrylamide gel was around 25kDa and 69kDa, respectively. doi:10.1371/journal.pone.0150288.g001

isomerization activity as compared to the control show that the recombinant Ppiases are enzy- matically active [28].

M.tb Ppiases display chaperone like activity as evident by surface hydrophobicity ANS has been commonly used as a fluorescent probe to establish surface hydrophobicity in proteins. A blue shift of fluorescence emission maxima and increase of fluorescence intensity is generally attributed to the hydrophobicity of a binding site [48].The relative fluorescence inten- sity and maximum emission wavelength of ANS alone and ANS bound to rPpiA and rPpiB was measured by exciting at 390nm. The maximum emission wavelength of ANS alone was found to be 540nm and a clear blue shift in the emission wavelength was observed in case of ANS bound to rPpiA and rPpiB (Fig 2). A significant increase in the fluorescent intensity could be noticed when rPpiA and rPpiB was bound to ANS. These results confirmed surface hydro- phobicity in PpiA and PpiB of M.tb. which is known to be associated with chaperone like func- tion [36]. We constructed 3D model of PpiB and compared it with PpiA that confirms the presence of hydrophobic patches on the surface of these proteins (Fig 3A). The ANS fluores- cence spectra and presence of hydrophobic patches on the surface of PpiA and PpiB clearly point to the likely function of M.tb Ppiases as a chaperone.

rPpiases protect MalZ from thermal aggregation Theelutionprofileoftherecombinantproteinsconfirmedthepresenceofoligomericstatesof rPpiA and rPpiB in solution, as oligomeric rearrangement in solution is one of the characteristic of chaperon like proteins Fig 3B).To experimentally demonstrate chaperone like activity of M.tb rPpiases, their ability to prevent thermal aggregation of a heterologous protein MalZ was assessed. MalZ is known to aggregate under elevated temperature[39]. Light scattering assay to monitor

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 5/14 M.tb Ppiases Have Chaperone Activity

Fig 2. ANS Florescence spectra reveal surface hydrophobicity in M.tb rPpiases. Concentration of ANS, rPpiA and rPpiB used were 20μM, 0.1 mg/ml and 0.1mg/ml, respectively. Blue shift in the position of peak and increase in the intensity of peak was observed upon addition of rPpiA and rPpiB. The ANS emission was scanned in the range of 400 to 600 nm. doi:10.1371/journal.pone.0150288.g002

aggregation was carried out by assessing absorbance (500nm) of thermally denatured MalZ in presence and absence of rPpiA and rPpiB. The thermal stability of rPpiases was also monitored and as expected for chaperones, both the rPpiases were highly stable at 47°C, exhibiting negligible aggregation. When rPpiA was co-incubated with MalZ in increasing molar ratio (10, 20, 40), it was able to increasingly prevent aggregation at 47°C(Fig 4A). In comparison to rPpiA, rPpiB could inhibit aggregation of MalZ at almost half the concentration (Fig 4B). The use of appropri- ate positive (GroEL) and negative control (lysozyme) confirmed the specificity of the aggregation inhibition activity by the rPpiases. Molecular chaperones are known to exhibit thermal stability and can protect proteins from thermal denaturation and aggregation [36] and these result there- fore, clearly demonstrate that M.tb peptidyl prolyl isomerases can protect proteins from thermal aggregation directly pointing to their chaperone like activity.

Fig 3. PpiA and PpiB have hydrophobic patches on the surface and are present in oligomeric form in solution. A. Green color shows the hydrophobic patches on the surface of protein structures of PpiA and PpiB, blue, red and limon yellow colors show negative, positive and polar residues, respectively. B. Elution profile of Gel Filtration Chromatography: Green color indicates elution profile of rPpiA and blue indicates elution profile of rPpiB. doi:10.1371/journal.pone.0150288.g003

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 6/14 M.tb Ppiases Have Chaperone Activity

Fig 4. M.tb rPpiases suppress aggregation of MalZ. The aggregation pattern was monitored by light scattering at O.D. 500 nm with excitation and emission slit width 5 and 2.5 nm, respectively. GroEL was used as a positive control. Lysozyme was used as a negative control. A. Increasing molar ratio of rPpiA (10, 20, 40) was used. B. Increasing molar ratio of rPpiB (5,10, 20) was used. doi:10.1371/journal.pone.0150288.g004 M.tb rPpiases protect Nde1 from thermal denaturation and consequent loss of restriction enzyme activity Having shown that rPpiases could protect MalZ from thermal aggregation, we further investi- gated the ability of rPpiases to protect the enzymatic activity of Nde1 restriction enzyme from thermal denaturation. Plasmid pET22b has a single restriction enzyme site for the enzyme Nde1. Upon heat denaturation Nde1 loses its ability to linearize pET22b plasmid (Fig 5, lane 3), but when heat denatured in the presence of rPpiA (lane 4) or rPpiB (lane 5) it retained the ability to digest and could linearize the pET22B plasmid DNA. BSA when used as a control could not

Fig 5. rPpiases can rescue E.coli from thermal shock: E.coli expression strain was transformed with Ec_VCy, pGEX6p-1 only, Ec_ppiA and Ec_ppiB. After heat treatment at 50°C different dilutions of 100ul culture was plated at one hour interval. The quantification of bacterial growth was carried out by counting the bacterial colony forming unit (cfu).E. coli transformed with ppiA and ppiB exhibited approximately 10 fold more survival compared with vector control. doi:10.1371/journal.pone.0150288.g005

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 7/14 M.tb Ppiases Have Chaperone Activity

protect Nde1 and as a consequence the enzyme activity was lost after heat denaturation (lane 6). These results further demonstrate that M.tb rPpiases cannot only protect proteins from thermal aggregation but can also preserve the functional activity of proteins under in vitro conditions.

rPpiases can rescue E. coli from thermal shock In order to assay for chaperone like function of rPpiases under physiological conditions, by checking if E.coli expressing M.tb PpiA and PpiB is resistant to thermal shock as compared to vector control. E.coli transformants were assessed for their ability to grow after thermal shock (50°C) as described earlier [36]. Results indicate that E. coli cells transformed with M.tb ppiA and ppiB showed more than ten folds survival after one hour and about 80 fold survival after 2 hours as compared to the E.coli transformed with vector alone (Fig 6). These results provide conclusive evidence that rPpiases show chaperone like function both under in in-vitro condi- tions and also under physiological conditions.

M.tb Ppiases confer protection to HEK293T cells against oxidative stress and hypoxia Results presented so far demonstrated the ability of M.tb Ppiases not only to act as a chaperone, under in-vitro conditions and but also under physiological conditions. Given the fact that chaperones help in maintaining cellular homeostasis under various stress conditions [49] and their presence increases tolerance to heat, toxins, hypoxic shock and increase cell longevity by maintaining proteostasis, we designed experiments to investigate whether expression of Ppiases in HEK cells cultured in-vitro impart resistance against hypoxia and oxidative stress. For this MTT assay was performed to determine viability of HEK cells expressing PpiA and PpiB pro- teins after oxidative stress and hypoxia. HEK cells transfected with vector alone was used as a

control. For oxidative stress cells were treated with different concentration of H2O2 (0, 10, 20, 30, 40 μM) whereas for hypoxic stress the cells were treated with varying concentrations of

Fig 6. rPpiases protected Nde1 is refractile to thermal denaturation. Enzymatic activity of thermal denatured Nde1 was assayed in presence and absence of rPpiA, rPpiB and control protein BSA. Lane M, 1-kb molecular size marker; lane 1, uncut pET22b; lane 2, pET22b digested with native Nde1; lane 3, with heat denatured Nde1; lane4, with rPpiA treated heat denatured Nde1; lane 5, with rPpiB treated heat denatured Nde1; lane 6, with BSA treated heat denatured Nde1. doi:10.1371/journal.pone.0150288.g006

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 8/14 M.tb Ppiases Have Chaperone Activity

Fig 7. M.tb Ppiases impart resistance to HEK293T cells against hypoxic and oxidative stress. MTT assay was performed to score the percent cell viability of HEK293t cells transiently expressing PpiA and PpiB under oxidative stress (panel A) or hypoxia (panel B) induced by treatment of cells with varying concentrations of H2O2 (0, 10, 20, 30 and 40μM) or CoCl2 (0, 50, 100, 150 and 200μM), respectively. The data are representative of three independent experiments. Error bar represents mean ± S.D. doi:10.1371/journal.pone.0150288.g007 μ CoCl2 (0, 50, 100, 150, 200 M). A significant increase in viability of cells expressing PpiA and PpiB (Fig 7), as compared to vector control, could be seen. These observations clearly implicate M.tb Ppiases, in aiding the intracellular survival of the pathogen amid the hostile environment of infected cells pointing to its likely role in-vivo.

Discussion A multifaceted interplay between the host immune response and pathogen virulence factors govern the outcome of the infection caused by Mycobacterium tuberculosis. Peptidyl-prolyl isomerase A of M.tb, also known as cyclophilin A is secreted in intraphagosomal niche and is known to be upregulated during infection [50]. Other than ubiquitous expression, prolyl isom- erases have large number of isoforms and are localized in different subcellular locations in higher organisms, e.g., in Saccharomyces, 8 different prolyl isomerases are targeted to endoplas- mic reticulum, mitochondria and nucleus from cytosol [51]. Isoform diversity and various sub- cellular localization are indicative of their functional importance. Recently, Ppiases have been studied in different contexts such as cell regulatory processes [52], role in different cancer sta- tus [53,54] and inflammation [55] and other abnormalities [56, 57]. In addition to their vitality in protein folding, peptidyl prolyl isomerases have also been implicated in many pathological conditions like diabetes, asthma, cancer and microbial infections. surA gene coding for Ppiase is associated with virulence of pathogenic strains of Salmonella, E. coli and Helicobacter pylori. Mutation in surA gene significantly affects virulence potential of these pathogens [58,59]. In the present study, we observed that M.tb Ppiases exhibit chaperone-like activity. Surface hydrophobicity is considered important during the interaction of molecular chaperones with misfolded proteins [60,61]. ANS binding analysis of rPpiases showed increase in absorbance and blue shift in the emission maxima, a reflection of its chaperone-like activity. Unlike Mal Z, M.tb rPpiases were refractile to thermal aggregation however; when MalZ was incubated along with rPpiA and rPpiB at 45°C aggregation was inhibited. About 90% of the enzymatic activity of thermally denatured Nde1was restored when incubated with rPpiA and rPpiB. Molecular

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 9/14 M.tb Ppiases Have Chaperone Activity

chaperones can transiently bind and stabilize an unstable conformation of a protein by pre- venting its misfolding and aggregation.With in-vitro evidence of rPpiases as a probable chaper- one, we further investigated its chaperone like function in-vivo. It was clearly evident that survival of E.coli transformed with ppiA and ppiB of M.tb was quantitatively higher compared to the vector control. Similar examples of proteins that showed chaperone-like activity have been reported earlier [36, 62–64]. The ability of M.tb to grow under reduced oxygen conditions and resist oxidative stress is

directly correlated to its ability to cause disease. Although O2is essentially needed for survival of M.tb, it can easily adapt itself to the hypoxic microenvironment of tissue lesions, sites of active TB and inside microphages [65]. In the Wayne model [66], under gradual hypoxic growth envi- ronment, bacteria move away from proliferative cycle and tend towards the latent form that is adapted to hypoxia and remain viable for extended period. Turning to hypoxia tolerant status appears to be the key response mechanism for coping stress, and other survival challenges [12]. In the present study, we sought to investigate the role of M.tb Ppiases in conferring protection to HEK293T cells under hypoxia and oxidative stress. A significantdifference in the survival of HEK cells transiently expressing PpiA and PpiB proteins as compared to the untransformed cells was evident. There are several other mycobacterial HSPs which could impart survival amid hostile host effector functions such as hypoxia and oxidative stress [67]. The induction of proinflamma- tory response by the human macrophages upon infection with M.tb is a natural defense strategy mounted by the host. Infection itself induces cellular stress resulting in an unfolded protein response (UPR). Human resistin—a pro-inflammatory cytokine [68], along with other chaper- ones, is over expressed in response to UPR and aids in protein folding [36]. M.tb once lodged within the macrophages may induce similar stress resulting in misfolding of mycobacterial pro- teins, conditions that may lead to molecular crowding detrimental for the bacteria. To combat such strategies mounted by the host, M.tb deploys its Ppiases to assist in folding of the misfolded proteins. Therefore, it is very likely that these Ppiases act as a connecting link between infection, inflammation, stress response and protein misfolding, thereby assisting in the survival of the M. tbwithin the hostile environment of the macrophages. In conclusion, our data demonstrating chaperone-like function of Mycobacterium tuberculosis Ppiases have implications in virulence and pathogenicity terms of possible role inenhancing the stress tolerance of the pathogen.

Supporting Information S1 Supporting Information. Strains and plasmids used in this study. (DOCX)

Acknowledgments We would like to thank Dr. Ashok Kumar Patel for his help in protein oligomerization study. SP, MK, AS and MB thank the UGC/CSIR/ICMR, Government of India (GoI) for the award of Research Fellowship. DT is a recipient of Research Associateship from Department of Biotech- nology, GoI, India. This research was supported by Centre of Excellence Grant (BT/PR12817/ COE/34/23/2015) to SEH and NZE from the Department of Biotechnology, Ministry of Sci- ence and Technology, GoI.

Author Contributions Conceived and designed the experiments: NZE SEH. Performed the experiments: SP MK AS DT AK MB. Analyzed the data: SP DT TKC. Contributed reagents/materials/analysis tools: NZE SEH. Wrote the paper: DT SEH NZE.

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 10 / 14 M.tb Ppiases Have Chaperone Activity

References 1. WHO TB Report. Global tuberculosis report. 2015. 2. Ehrt S, Schnappinger D. Mycobacterial survival strategies in the phagosome: defence against host stresses. Cell Microbiol. 2009; 11: 1170–1178. doi: 10.1111/j.1462-5822.2009.01335.x PMID: 19438516 3. Russell DG. Mycobacterium tuberculosis: here today, and here tomorrow. Nat Rev Mol Cell Biol. Nature Publishing Group; 2001; 2: 569–77. doi: 10.1038/35085034 PMID: 11483990 4. Henderson B. Prokaryotic and Eukaryotic Heat Shock Proteins in Infectious Disease. Pockley AG, Cal- derwood SK, Santoro MG, editors. Proteins. Dordrecht: Springer Netherlands; 2010; 4: 185–209. doi: 10.1007/978-90-481-2976-8 5. Ng VH, Cox JS, Sousa AO, MacMicking JD, McKinney JD. Role of KatG catalase-peroxidase in myco- bacterial pathogenisis: Countering the phagocyte oxidative burst. Mol Microbiol. 2004; 52: 1291–1302. doi: 10.1111/j.1365-2958.2004.04078.x PMID: 15165233 6. Cirillo SLG, Subbian S, Chen B, Weisbrod TR, Jacobs WR, Cirillo JD. Protection of Mycobacterium tuberculosis from reactive oxygen species conferred by the mel2 locus impacts persistence and dis- semination. Infect Immun. 2009; 77: 2557–67. doi: 10.1128/IAI.01481-08 PMID: 19349422 7. Mukhopadhyay S, Nair S, Hasnain SE. Nitric oxide: Friendly rivalry in tuberculosis. Curr Signal Trans- duct Ther. Bentham Science Publishers; 2007; 2: 121–128. doi: 10.2174/157436207780619536 8. Divangahi M, Chen M, Gan H, Desjardins D, Hickman TT, Lee DM, et al. Mycobacterium tuberculosis evades macrophage defenses by inhibiting plasma membrane repair. Nat Immunol. 2009; 10: 899– 906. doi: 10.1038/ni.1758 PMID: 19561612 9. Shukla S, Richardson ET, Athman JJ, Shi L, Wearsch PA, McDonald D, et al. Mycobacterium tubercu- losis lipoprotein LprG binds lipoarabinomannan and determines its cell envelope localization to control phagolysosomal fusion. PLoS Pathog. 2014; 10: e1004471. doi: 10.1371/journal.ppat.1004471 PMID: 25356793 10. Huang D, Bao L. Mycobacterium tuberculosis EspB protein suppresses interferon-γ-induced autop- hagy in murine macrophages. J Microbiol Immunol Infect. 2014; doi: 10.1016/j.jmii.2014.11.008 11. Kugelberg E. Immune evasion: Mycobacteria hide from TLRs. Nat Rev Immunol. Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved.; 2014; 14: 62–3. doi: 10.1038/ nri3604 12. Gupta A, Kaul A, Tsolaki AG, Kishore U, Bhakta S. Mycobacterium tuberculosis: immune evasion, latency and reactivation. Immunobiology. 2012; 217: 363–74. doi: 10.1016/j.imbio.2011.07.008 PMID: 21813205 13. Tundup S, Mohareer K, Hasnain SE. Mycobacterium tuberculosis PE25/PPE41 protein complex induces necrosis in macrophages: Role in virulence and disease reactivation? FEBS Open Bio. 2014; 4: 822–8. doi: 10.1016/j.fob.2014.09.001 PMID: 25379378 14. Stebbins CE. Structural microbiology at the pathogen-host interface. Cell Microbiol. 2005; 7: 1227–36. doi: 10.1111/j.1462-5822.2005.00564.x PMID: 16098211 15. Wang P, Heitman J. The cyclophilins. Genome Biol. 2005; 6: 226. doi: 10.1186/gb-2005-6-7-226 PMID: 15998457 16. Galat A. Peptidylproline cis-trans-isomerases: immunophilins. Eur J Biochem. 1993; 216: 689–707. PMID: 8404888 17. Kromina K a., Ignatov a. N, Abdeeva I a. Role of peptidyl-prolyl-cis/trans-isomerases in pathologic pro- cesses. Biochem Suppl Ser. 2008; 2: 195–202. doi: 10.1134/S199074780803001X 18. Bell A, Monaghan P, Page AP. Peptidyl-prolyl cis-trans isomerases (immunophilins) and their roles in parasite biochemistry, host-parasite interaction and antiparasitic drug action. Int J Parasitol. 2006; 36: 261–76. doi: 10.1016/j.ijpara.2005.11.003 PMID: 16443228 19. Sykes K, Gething MJ, Sambrook J. Proline isomerases function during heat shock. Proc Natl Acad Sci U S A. 1993; 90: 5853–7. PMID: 7685914 20. Mark PJ, Ward BK, Kumar P, Lahooti H, Minchin RF, Ratajczak T. Human cyclophilin 40 is a heat shock protein that exhibits altered intracellular localization following heat shock. Cell Stress Chaper- ones. 2001; 6: 59–70. PMID: 11525244 21. Arevalo-Rodriguez M, Heitman J. Cyclophilin A is localized to the nucleus and controls meiosis in Sac- charomyces cerevisiae. Eukaryot Cell. 2005; 4: 17–29. doi: 10.1128/EC.4.1.17–29.2005 PMID: 15643056 22. Lu KP, Hanes SD, Hunter T. A human peptidyl-prolyl isomerase essential for regulation of mitosis. Nature. 1996; 380: 544–7. doi: 10.1038/380544a0 PMID: 8606777

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 11 / 14 M.tb Ppiases Have Chaperone Activity

23. Wu X, Wilcox CB, Devasahayam G, Hackett RL, Arévalo-Rodríguez M, Cardenas ME, et al. The Ess1 prolyl isomerase is linked to chromatin remodeling complexes and the general transcription machinery. EMBO J. 2000; 19: 3727–38. doi: 10.1093/emboj/19.14.3727 PMID: 10899126 24. Yang WM, Yao YL, Seto E. The FK506-binding protein 25 functionally associates with histone deacety- lases and with transcription factor YY1. EMBO J. 2001; 20: 4814–25. doi: 10.1093/emboj/20.17.4814 PMID: 11532945 25. Thapar R. Roles of Prolyl Isomerases in RNA-Mediated Gene Expression. Biomolecules. 2015; 5: 974–99. doi: 10.3390/biom5020974 PMID: 25992900 26. Krishnan N, Titus MA, Thapar R. The prolyl isomerase pin1 regulates mRNA levels of genes with short half-lives by targeting specific RNA binding proteins. PLoS One. Public Library of Science; 2014; 9: e85427. doi: 10.1371/journal.pone.0085427 27. Keckesova Z, Ylinen LMJ, Towers GJ. Cyclophilin A Renders Human Immunodeficiency Virus Type 1 Sensitive to Old World Monkey but Not Human TRIM5 α Antiviral Activity. J Virol. 2006; 80: 4683–4690. doi: 10.1128/JVI.80.10.4683 PMID: 16641261 28. Henriksson LM, Johansson P, Unge T, Mowbray SL. X-ray structure of peptidyl-prolyl cis-trans isomer- ase A from Mycobacterium tuberculosis. Eur J Biochem. 2004; 271: 4107–13. doi: 10.1111/j.1432- 1033.2004.04348.x PMID: 15479239 29. Bhaduri A, Misra R, Maji A, Bhetaria PJ, Mishra S, Arora G, et al. Mycobacterium tuberculosis cyclophi- lin A uses novel signal sequence for secretion and mimics eukaryotic cyclophilins for interaction with host protein repertoire. PLoS One. 2014; 9: e88090. doi: 10.1371/journal.pone.0088090 PMID: 24505389 30. Gu S, Chen J, Dobos KM, Bradbury EM, Belisle JT, Chen X. Comprehensive proteomic profiling of the membrane constituents of a Mycobacterium tuberculosis strain. Mol Cell proteomicsMCP. 2003; 2: 1284–96. doi: 10.1074/mcp.M300060-MCP200 31. Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C, Harris D, et al. Deciphering the biology of Myco- bacterium tuberculosis from the complete genome sequence. Nature. Nature Publishing Group; 1998; 393: 537–544. PMID: 9634230 32. Sassetti CM, Boyd DH, Rubin EJ. Genes required for mycobacterial growth defined by high density mutagenesis. Mol Microbiol. 2003; 48: 77–84. PMID: 12657046 33. Banerjee S, Nandyala AK, Raviprasad P, Ahmed N, Hasnain SE. Iron-dependent RNA-binding activity of Mycobacterium tuberculosis aconitase. J Bacteriol. 2007; 189: 4046–52. doi: 10.1128/JB.00026-07 PMID: 17384188 34. Banerjee S, Nandyala A, Podili R, Katoch VM, Murthy KJR, Hasnain SE. Mycobacterium tuberculosis (Mtb) isocitrate dehydrogenases show strong B cell response and distinguish vaccinated controls from TB patients. Proc Natl Acad Sci U S A. 2004; 101: 12652–7. doi: 10.1073/pnas.0404347101 PMID: 15314217 35. Fischer G, Bang H, Mech C. Determination of enzymatic catalysis for the cis-trans-isomerization of pep- tide binding in proline-containing peptides. Biomed Biochim Acta. 1984; 43: 1101–11. PMID: 6395866 36. Suragani M, Aadinarayana VD, Pinjari AB, Tanneeru K, Guruprasad L, Banerjee S, et al. Human resis- tin, a proinflammatory cytokine, shows chaperone-like activity. Proc Natl Acad Sci U S A. 2013; 110: 20467–72. doi: 10.1073/pnas.1306145110 PMID: 24282299 37. Arnold K, Bordoli L, Kopp J, Schwede T. The swiss-model workspace: A web-based environment for protein structure homology modelling. Bioinformatics. 2006; 22(2):195–201 PMID: 16301204 38. Delano WL. 2002. The PyMOL molecular graphics system. Delano Scientific, San Carlos, CA. http:// www.pymol.org/. 39. Paul S, Chaudhuri TK. Chaperone mediated solubilization of 69-kDa recombinant maltodextrin glucosi- dase in Escherichia coli. J Appl Microbiol. 2008; 104: 35–41. doi: 10.1111/j.1365-2672.2007.03519.x PMID: 18171380 40. Goyal M, Chaudhuri TK, Kuwajima K. Irreversible denaturation of maltodextrin glucosidase studied by differential scanning calorimetry, circular dichroism, and turbidity measurements. PLoS One. Public Library of Science; 2014; 9: e115877. doi: 10.1371/journal.pone.0115877 PMID: 25548918 41. Bhuwan M, Arora N, Sharma A, Khubaib M, Pandey S, Chaudhuri TK, et al. Interaction of Mycobacte- rium tuberculosis virulence factor RipA with MoxR1, a chaperone, is required for transport through TAT secretion sustem. mBio; 2016. In press. 42. Wijeratne SS, Cuppett SL, Schlegel V. Hydrogen peroxide induced oxidative stress damage and anti- oxidant enzyme response in Caco-2 human colon cells. J Agric Food Chem. 2005; 53(22):8768–74. PMID: 16248583

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 12 / 14 M.tb Ppiases Have Chaperone Activity

43. Nakamura J, Purvis ER, Swenberg JA. Micromolar concentrations of hydrogen peroxide induce oxida- tive DNA lesions more efficiently than millimolar concentrations in mammalian cells. Nucleic Acids Res. 2003; 31(6):1790–5. PMID: 12626721 44. Triantafyllou A, Liakos P, Tsakalof A, Georgatsou E, Simos G, Bonanou S. Cobalt induces hypoxia- inducible factor-1alpha (HIF-1alpha) in HeLa cells by an iron-independent, but ROS-, PI-3K- and MAPK-dependent mechanism. Free Radic Res. 2006; 40(8):847–56. PMID: 17015263 45. Piret JP, Mottet D, Raes M, Michiels C. CoCl2, a chemical inducer of hypoxia-inducible factor-1, and hypoxia reduce apoptotic cell death in hepatoma cell line HepG2. Ann N Y Acad Sci. 2002; 973:443–7. PMID: 12485908 46. Khan N, Rahim SS, Boddupalli CS, Ghousunnissa S, Padma S, Pathak N, et al. Hydrogen peroxide inhibits IL-12 p40 induction in macrophages by inhibiting c-rel translocation to the nucleus through acti- vation of calmodulin protein. Blood. American Society of Hematology; 2006; 107: 1513–20. doi: 10. 1182/blood-2005-04-1707 47. Piret J- P, Mottet D, Raes M, Michiels C. CoCl2, a chemical inducer of hypoxia-inducible factor-1, and hypoxia reduce apoptotic cell death in hepatoma cell line HepG2. Ann N Y Acad Sci. 2002; 973: 443–7. PMID: 12485908 48. Gasymov OK, Glasgow BJ. ANS fluorescence: potential to augment the identification of the external binding sites of proteins. Biochim Biophys Acta. 2007; 1774: 403–11. doi: 10.1016/j.bbapap.2007.01. 002 PMID: 17321809 49. Henderson B, Martin A. Bacterial virulence in the moonlight: Multitasking bacterial moonlighting pro- teins are virulence determinants in infectious disease. Infect Immun. 2011; 79: 3476–3491. doi: 10. 1128/IAI.00179-11 PMID: 21646455 50. Målen H, Berven FS, Fladmark KE, Wiker HG. Comprehensive analysis of exported proteins from Mycobacterium tuberculosis H37Rv. Proteomics. 2007; 7: 1702–18. doi: 10.1002/pmic.200600853 PMID: 17443846 51. Arevalo-Rodriguez M, Wu X, Hanes SD, Heitman J, others. Prolyl isomerases in yeast. Front Biosci. 2004; 9: 2420–2446. PMID: 15353296 52. Lin C-H, Li H-Y, Lee Y-C, Calkins MJ, Lee K-H, Yang C-N, et al. Landscape of Pin1 in the cell cycle. Exp Biol Med. 2015; 240: 403–8. doi: 10.1177/1535370215570829 53. Yao Q, Li M, Yang H, Chai H, Fisher W, Chen C. Roles of cyclophilins in cancers and other organ sys- tems. World J Surg. 2005; 29: 276–80. doi: 10.1007/s00268-004-7812-7 PMID: 15706440 54. Chen J, Chen S, Wang J, Zhang M, Gong Z, Wei Y, et al. Cyclophilin J is a novel peptidyl-prolyl isomer- ase and target for repressing the growth of hepatocellular carcinoma. PLoS One. 2015; 10: e0127668. doi: 10.1371/journal.pone.0127668 PMID: 26020957 55. Liu T, Schneider RA, Lee NY, Hoyt DG. Peptidylprolyl cis/trans isomerase, NIMA-interacting 1 (PIN1) regulates pulmonary effects of endotoxin and tumor necrosis factor-α in mice. Biochem Biophys Res Commun. 2014; 452: 468–72. doi: 10.1016/j.bbrc.2014.08.089 PMID: 25159840 56. Tsurubuchi T, Allender E V, Siddiqui MR, Shim K-W, Ichi S, Boshnjaku V, et al. A critical role of noggin in developing folate-nonresponsive NTD in Fkbp8 -/- embryos. Child’s Nerv Syst. 2014; 30: 1343–53. doi: 10.1007/s00381-014-2428-1 57. Restelli M, Lopardo T, Lo Iacono N, Garaffo G, Conte D, Rustighi A, et al. DLX5, FGF8 and the Pin1 isomerase control ΔNp63α protein stability during limb development: a regulatory loop at the basis of the SHFM and EEC congenital malformations. Hum Mol Genet. 2014; 23: 3830–42. doi: 10.1093/hmg/ ddu096 PMID: 24569166 58. Sydenham M, Douce G, Bowe F, Ahmed S, Chatfield S, Dougan G. Salmonella enterica serovar typhi- murium surA mutants are attenuated and effective live oral vaccines. Infect Immun. 2000; 68: 1109–15. PMID: 10678914 59. Basak C, Pathak SK, Bhattacharyya A, Pathak S, Basu J, Kundu M. The secreted peptidyl prolyl cis, trans-isomerase HP0175 of Helicobacter pylori induces apoptosis of gastric epithelial cells in a TLR4- and apoptosis signal-regulating kinase 1-dependent manner. J Immunol (Baltimore, Md1950). 2005; 174: 5672–80. 60. Das KP, Surewicz WK. Temperature-induced exposure of hydrophobic surfaces and its effect on the chaperone activity of alpha-crystallin. FEBS Lett. 1995; 369: 321–5. PMID: 7649280 61. Yang H, Huang S, Dai H, Gong Y, Zheng C, Chang Z. The Mycobacterium tuberculosis small heat shock protein Hsp16.3 exposes hydrophobic surfaces at mild conditions: conformational flexibility and molecular chaperone activity. Protein Sci. 1999; 8: 174–179. doi: 10.1110/ps.8.1.174 PMID: 10210195 62. Freeman BC, Toft DO, Morimoto RI. Molecular chaperone machines: chaperone activities of the cyclo- philin Cyp-40 and the steroid aporeceptor-associated protein p23. Science (80-). 1996; 274: 1718– 1720. doi: 10.1126/science.274.5293.1718

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 13 / 14 M.tb Ppiases Have Chaperone Activity

63. Ou WB, Luo W, Park YD, Zhou HM. Chaperone-like activity of peptidyl-prolyl cis-trans isomerase dur- ing creatine kinase refolding. Protein Sci. 2001; 10: 2346–53. doi: 10.1110/ps.23301 PMID: 11604540 64. Lilie H, Lang K, Rudolph R, Buchner J. Prolyl isomerases catalyze antibody folding in vitro. Protein Sci. 1993; 2: 1490–1496. doi: 10.1002/pro.5560020913 PMID: 8104614 65. Boon C, Dick T. Mycobacterium bovis BCG Response Regulator Essential for Hypoxic Dormancy. J Bacteriol. 2002; 184: 6760–6767. doi: 10.1128/JB.184.24.6760 PMID: 12446625 66. Wayne LG. Dormancy of Mycobacterium tuberculosis and latency of disease. Eur J Clin Microbiol Infect Dis. 1994; 13: 908–14. PMID: 7698116 67. Yuan Y, Crane DD, Simpson RM, Zhu YQ, Hickey MJ, Sherman DR, et al. The 16-kDa alpha-crystallin (Acr) protein of Mycobacterium tuberculosis is required for growth in macrophages. Proc Natl Acad Sci U S A. 1998; 95: 9578–83. PMID: 9689123 68. Silswal N, Singh AK, Aruna B, Mukhopadhyay S, Ghosh S, Ehtesham NZ. Human resistin stimulates the pro-inflammatory cytokines TNF-α and IL-12 in macrophages by NF-κB-dependent pathway. Bio- chem Biophys Res Commun. 2005; 334: 1092–1101. doi: 10.1016/j.bbrc.2005.06.202 PMID: 16039994

PLOS ONE | DOI:10.1371/journal.pone.0150288 March 16, 2016 14 / 14

201 4

International Journal of Medical Microbiology 304 (2014) 620–625

Contents lists available at ScienceDirect

International Journal of Medical Microbiology

j ournal homepage: www.elsevier.com/locate/ijmm

Short Communication

Mycobacterium avium subsp. paratuberculosis is not discerned in

diabetes mellitus patients in Hyderabad, India

a a a

Pittu Sandhya Rani , Sankara Narayana Doddam , Sashank Agrawal ,

b c a a,d,∗

Seyed E. Hasnain , Leonardo A. Sechi , Ashutosh Kumar , Niyaz Ahmed

a

Pathogen Biology Laboratory, Department of Biotechnology and Bioinformatics, University of Hyderabad, Hyderabad, India

b

Kusuma School of Biological Sciences, Indian Institute of Technology, New Delhi, India

c

Department of Biomedical Sciences, University of Sassari, Sassari, Italy

d

Institute of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: Mycobacterium avium ssp. paratuberculosis (MAP) is an obligate intracellular pathogen. It causes chronic

Received 12 November 2013

intestinal inflammation in ruminants known as Johne’s disease and is associated with human Crohn’s

Received in revised form 18 April 2014

disease. Furthermore, association of MAP with other autoimmune diseases, such as type-1 diabetes,

Accepted 21 April 2014

has been established in patients from Sardinia (Italy) which is a MAP endemic and genetically isolated

region. Due to largest livestock population and consequently high MAP prevalence amidst a very high

Keywords:

diabetes incidence in India, we sought to test this association on a limited number of patient samples from

Mycobacterium avium ssp. paratuberculosis

Hyderabad. Our results of ELISA with MAP lysate and MAP-specific protein MAP3738c as well as PCR/real-

MAP3738c

IS900 time PCR of MAP-specific sequences IS900 and/or f57 indicated that, in contrast to Sardinian diabetic

f57gene patients, MAP infection in blood is not discerned in diabetic patients in Hyderabad. The association of a

mycobacterial trigger with diabetes therefore could well be a population-specific phenomenon, highly

dependent on genetic repertoire and the environment of susceptible populations. However, a larger study

is needed in order to confirm this.

© 2014 Elsevier GmbH. All rights reserved.

Introduction inflammation in cattle and ruminants known as paratuberculosis

(Johne’s disease) (Chiodini et al., 1984). The infected cattle shed

Diabetes mellitus (DM) is a multifactorial disease that is alarm- MAP bacilli in faeces and milk. Furthermore, studies have reported

ingly increasing worldwide. In India alone, according to the that sterilization of water by chlorination, and of milk by pas-

statistics of the International Diabetes Federation, about 50 mil- teurization have minimal effects on MAP (Pierce, 2009). MAP has

lion individuals are suffering from diabetes and this may increase also been presumed to cause chronic inflammatory bowel disease

to 87 million by 2030 (Unwin et al., 2009). Of the two types of dia- in humans known as Crohn’s disease (Naser et al., 2004). It was

betes mellitus, type 1 diabetes mellitus (T1DM) is characterized reviewed/reported that PCR amplification of MAP-specific inser-

by the presence of autoantibodies whereas type 2 diabetes melli- tion sequence IS900 and detection of MAP-specific proteins such

tus (T2DM) is characterized by insulin resistance. Gut microbiota, as MAP3733c and MAP3738c conveyed association of MAP with

environmental microorganisms and dietary factors act as puta- T1DM (Rani et al., 2010; Cossu et al., 2011). However, all these

tive triggers for the development of DM in genetically susceptible studies were made with a single cohort of diabetic and control sub-

individuals (Rani et al., 2010). Mycobacterium avium ssp. paratu- jects from the Mediterranean island of Sardinia. Similar association

berculosis (MAP), an acid-fast bacillus causes chronic intestinal studies involving other world populations have not been described

so far. The 2009–2010 annual report of the department of animal

husbandry, dairying and fisheries of the Government of India, has

indicated that India has approximately 304 million milk produc-

∗

Corresponding author at: S-51, Pathogen Biology Laboratory, Department of

ing cows and buffalos, making it the world’s largest dairy herd

Biotechnology and Bioinformatics, University of Hyderabad, Gachibowli, Hyderabad

(Department of Animal Husbandry, 2011). In parallel, there have

500046, India. Tel.: +91 40 23134585; fax: +91 40 23134585.

been reports of high incidence of paratuberculosis or Johne’s dis-

E-mail addresses: [email protected], [email protected],

[email protected] (N. Ahmed). ease in Indian cattle (Singh et al., 2008). With India being the largest

http://dx.doi.org/10.1016/j.ijmm.2014.04.010

1438-4221/© 2014 Elsevier GmbH. All rights reserved.

P.S. Rani et al. / International Journal of Medical Microbiology 304 (2014) 620–625 621

producer of milk with significant prevalence of paratuberculosis

among the herds and also having highest diabetes and tuberculo-

sis incidences, it is important to identify the prevalence of MAP

in the diabetic population of India. Given this, we were interested

in identifying if MAP persists or can be detected in diabetic popu-

lations other than Sardinians, such as in India. Accordingly, using

peripheral blood samples of diabetic and non-diabetic individuals

from Hyderabad we performed a series of molecular diagnostic

assays based on ELISA with MAP cell lysate and MAP-specific pro-

tein MAP3738c used as antigens and PCR/real-time PCR analysis of Fig. 1. 12% SDS gel showing purified MAP 3738c protein; M: marker, E1: elution1

and E2: elution2.

MAP-specific insertion sequences IS900 and/or MAP-specific gene

f57. We did not find any circulating MAP DNA or MAP antibodies in

the blood of South Indian diabetic patients from Hyderabad area. Indirect ELISA of human insulin, MAP lysate and recombinant

MAP3738c protein

Materials and methods

The human insulin, MAP lysate and purified recombinant

MAP3738c protein (approximately 500 ng each in a bicarbonate

Cohort characteristics and blood samples

buffer) were coated to the ELISA plates individually and incubated

◦

overnight at 4 C. The plates were then washed with PBST (PBS with

Our human study group comprised a total of 90 individuals of

0.05% Tween 20) thrice. The uncoated sites were blocked by a block-

either sex; 30 healthy (18 males and 12 females with an average age

ing buffer [1% bovine serum albumin (BSA, Sigma) in PBS] for 2 h

of 35 years), 30 T1DM (17 male and 13 female with an average age ◦

at 37 C. The plates were then washed thrice with PBST. Human

of 29 years) and 30 T2DM (15 male and 15 female with an average

serum samples (healthy, T1DM, T2DM) were diluted 1:50 with PBS

age of 40 years). The individuals lived in a semi-urban environment

and added to the ELISA plates as appropriate and incubated for 1 h at

and had no co-morbidities or acute infections. The diabetic indi- ◦

37 C. The plates were washed with PBST thrice and incubated with

viduals were classified and identified as T1DM or T2DM based on ◦

peroxidase-conjugated anti-human IgG (Sigma; 1:15,000) at 37 C

established clinical criteria and were negative for Mycobacterium

for 1 h. The plates were then washed five times with PBST and then

tuberculosis infection by routine clinical diagnosis. Five millilitre of

with PBS. The plates were developed with o-phenylenediamine

blood was collected from each subject, and the sera obtained were

◦ (Sigma) and H2O2 in citrate buffer. The reaction was stopped with

stored at −80 C. Written informed consent was obtained from all

2N H2SO4. The intensity of the colour formed was read at 490 nm

the individuals and the study was approved by the Institutional

using an ELISA reader (infinite M200, TECAN).

Ethics Committee and Institutional Biosafety Committee of the Uni-

versity of Hyderabad. The total DNA was isolated from blood using

PCR assay for detection of 16S rDNA, IS900 and f57 gene sequences

® ◦

DNeasy Blood and Tissue kit (Qiagen) and stored at −20 C for

PCR analysis. We also tested sheep sera representing 9 healthy and

To detect bacterial DNA in the total DNA extracted from

9 MAP-infected sheep. These sheep sera were obtained from the

human blood, 16S rDNA PCR analysis was carried out using

collection of the Dept. of Biomedical Sciences, University of Sassari, - 

universal primers fD1 5 AGAGTTTGATCCTGGCTCAG-3 and rP2

Italy.  

5 -ACGGCTACCTTGTTACGACTT-3 (Weisburg et al., 1991). For

detection of MAP DNA in the total DNA extracted, PCR analy-

Preparation of MAP cell lysate and purification of recombinant sis was carried out using primers for MAP-specific fragments of

MAP 3738c protein IS900 and f57 as described previously (Vansnick et al., 2004).



The IS900 forward primer 5 -GGAGGTGGTTGTGGCACAACCTGT-

  

The MAP lysate from strain 1515 (ATCC43015) was prepared 3 , IS900 reverse primer 5 -CGATCAGCCACCAGATCGGAA-3 , f57

 

as described previously (Rani et al., 2012) and total protein con- forward primer 5 -CCTGTCTAATTCGATCACGGACTAGA-3 and f57

 

centration of the lysate was estimated using Bradford’s reagent. reverse primer 5 -TCAGCTATTGGTGTACCGAATGT-3 were used.

␮ ×

The expression plasmid encoding MAP3738c was provided by Prof. The reaction mixture consisted of 2 l 10 Dream Taq buffer

␮ ␮ ␮

Leonardo A. Sechi. For purification of recombinant MAP3738c pro- (Thermo Scientific, USA), 1 l 200 M dNTPs, 1 l 10 mM forward

␮ ␮

tein, LB broth was inoculated with an overnight grown culture of primer, 1 l 10 mM reverse primer, 5 l template DNA (approxi-

recombinant BL-21 E. coli cells transformed with the MAP3738c mately 100 ng total DNA). As a positive control, human sera samples

◦

␮

expression construct and placed in a shaking incubator at 37 C were spiked with 50 ng of MAP DNA and 0.2 l of Dream Taq poly-

until an OD600 of approximately 0.5 was obtained. The culture was merase (Thermo Scientific, USA). The final volume was made to

␮

then induced with 1 mM isopropyl-␤-d-1-thiogalactopyranoside 20 l with autoclaved milliQ water. The amplification of 16S rDNA

◦

(Sigma) and incubated at 37 C for further 3 h. The bacteria were was carried out in a thermal cycler (Eppendorf, Germany) with an

◦ ◦

then pelleted at 3200 × g for 10 min. The pellet was suspended in initial denaturation of 94 C for 5 min followed by 30 cycles of 94 C

◦ о ◦

resuspension buffer (50 mM Tris, 150 mM NaCl, 15 mM imidazole, for 30 s, 62 C for 1 min, 72 C for 2 min and final extension of 72 C

0.3% Sarcosyl and 1 mM PMSF). After sonication, the cell lysate for 10 min. The PCR amplification of both IS900 and f57 were car-

◦ о

was centrifuged at 10,000 × g for 30 min at 4 C. The supernatant ried out with an initial denaturation of 94 C for 5 min followed by

+2 ◦ ◦ ◦

was collected and loaded onto Ni -NTA beads in a column by 30 cycles of 94 C for 15 s, 58 C for 40 s, 72 C for 45 s and a final

◦ ◦

gently rocking at 4 C for 1 h. After passage of the flow-through, extension of 72 C for 10 min.

the column was washed with wash buffer (50 mM Tris, 500 mM

NaCl, 30 mM imidazole). The protein was eluted by using elution Real-time PCR of IS900 from blood samples

buffer-1 (50 mM Tris, 500 mM NaCl, 100 mM imidazole) followed

by elution buffer-2 (50 mM Tris, 500 mM NaCl, 300 mM imida- To increase the robustness of the study we performed real-

zole). The purity of the protein was checked by 12% SDS-PAGE time PCR of IS900. The MAP IS900 specific primers (IS900qPCR

◦   

−

(Fig. 1). The purified protein was stored at 20 C until further forward 5 -GATGGCCGAAGGAGATTG-3 and IS900qPCR reverse 5 -



use. CACAACCACCTCCGTAACC-3 ) were used as described previously

622 P.S. Rani et al. / International Journal of Medical Microbiology 304 (2014) 620–625

Fig. 2. ELISA of healthy and diabetes patient sera samples with MAP lysate (A), MAP3738c protein (B) and human insulin (C). (D) Indirect ELISA of healthy sheep and MAP

infected sheep sera samples with MAP3738c. The median value for each population is indicated by a horizontal line.

®

(Slana et al., 2008). The reaction mixture consisted of 5 ␮l SYBR and no difference between diabetic and non-diabetic individuals

Green PCR Master mix (Takara), 0.2 ␮l 10 mM forward primer, was found (data not shown).

0.2 ␮l 10 mM reverse primer, 1 ␮l template DNA (approximately Further, to identify whether MAP3738c could be a potent

20 ng template DNA was used for all the samples analyzed in the immunogen, we performed indirect ELISA with Johne’s disease-

assay including sera from two Sardinian T1DM patients and a sam- positive sheep sera and healthy sheep sera. We observed a

ple spiked with 10 ng MAP DNA used as a positive control) and the significantly higher immunoreactivity to MAP3738c in serum from

final volume was made to 10 ␮l with autoclaved milliQ water. The Johne’s disease-positive sheep sera than in sera from healthy sheep

4

analysis was carried out in a Mastercycler ep Real plex (Eppendorf, (Fig. 2D).

◦

Germany) with initial denaturation of 95 C for 1 min 10 s, followed

◦ ◦

by 40 cycles of 95 C for 15 s and 60 C for 30 s.

MAP DNA not detected in circulating blood cells

Statistical analysis The 16S rDNA PCR resulted in positive amplification of 1.5-kb

amplicons (Fig. 3A). The 16S rDNA amplification was observed in

Mann–Whitney’s U test (or Wilcoxon rank sum test) was 20 samples (66%) of healthy, 23 samples (76%) of T1DM and 26

used for the analysis of ELISA data. Graph Pad Prism version 5 samples (86%) of T2DM. However, all of these samples failed to

(www.graphpad.com) was used for plotting of ELISA data, calcula- amplify the IS900 and f57 gene fragments. The positive controls

tion of median values and for calculating level of significance using (representing total DNA extracted from blood samples spiked with

p value. MAP DNA) readily amplified a 452-bp fragment of IS900 and a 432-

bp fragment of the f57 gene (Fig. 3B, C). Moreover, real-time PCR

Results analysis of IS900 indicated amplification in positive control (sample

spiked with MAP DNA) and blood samples from the Sardinian T1DM

patients (2 individuals), but no such amplification of IS900 was dis-

Immunoreactivity to MAP antigens in diabetic sera

cerned in the diabetic and non-diabetic samples from Hyderabad

(Fig. 3D, E).

No differences in serum immunoreactivity to MAP cell lysate

or MAP3738c were observed between diabetic cases and healthy

controls (Fig. 2A and B). As a positive control, an ELISA was per- Discussion

formed with human insulin. Sera from diabetic individuals showed

a significantly higher immunoreactivity to insulin than sera from The association of MAP infection with T1DM has been estab-

non-diabetic controls (Fig. 2C). The negative control used was BSA, lished in the Sardinian population (Rani et al., 2010; Cossu et al.,

P.S. Rani et al. / International Journal of Medical Microbiology 304 (2014) 620–625 623

Fig. 3. (A) 16S rDNA amplification from samples corresponding to healthy (H), T1DM (T1) T2DM (T2) samples and a positive control (+ve); M = size marker. (B) IS900

amplification was observed only in the positive control. (C) f57 amplification was detected only in the positive control. (D) Real-time PCR analysis to detect MAP IS900

sequence; IS900 amplification was observed in the positive control and total DNA from T1DM patients of Sardinia but no amplification of IS900 was observed in diabetic and

non-diabetic samples from Hyderabad. (E) Melting curve of MAP IS900.

2011; Paccagnini et al., 2009). Sardinia, a Mediterranean island in healthy blood samples has previously been reported; it is estab-

has 3 million sheep with high rates of MAP infection among live- lished that healthy blood contains small amounts of microbial DNA

stock (Rani et al., 2010). Further, genetic analysis of this isolated of Riemerella, Stenotrophomonas and Pseudomonas putida as well as

population indicated a peculiar gene pool with single nucleotide that of staphylococcus and streptococcus species (McLaughlin et al.,

polymorphisms of the SLC11A1 gene conferring susceptibility to 2002; Jordana-Lluch et al., 2013; Tedeschi et al., 1976; Tedeschi and

T1DM in presence of MAP (Paccagnini et al., 2009). On the other Amici, 1972; Nikkari et al., 2001).We carried out PCR and real-time

side, India having the world’s largest livestock population and PCR analysis for the detection of MAP-specific insertion sequence

fast emerging as the ‘diabetes capital of the world’ qualifies very IS900 and/or MAP-specific gene f57 using total DNA extracted from

well to be studied for any plausible MAP link to autoimmune the blood of healthy, T1DM and T2DM patients. Approximately,

diseases. Given this background, we sought to explore the asso- around 12–18 copies of IS900 and a single copy of the f57 gene

ciation of MAP infection with T1DM in Hyderabad subjects. The exist in the MAP genome. This indicates that IS900 is an ideal tar-

ELISA assay of healthy, T1DM and T2DM sera with MAP cell lysate get for sensitive detection of MAP, whereas, f57 is a novel target

indicated a generic response to MAP lysate in all the Indian sera for specific amplification and quantification of MAP DNA in a sam-

samples. We believe it could be due to BCG vaccination or exposure ple (Bull et al., 2000; Poupart et al., 1993). The PCR and real time

to saprophytic mycobacteria (see later). To rule out this possi- PCR of IS900 and/or f57 indicated no amplification of MAP DNA

bility, we performed indirect ELISA with MAP-specific protein in all 90 Indian samples tested. These results obtained based on

MAP3738c, a 31-kDa hypothetical protein encoded by a MAP- a modest number of samples from a non-Sardinian cohort should

specific pathogenicity island (Stratmann et al., 2004). The PCR not be seen as contradictory to the other established studies link-

assays are more sensitive and constitute a simple approach to ing MAP to autoimmune diseases in general and T1DM in particular

detect slow growing bacteria such as MAP. To identify the pres- such as the studies based on Sardinian and mainland Italian popu-

ence of bacterial DNA in total DNA isolated from blood samples, lations (Cossu et al., 2011; Masala et al., 2014; Cossu et al., 2013).

we carried out 16S rDNA amplification. As indicated in the Results, Of course, association of a particular environmental or infectious

16S rDNA amplification was not equally distributed among the agent with causation of a disease or its progression can be popula-

healthy and diabetic individuals. The presence of microbial DNA tion dependent. This can be explained from the example of another

624 P.S. Rani et al. / International Journal of Medical Microbiology 304 (2014) 620–625

autoimmune disease, Crohn’s disease, for which the MAP link was References

not evident in Indians, perhaps due to different genetic background

Ahmed, N., Saini, V., Raghuvanshi, S., Khurana, J.P., Tyagi, A.K., Tyagi, A., Hasnain,

(Sasikala et al., 2009). Genetic studies have reported that criti-

S.E., 2007. Molecular analysis of a leprosy immunotherapeutic bacillus provides

cal NOD2 mutations that play a significant role in susceptibility

insights into Mycobacterium evolution. PLoS ONE 2, e968.

to Crohn’s disease were not found in Crohn’s disease patients in Booth, R.J., Williams, D.L., Moudgil, K.D., Noonan, L.C., Grandison, P.M., McKee,

J.J., Prestidge, R.L., Watson, J.D., 1993. Homologs of Mycobacterium leprae 18-

India (Pugazhendhi et al., 2008). The outcome of MAP infection and

kilodalton and Mycobacterium tuberculosis 19-kilodalton antigens in other

its association with autoimmune diseases such as T1DM has also

mycobacteria. Infect. Immun. 61, 1509–1515.

been linked to SLC11A1 polymorphisms (Paccagnini et al., 2009); Bull, T.J., Hermon-Taylor, J., Pavlik, I., El Zaatari, F., Tizard, M., 2000. Characterization

of IS900 loci in Mycobacterium avium subsp. paratuberculosis and development

this status is currently not available in the context of the Indian

of multiplex PCR typing. Microbiology 146, 2185–2197.

diabetic population, although the SLC11A1 haplotypes of two dif-

Chiodini, R., Van Kruiningen, J.H., Merkal, J.R.S., 1984. Ruminant paratuberculo-

ferent populations cannot be the same. Moreover, to eventually sis (Johne’s disease): the current status and future prospects. Cornell Vet. 74,

exclude this association in India, we need to enlarge the number 217–226.

Cossu, A., Rosu, V., Paccagnini, D., Cossu, D., Pacifico, A., Sechi, L.A., 2011. MAP3738c

of patients screened and also to target other epitopes, because

and MptD are specific tags of Mycobacterium avium subsp. paratuberculosis

the protein MAP3738c, which we used herein, is an intracellu-

infection in type I diabetes mellitus. Clin. Immunol. 141, 49–57.

lar protein and may not be fully exposed in T1DM patients. Also, Cossu, A., Ferrannini, E., Fallahi, P., Antonelli, A., Sechi, L.A., 2013. Antibodies recog-

nizing specific Mycobacterium avium subsp. Paratuberculosis MAP3738c protein

it would have been advantageous to use other immunodominant

in type 1 diabetes mellitus children are associated with serum Th1 (CXCL10)

peptides/epitopes such as MAP3865c133-141which has been used

chemokine. Cytokine 2, 337–339.

to detect MAP infection inT1DM patients at the onset (Masala et al., Department of Animal Husbandry, Dairying and Fisheries, Govt. of India,

2011. Annual Report (2010–2011), http://dahd.nic.in/dahd/WriteReadData/

2011, 2013). Our observations suggest that the Indian diabetic pop-

Annual%20Report%202010-11%20English.pdf

ulation appears to be different from Sardinians with respect to MAP

Jordana-Lluch, E., Carolan, H.E., Gimenez, M., Sampath, R., Ecker, D.J., Quesada, M.D.,

infection being an underlying factor in T1DM and this could be due Modol, J.M., Armestar, F., Blyn, L.B., Cummins, L.L., Ausina, V., Martro, E., 2013.

Rapid diagnosis of bloodstream infections with PCR followed by mass spectrom-

to many other associated parameters besides a completely differ-

etry. PLoS ONE 8, e62108.

ent genetic background. Some of the important non-genetic factors

Masala, S., Paccagnini, D., Cossu, D., Brezar, V., Pacifico, A., Ahmed, N., Mallone, R.,

in this regard include: (1) a possible historical exposure of Indians Sechi, L.A., 2011. Antibodies recognizing Mycobacterium avium paratuberculosis

epitopes cross-react with the beta-cell antigen ZnT8 in Sardinian type 1 diabetic

to mycobacteria of animal origins probably due to rural agrarian

patients. PLoS One 10, e26931.

lifestyles with high dependence on livestock/cattle, and, to sapro-

Masala, S., Zedda, M.A., Cossu, D., Ripoli, C., Palermo, M., Sechi, L.A., 2013. Zinc trans-

phytic mycobacteria as a result of poor sanitation and hygiene; porter 8 and MAP3865c homologous epitopes are recognized at T1D onset in

(2) consequently, a different type of gut/body microbiota might sardinian children. PLoS One 5, e63371.

Masala, S., Cossu, D., Piccinini, S., Rapini, N., Massimi, A., Porzio, O., Pietrosanti, S.,

have evolved that governed immunological repertoires specifying

Lidano, R., Bitti, M.L., Sechi, L.A., 2014. Recognition of zinc transporter 8 and

immune regulation and/or deregulation in Indian populations; (3)

MAP3865c homologous epitopes by new-onset type 1 diabetes children from

compulsory BCG vaccination giving cross protection from MAP as in continental Italy. Acta Diabetol., http://dx.doi.org/10.1007/s00592-014-0558-2.

McLaughlin, R.W., Vali, H., Lau, P.C.K., Palfree, R.G.E., Ciccio, A.D., Sirois, M., Ahmad,

case of leprosy (it is noteworthy that saprophytic mycobacteria as

D., Villemur, R., Desrosiers, M., Chan, E.C.S., 2002. Are there naturally occurring

well as vaccine strains and pathogens have shared genetic lineages

pleomorphic bacteria in the blood of healthy humans? J. Clin. Microbiol. 12,

(Booth et al., 1993; Yadava and Mukherjee, 1993; Ahmed et al., 4771–4775.

Naser, S., Ghobrial, A., Romero, G., Valentine, C.J.F., 2004. Culture of Mycobacterium

2007) and have common antigenic epitopes); (4) certain cultural

avium subspecies paratuberculosis from the blood of patients with Crohn’s dis-

and culinary practices such as widespread vegetarianism, deep fry-

ease. Lancet 364, 1039–1044.

ing or steam cooking of meat, and compulsory boiling of milk before Nikkari, S., Mclaughlin, I.J., Wanli, B.I., Dodge, D.E., Relman, D.A., 2001. Does blood

of healthy subjects contain bacterial ribosomal DNA? J. Clin. Microbiol. 39,

consumption in households could have historically prevented MAP

1956–1959.

infection of humans in India. However, there are no direct sup-

Paccagnini, D., Sieswerda, L., Rosu, V., Masala, S., Pacifico, A., Gazouli, M.,

portive evidences to independently confirm or negate these factors Ikonomopoulos, J., Ahmed, N., Zanetti, S., Sechi, L.A., 2009. Linking chronic

infection and autoimmune diseases: Mycobacterium avium subspecies paratu-

through mechanistic, genetic or epidemiological studies. Hence, it

berculosis, SLC11A1 polymorphisms and type-1 diabetes mellitus. PLoS One 4

would be pertinent to recommend that a larger study may be car-

(9), e7109.

ried out with an extended cohort representative of whole of India Pierce, E., 2009. Possible transmission of Mycobacterium avium subspecies paratu-

and a spectrum of regions and cultures, and the SLC11A1 and other berculosis through potable water: lesson from an urban cluster of Crohn’s

disease. Gut Pathogens 1, 17.

important polymorphism in Indian populations should be analyzed

Poupart, P., Coene, M., Van Heuverswyn, H., Cocito, C., 1993. Preparation of a spe-

to correlate genetic association with MAP infection and T1DM.

cific RNA probe for detection of Mycobacterium paratuberculosis and diagnosis

of Johne’s disease. J. Clin. Microbiol. 31, 1601–1605.

Pugazhendhi, S., Amte, A., Balamurugan, R., Subramanian, V., Ramakrishna, B.S.,

Funding 2008. Common NOD2 mutations are absent in patients with Crohn’s disease

in India. Indian J. Gastroenterol. 27, 201–203.

Rani, P.S., Sechi, L.A., Ahmed, N., 2010. Mycobacterium avium subsp.paratuberculosis

This research was partly funded through a grant from as a trigger of type-1 diabetes: destination Sardinia, or beyond? Gut Pathogens

the Department of Biotechnology (DBT) of the Indian Govern- 2, 1.

Rani, P.S., Tulsian, N.K., Sechi, L.A., Ahmed, N., 2012. In vitro cytokine profiles and

ment (No.BT/PR8040/Med/14/1224/2006). PSR received a Senior

viability of different human cells treated with whole cell lysate of Mycobacterium

Research Fellowship from the DBT. We also would like to acknowl-

avium subsp. Paratuberculosis. Gut Pathogens 4, 10.

edge partial support from UM–HIR grant Molecular Genetics (Ref. Sasikala, M., Reddy, D.N., Pratap, N., Sharma, S.K., Balkumar, P.R., Sekaran,

UM.C/625/1HIR/MOHE/CHAN-02). A., Banerjee, R., Reddy, D.B., 2009. Absence of Mycobacterium avium ss

paratuberculosis-specific IS900 sequence in intestinal biopsy tissues of Indian

patients with Crohn’s disease. Indian J. Gastroenterol. 28, 169–174.

Singh, S.V., Singh, A.V., Singh, R., Sharma, S., Shukla, N., Misra, S., Singh, P.K., Sohal,

Conflict of interests J.S., Kumar, H., Patil, P.K., Misra, P., Sandhu, K.S., 2008. Sero-prevalence of

bovine Johne’s disease in buffaloes and cattle population of North India using

indigenous ELISA kit based on native Mycobacterium avium subspecies paratu-

The authors declare that they have no competing interests.

berculosis ‘Bison type’ genotype of goat origin. Comp. Immunol. Microbiol.

Infect. Dis. 31, 419–433.

Slana, I., Kralik, P., Kralova, A., Pavlik, I., 2008. On-farm spread of Mycobacterium

Acknowledgements avium subsp. paratuberculosis in raw milk studied by IS900 and F57 competitive

real time quantitative PCR and culture examination. Int. J. Food Microbiol. 128,

250–257.

We acknowledge help from Vynika Goud, Abid Hussain and

Stratmann, J., Strommenger, B., Goethe, R., Dohmann, K., Gerlach, G.F., Stevenson, K.,

Mahabubunnisa Begum during the collection of blood samples. Li, L.L., Zhang, Q., Kapur, V., Bull, T.J., 2004. A 38-kilobase pathogenicity island

P.S. Rani et al. / International Journal of Medical Microbiology 304 (2014) 620–625 625

specific for Mycobacterium avium subsp. paratuberculosis encodes cell surface Vansnick, E., De Rijk, P., Vercammen, F., Geysen, D., Rigouts, L., Portaels, F., 2004.

proteins expressed in the host. Infect. Immun. 72, 1265–1274. Newly developed primers for the detection of Mycobacterium avium subspecies

Tedeschi, G., Amici, G.D., 1972. Mycoplasma-like microorganisms probably related paratuberculosis. Vet. Microbiol. 3, 197–204.

to L forms of bacteria in the blood of healthy persons. Cultural, morphological Weisburg, W.G., Barns, S.M., Pelletier, D.A., Lane, D.J., 1991. 16S ribosomal DNA

and histochemical data. Ann. Sclavo 14, 430–442. amplification for phylogenetic study. J. Bacteriol. 173, 697–703.

Tedeschi, G., Amici, G., Sprovieri, D., Vecchi, G.A., 1976. Staphylococcus epider- Yadava, A., Mukherjee, R., 1993. An immunodominant 30-kDa antigen of a

midis in the circulating blood of normal and thrombocytopenic human subjects: candidate anti-leprosy vaccine, Mycobacterium w, shares T and B cell deter-

immunological data. Experientia 15, 1600–1602. minants with M. leprae and M. tuberculosis. Med. Microbiol. Immunol. 182,

Unwin, N., Whiting, D., Gan, D., Jacqmain, O., Ghyoot, G. (Eds.), 2009. IDF diabetes 243–253.

atlas. , 4th ed. International Diabetes Federation, Belgium. Cytokine 68 (2014) 110–117

Contents lists available at ScienceDirect

Cytokine

journal homepage: www.journals.elsevier.com/cytokine

Helicobacter pylori protein HP0986 (TieA) interacts with mouse TNFR1 and triggers proinflammatory and proapoptotic signaling pathways in cultured macrophage cells (RAW 264.7) ⇑ Suhail A. Ansari a, Savita Devi a, Shivendra Tenguria a, Ashutosh Kumar a, Niyaz Ahmed a,b, a Pathogen Biology Laboratory, Department of Biotechnology and Bioinformatics, University of Hyderabad, Hyderabad 500046, India b Institute of Biological Sciences, University of Malaya, Kuala Lumpur 50603, Malaysia article info abstract

Article history: HP0986 protein of Helicobacter pylori has been shown to trigger induction of proinflammatory cytokines Received 17 September 2013 (IL-8 and TNF-a) through the activation of NF-jB and also to induce Fas mediated apoptosis of human Received in revised form 9 February 2014 macrophage cells (THP-1). In this study, we unravel mechanistic details of the biological effects of this Accepted 24 March 2014 protein in a murine macrophage environment. Up regulation of MCP-1 and TNF-a in HP0986-induced Available online 21 April 2014 RAW 264.7 cells occurred subsequent to the activation and translocation of NF-jB to the cell nucleus. Further, HP0986 induced apoptosis of RAW 264.7 cells through Fas activation and this was in agreement Keywords: with previous observations made with THP-1 cells. Our studies indicated activation of TNFR1 through HP0986 interaction with HP0986 and this elicited the aforementioned responses independent of TLR2, TLR4 or Murine macrophages (RAW 264.7) MCP-1 and TNF-a TNFR2. We found that mouse TNFR1 activation by HP0986 facilitates formation of a complex comprising Fas of TNFR1, TRADD and TRAF2, and this occurs upstream of NF-jB activation. Furthermore, FADD also Caspase-8 forms a second complex, at a later stage, together with TNFR1 and TRADD, resulting in caspase-8 activa- tion and thereby the apoptosis of RAW 264.7 cells. In summary, our observations reveal finer details of the functional activity of HP0986 protein in relation to its behavior in a murine macrophage cell environ- ment. These findings reconfirm the proinflammatory and apoptotic role of HP0986 signifying it to be an important trigger of innate responses. These observations form much needed baseline data entailing future in vivo studies of the functions of HP0986 in a murine model. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Genetic heterogeneity of H. pylori is reflected in its varied func- tional effects occurring due to variations in the allelic structures Helicobacter pylori infection is the known cause of infectious and functional repertoires of certain virulence factors. Among the gastritis, peptic ulcer disease and gastric cancer. Although more several known virulence factors, cytotoxin-associated gene (cag) than 50% of world’s population is infected with H. pylori, only Pathogenicity Island (cagPAI) and vacuolating cytotoxin (VacA) 10–20% of total infected cases progress to form clinically signifi- are most extensively studied [6–9]. In addition, factors contribut- cant morbidity [1]. The infection initiates with the adherence of ing towards the adherence of H. pylori to gastric epithelial cells H. pylori to the gastric epithelium with concomitant immune (SabA, BabA, AlpA/B, HopZ and OipA) also play crucial roles during responses leading to infiltration of neutrophils and lymphocytes. pathogenesis [10–13]. However, the diversity in expression of The inflammation developed during this process is often not suffi- these virulence factors in clinical strains is reflective of the exis- cient enough to clear off the infection. This leads to persistence of tence of selection pressure on the colonizing strains infecting the H. pylori infection for decades [2]. However, variations have been same or different individuals. Such variability in strain specific vir- observed in the course and severity of infection in different indi- ulence factors is also responsible for differences in virulence poten- viduals which can be attributed to bacterial strain heterogeneity, tials of H. pylori isolates [10,14,15]. Many H. pylori strain-specific host genetic susceptibility and environmental factors [3–5]. genes that encode putative virulence factors of unknown functional mechanisms are encoded within highly variable chro- ⇑ Corresponding author at: Department of Biotechnology and Bioinformatics, mosomal plasticity zones (PZs) [3]. Most PZ genes are predicted University of Hyderabad, School of Life Sciences, HCU Post, Hyderabad 500046, to be of unknown function but the encoded products of some are India. Tel.: +91 40 66794585. considered to influence the virulence potential of infecting strains. E-mail address: [email protected] (N. Ahmed). http://dx.doi.org/10.1016/j.cyto.2014.03.006 1043-4666/Ó 2014 Elsevier Ltd. All rights reserved. S.A. Ansari et al. / Cytokine 68 (2014) 110–117 111

The HP0986 protein encoded within the PZ of strain 26695 is one Forward primer: TGTGTCCGTCGTGGATCTGA and Reverse primer: such example. Our previous studies have shown it to be naturally CCTGCTTCACCACCTTCTTGA. The relative expression levels of expressed and presented to the immune system and addition of MCP-1 and TNF-a were normalized to expression values of GAPDH purified recombinant HP0986 to cultured macrophages and gastric taken as a housekeeping gene reference. epithelial cells were shown to activate NF-jB leading to up regula- tion of proinflammatory cytokines such as TNFa and IL-8 [16,17]. 2.4. Western blot analysis Also, HP0986 actively induced apoptosis via TNFR1 and Fas medi- ated pathways in THP-1 cells [16], although specific steps and RAW 264.7 cells were harvested and centrifuged at 500g for details of these pathways have remained elusive. These pathways 5 min. The whole cell extract was prepared using non-denaturing urgently required to be dissected. Further, there was a need to con- lysis buffer and according to the protocol described earlier [18]. firm HP0986 induced signaling in a murine setup in order to widen Cytoplasmic and nuclear extracts were prepared using the NE- the study towards reproduction of the effects in vivo. PER Nuclear and Cytoplasmic Extraction kit (Thermo Fisher Scien- tific, USA) according to the manufacturer’s instruction. The extracts were stored at À80 °C until use. Western blotting was preformed 2. Materials and methods as per the previous description [19]. Primary antibodies used were against p65 (NF-jB), procaspase-8, procaspase-3, b-actin (Santa 2.1. Murine macrophage (RAW 264.7) cell culture Cruz Biotechnology Inc., USA), lamin B and IjB-alpha (Sigma Aldrich, USA). Blots were developed using Super Signal West Pico RAW 264.7 cells were procured from NCCS, Pune, India. The Kit (Thermo Fisher Scientific, USA). b-actin was used as whole cells were grown in RPMI 1640 medium (HyCloneTM, Thermo cell/cytoplasmic extract marker and for nuclear extract, lamin B Fisher Scientific, USA) supplemented with 10% fetal bovine serum, was used as loading control. 2 mM glutamine and 1% anti-bacterial and anti-mycotic agent (Gibco, Life TechnologiesTM). Approximately 0.5 Â 106 cells were 2.5. Immunoprecipitation assays grown until 60–70% became confluent in 12 well culture plates. Cells were induced with increasing concentration of recombinant For interaction analysis, cell extracts were prepared with RIPA (His-tagged) HP0986 (0.5 lg, 1.0 lg, 2.5 lg and 5.0 lg/ml). Cells buffer as described before [18]. An aliquot of 0.5 mg of the cell without any stimulus and cells with proteinase-K treated HP0986 lysate was incubated overnight with HP0986 (10.0 lg) and were used as negative controls. Cells treated with LPS (5.0 lg/ml) polyclonal anti-HP0986 antibody (10.0 lg) at 4 °C. The immune from Escherichia coli 055:B5 (Sigma Aldrich, USA) served as positive complex was pulled down with protein-G agarose beads (Santa control. Culture supernatants from each treated set were collected Cruz Biotechnology Inc., USA). The complexes were run on 12% after 10 h of incubation and stored at À80 °C until used for assays. SDS–PAGE and transferred over to PVDF membrane for western blot analysis using the primary antibodies directed against TNFR1 2.2. Cytokines assay and TNFR2 (Santacruz Biotechnology Inc, USA), TLR2 and TLR4 (Imgenex, India). The immunoblots were developed with Super Cytokines/chemokines in the culture supernatant were ana- Signal West Pico Kit (Thermo Fisher Scientific, USA). Primary lyzed by multiplexed bead based immunoassays using commer- antibodies against TRAF2, TRADD and FADD (Santa Cruz Biotech- cially available Cytometry Bead Array (CBA) Kit (BD Biosciences, nology Inc., USA) were used to analyze the respective co-precipi- USA). The assay was performed according to manufacturer’s tated molecules along with the HP0986-receptor complex by instructions. The samples’ flow data were collected using a flow Immunoblotting. cytometer (BD FACS CantoII, BD Biosciences, USA) and analyzed using FCAP Array multiplex analysis software (BD Biosciences, 2.6. Flow cytometric analysis USA). Cytokine/chemokine concentrations were determined with a standard curve plotted using the intuitive FCAP array software. Cells were grown in 12 well plates till they became approxi- The values were represented as mean ± SE. mately 50–60% confluent. Cells were then treated with increasing doses of HP0986 (1.0, 2.5, 5.0 and 10.0 lg/ml) and incubated in a

2.3. Real-time PCR CO2 incubator at 37 °C for 24 h for TNFR1 expression study. Simi- larly, for the Fas expression analysis, cells were treated with increas- Approximately 2 Â 106 RAW 264.7 cells per well and per 2 ml of ing doses of HP0986 (1.0, 2.5 and 10.0 lg/ml) and for different time RPMI media were grown until 70–80% confluent in a 6 well culture periods (12 h, 24 h and 48 h). Unstimulated cells were taken as neg- plate. Cell were treated with recombinant (His-tagged) isocitrate ative control for both Fas and TNFR1 expressions whereas TNF-a dehydrogenase (ICD) (2.5 lg/ml), LPS (5.0 lg/ml) and HP0986 (0.2 lg/ml) treated cells acted as positive control. After incubation, (2.5 lg/ml). An aliquot of cells was also treated with neutralizing cells were harvested in 0.2% EDTA solution (in PBS) and counted. anti-mouse TNFR1 antibody (5.0 lg/ml) before HP0986 (2.5 lg/ Cells were centrifuged at 300g for 5 min and supernatant was com- ml) induction. The cells were incubated for 6 h. After incubation, pletely removed. Cells were then resuspended in appropriate vol- cells from each well were harvested in 1 ml TRI reagent (Sigma ume of blocking buffer (5% horse serum in PBS) to get 1 Â 106 Aldrich, USA). Total RNA was extracted as per the manufacturer’s cells per ml of buffer. The resuspended cells were kept on ice for instruction. One microgram of total RNA was taken for generation 30–45 min to ensure the blocking of the Fc receptor sites present of cDNA using Verso cDNA Synthesis Kit (Thermo Fisher Scientific, on macrophage cells and therefore to enhance the specific binding USA) according to the protocol described in their manual. The mRNA of the primary antibodies to their respective antigenic sites on cells levels of MCP-1 and TNF-a were measured with the help of SYBR surface. Approximately 0.1 Â 106 cells (100 ll) were taken in a poly- green (Applied Biosystems, Life TechnologiesTM) in a MastercyclerÒ styrene tube and the subsequent staining steps were performed in ep realplex thermal cycler (Eppendorf, Germany). Following primers the same buffer. For TNFR1 expression analysis, cells were incu- were used for the amplifications: Mouse TNF-a Forward primer: bated with rabbit anti-TNFR1 antibody (Santacruz Biotechnology TCCCAGGTTCTCTTCAAGGGA, Reverse primer: GGTGAGGAGCACG- Inc. USA) for 45 min on ice, followed by incubation with secondary TAGTCGG; Mouse MCP-1 Forward primer: GAAGGAATGGGTCCAGACAT, FITC-conjugated anti-rabbit antibody (Sigma Aldrich, USA) for Reverse primer: ACGGGTCAACTTCACATTCA and Mouse GAPDH another 45 min. For the Fas expression assay, approximately 112 S.A. Ansari et al. / Cytokine 68 (2014) 110–117

0.1 Â 106 (100 ll) cells were incubated directly with FITC-conju- at the protein concentration of 2.5 lg/ml (Fig. 1A, B). Cells treated gated anti-CD95 antibody (BD Biosciences, USA) for 30 min on ice. with even higher dose (5.0 lg/ml) of HP0986 did not show any fur- After the incubation with labeled antibodies, the cells were washed ther increase in MCP-1and TNF-a levels. Proteinase-K-treated twice and resuspended in 500 ll of the same buffer for immediate HP0986 failed to induce the production of both MCP-1 and TNF- data acquisition on flow cytometer. Unstained cells were taken as a suggesting that the response was indeed triggered by HP0986. internal negative control in both the assays to rule out any interfer- Cells treated with LPS (5.0 lg/ml) were taken as a positive control ence due to background fluorescence. All acquired data were for the induction of both MCP-1 and TNF-a (Fig. 1A, B). In order to analyzed with Flow Jo software (Tree Star Inc., USA). further confirm the effect specific to HP0986, His-tagged isocitrate dehydrogenase (ICD) of H. pylori purified in a similar manner was 2.7. MTT-assay for cell viability measurement used as an internal negative control. A relative mRNA quantifica- tion analysis using real-time PCR showed that cells treated with The RAW 264.7 cells’ viability after HP0986 treatment was ana- His-tagged ICD protein (2.5 lg/ml) failed to trigger MCP-1 and lyzed by MTT (3-(4, 5-Dimethylthiazol-2-yl)-2, 5-diphenyltetrazo- TNF-a mRNA expression above the control level and therefore rul- lium bromide (Amresco, USA) assay. The HP0986 induced cells ing out the possible involvement of 6x His tag of recombinant, were incubated for 48 h in a 96 well plate. An aliquot of 20 llof purified HP0986 in the above responses (Fig. 1C, D). Thus, our data MTT (5 mg/ml) was added to each well and kept for 3 h at 37 °C. reveal HP0986 specific proinflammatory cytokine responses in The resultant insoluble Formazan complex was dissolved in RAW 264.7 cells. dimethyl sulfoxide (Sigma Aldrich, USA). The colored solution thus formed was measured at 570 nm [20]. 3.2. MCP-1 and TNF-a secretion is mediated by activation of NF-jB

2.8. HP0986 mediated apoptosis analysis In order to determine whether the MCP-1 and TNF-a secretion is regulated by NF-jB activation, we analyzed the NF-jB levels in HP0986 induced apoptosis in RAW 264.7 cells was analyzed by nuclear and cytoplasmic fractions of HP0986 stimulated RAW AnnexinV-FITC/PI assay. Cells were processed and staining was 264.7 cells. We observed an increase in nuclear translocation of performed according to manufacturer’s instructions (BD Biosci- p65 in HP0986 treated RAW 264.7 cells in a time dependent man- ences, USA) and the sample data were acquired on flow cytometer ner. A clear and enhanced translocation of p65 in the nuclear (BD FACS Canto II, BD Biosciences, USA). The percentage of cells extract of HP0986 treated (2.5 lg/ml) RAW 264.7 cells was + undergoing apoptosis were expressed as total of % AnnexinV observed with increase in time of incubation (0.5, 1.0, 2.0 and + and PI cells after subtracting the background fluorescence. Fur- 3.0 h). The time course followed for this study indicated that max- thermore, apoptosis in HP0986 treated RAW 264.7 cells could be imum level of nuclear localization of p65 was observed at 3 h post visualized through DNA fragmentation assay. The analysis was per- treatment when compared to the control (Fig. 2B). The correspond- formed as per the method described previously [21]. In brief, ing observed decrease in levels of p65 from cytoplasmic extracts 6 2 Â 10 cells were washed with PBS and pelleted by centrifugation during the same time period was also consistent with increased at 400g for 5 min. Cell pellets were then treated with a lysis buffer translocation to the nucleus. TNF-a (50 ng/ml) was taken as posi- (1% NP-40 in 20 mM EDTA, 50 mM Tris–HCl, pH, 7.5) by repeated tive control for NF-jB translocation analysis. In a further study, pipetting up and down. Homogenates were then centrifuged at western blot analysis showed noticeable degradation of IjB-alpha 13,000g for 5 min. Supernatants were then treated with RNase A in the cytosol after stimulation with HP0986 (2.5 lg/ml). A visibly at 37 °C for 2 h followed by proteinase-K at 37 °C for 2 h. Then a clear degradation of IjB-alpha was observed at 30 min, 1 h and 2 h ½ volume of 10 M ammonium acetate was added to DNA solutions. post treatment with HP0986 (Fig. 2A). Taken together, we found To further facilitate the DNA precipitation, 2.5 volume of ethanol that time course for IjB-alpha degradation preceded NF-jB was added and centrifuged. DNA pellets were dissolved in gel load- activation. ing buffer, run and analyzed on 1% agarose gel. 3.3. Interaction of HP0986 with murine TNFR1 2.9. Statistical analysis Host receptor involved in eliciting the cytokine response was Results were expressed as means ± standard error (SE). Compar- analyzed by pull down assay which showed mouse TNFR1 cap- isons of cytokine inductions, TNFR1 and Fas expressions were per- tured together with HP0986 when probed with anti-TNFR1 anti- formed by two-tailed t-test using GraphPad Prism version 5.01 for body. However, HP0986 did not involve in any complex windows (Graph Pad software, San Diego, California USA: formation with TNFR2/TLR2 or TLR4 as revealed by pull-down www.graphpad.com). The differences between two values were assay using anti-HP0986 antibody followed by western blot analy- considered significant if p < 0.05. sis with anti-TNFR2, anti-TLR2 and anti-TLR4 antibodies (to rule out any possible HP0986 binding to receptors other than TNFR1) 3. Results (Fig. 3A). Furthermore, an increase in cell surface TNFR1 expression was observed by flow cytometry in HP0986 treated RAW 264.7 3.1. HP0986 induces expression and secretion of MCP-1 and TNF-a cells in a dose (1.0, 2.5, 5.0 and 10.0 lg/ml) dependent manner from murine macrophages (Fig. 3B). These observations confirmed the interaction between HP0986 and murine-TNFR1, an event upstream of inflammatory The N-terminal His-tagged protein was over expressed in BL- response, and also ruled out any interaction between HP0986 21(DE3) cells and purified under native conditions using Co+2- and TNFR2/TLR2 or TLR4. Moreover, RAW 264.7 cells pretreated NTA affinity column. RAW 264.7 cells treated with HP0986 and with neutralizing anti-mouse TNFR1 antibody (5 lg/ml) prior to incubated for 10 h released MCP-1 and TNF-a into the culture induction by HP0986 (2.5 lg/ml) revealed reduced translocation supernatant in a dose dependent manner. A significant secretion of p65 to the nucleus (Fig. 3C) and also decreased induction of of MCP-1 and TNF-a was observed when 1 ml aliquot of media MCP-1 and TNF-a (Fig. 1C and D). These observations clearly sug- with about 1 million cells was treated with increasing doses of gest a link between TNFR1 stimulation through HP0986 binding HP0986 (0.5 lg, 1.0 lg, 2.5 lg and 5.0 lg/ml). Highest secretion and NF-jB activation leading to the MCP-1 and TNF-a secretion of both MCP-1 (p = 0.0002) and TNF-a (p < 0.0001) were observed in murine macrophages. S.A. Ansari et al. / Cytokine 68 (2014) 110–117 113

Fig. 1. (A) Secretion of MCP-1 and (B) TNF-a by murine macrophage (RAW 264.7) cells after treatment with HP0986. Uninduced cells, and cells stimulated with proteinase-K treated HP0986 were taken as negative control. LPS (5 lg/ml) treated cells were taken as positive control. Results are shown as mean ± SE. (C) Estimation of relative mRNA levels for MCP-1 and (D) TNF-a by real time PCR in HP0986 treated RAW 264.7 cells. Unstimulated cells, His-tagged ICD treated cells and the cells pretreated with neutralizing anti-mouse TNFR1 before HP0986 induction also served as negative controls for this study. Results are shown as mean ± SE.

Fig. 2. (A) IjB-alpha degradation analysis in HP0986 treated murine macrophage (RAW 264.7) cells at different time intervals by western blotting. Reprobing of the blot with b-actin antibody was done to ensure equal loading of the protein. (B) Time dependent translocation analysis of p65 (NF-jB) in cytoplasmic and nuclear fractions from RAW 264.7 cells treated with HP0986. TNF-a treated cells were taken as positive control for this study. b-actin was used as a marker for equal loading of cytoplasmic fractions. Lamin B was used as a marker entailing nuclear fraction.

3.4. HP0986 induces Fas mediated apoptosis in murine macrophages shown to induce increased expression of Fas on murine (RAW 264.7) macrophages (RAW 264.7) in a dose (1.0, 2.5 and 10 lg/ml) and time (12, 24 and 48 h) dependent manner. However, at highest Fas receptor on the cell surface induces apoptosis upon interac- time point of 48 h incubation, the increment in protein doses did tion with the ligand [22]. According to our previous observation not reflect increase in Fas expression possibly due to apoptosis with human macrophages (THP-1) that suggested a link between of the cells (Fig. 4). However, expression of Fas on murine TNFR1 activation by HP0986 and Fas expression [16], we per- macrophages was significantly higher than unstimulated cells. formed a similar analysis with murine macrophages. HP0986 were AnnexinV-FITC/PI assay also indicated an increase in apoptosis of 114 S.A. Ansari et al. / Cytokine 68 (2014) 110–117

Fig. 3. (A) Pull down assays for binding analysis of HP0986 with receptors (TNFR1, TNFR2, TLR2 or TLR4). HP0986 bound receptor from the mouse macrophage cell lysate was pulled down with protein G-agarose beads and immunoblots were probed with anti-TNFR1, anti-TLR2, anti-TLR4 and anti-TNFR2 antibodies. Cell lysates incubated with protein-G beads and protein-G beads alone were used as negative controls for this assay. (B) Flow cytometry analysis showing increased effect of HP0986 on TNFR1 expression on murine macrophage (RAW 264.7) cells. Uninduced cells were taken as negative control while TNF-a treated cells were analyzed as positive control for TNFR1 expression. (C) Western blot analysis showing reduced translocation of p65 (NF-jB) in RAW 264.7 cells pretreated with neutralizing anti-TNFR1 antibody before induction with HP0986 when compared to cells treated with HP0986 alone. TNF-a treated cells served as positive control and unstimulated cells as negative control for this analysis.

HP0986 (5.0 lg/ml) treated RAW 264.7 cells. In the same analysis, from the RAW 264.7 cells treated with HP0986 and incubated for upon HP0986 (5.0 lg/ml) stimulation, we observed decreased 30 min (Fig. 6). This suggests that TRADD forms a complex with apoptosis of cells pretreated with neutralizing anti-mouse TNFR1 TNFR1 after interacting with HP0986 which in turn provides a antibody (5.0 lg/ml). His-tagged ICD (5.0 lg/ml) failed to induce platform for the recruitment of TRAF2 [25] leading to the NF-jB apoptosis and therefore ruling out the possible interference by activation. Further, we also observed in a similar study, the 6 Â His-tag attached to recombinant HP0986. Staurosporin co-precipitation of FADD along with TNFR1-HP0986 complex in (50 nM) treated cells in all cases were taken as positive control the lysate from cells treated with HP0986 for 24 h (Fig. 6). However for apoptosis (Fig. 5A). Further, we observed an increase in procas- since FADD interacts with both TRADD and Fas, this indicates two pase-8 and procaspase-3 degradation in RAW 264.7 cells induced possible pathways for HP0986 induced apoptosis: one mediated with HP0986 for 48 h in a dose (2.5, 5.0 and 10.0 lg/ml) dependent through TNFR1 and the other involving Fas activation [25,26]. manner when compared with uninduced cells (Fig. 5B). As a posi- tive control, Staurosporin (50 nM) also showed an enhanced degra- dation of pro-caspase-8 and procaspase-3. b-actin was taken as 4. Discussion loading control for the above analysis. This observation is also in agreement with earlier reports suggesting that Fas engagement HP0986 or TieA is an importantly proinflammatory protein and with its ligand results in accumulation, cleavage and therefore acti- putative type II endonuclease encoded by the plasticity region of H. vation of caspase-8 which further activates the effector caspases pylori [16,17]. The protein has been profiled either (genotypically) [23,24]. Furthermore, apoptosis process in RAW 264.7 cells was using patient biopsies and clinical isolates corresponding to differ- validated through DNA fragmentation in HP0986 (5.0 lg/ml) trea- ent populations or through (serological) testing for humoral ted cells (DNA ladder assay) as compared to unstimulated cells immune responses and in (cell signaling) studies involving cul- (Fig. 5C). tured human cells, and was also shown to induce proinflammatory cytokines through TNFR1 and NF-jB mediated signaling pathways 3.5. TNFR1 activation by HP0986 and recruitment of TRADD, TRAF2 [16,17]. However, its function has not been verified in an animal and FADD model such as mouse. The present study therefore assumes an important extension of the works of Alvi et al. [16] and Devi Both TRADD and TRAF2 were co-precipitated together with et al. [17] to reproduce the interaction of this protein with mouse TNFR1-HP0986 complex as shown in immunoblots using lysate immune apparatus (including TNFR1 and other receptors) and S.A. Ansari et al. / Cytokine 68 (2014) 110–117 115

Fig. 4. Fas expression on the surface of murine macrophage cells (RAW 264.7) induced by HP0986 in a dose and time dependent fashion and analyzed using FITC-conjugated anti-Fas antibody by flow cytometry. Cells without protein stimulus and/or unstained uninduced cells served as negative controls and TNF-a treated cells as positive control. Results are also shown as mean ± S.E in bar diagrams. characterize the downstream signaling events to unravel finer (Fig. 2). NF-jB activation is a key determinant of the immune cell details of the apoptotic pathways in order to understand the com- response to the bacterial infection which in turn regulates expres- plete scenario of the proposed ‘survival advantage’ to the bacte- sion of many inflammation promoting genes encoding several che- rium [16,17] in a mammalian system. mokines, cytokines and members of the IgG superfamily [38–42]. In that pursuit, the strongest observation made in this study Fas expression by RAW 264.7 cells after HP0986 induction entails consistent proinflammatory and apoptotic behavior of (Fig. 4) was in accordance with our previous observations in HP0986 in a murine macrophage cell (RAW 264.7), similar to what THP-1 cells [16]. This also suggests that the virulence potential has been observed earlier for the human cell lines and PBMCs. and activity of HP0986 is similarly placed in the mouse system. Induction of proinflammatory cytokines by the mouse cells con- In corroboration with previous reports, we also showed by immu- firms HP0986 to be an important antigen that excites mammalian noprecipitation method the interaction between HP0986 and mur- cells. As a potent chemoattractant for monocytes/macrophages and ine TNFR1 (Fig. 3A). Furthermore, RAW 264.7 cells pretreated with T-lymphocytes, MCP-1 is reported to be actively involved in the neutralizing antibody against TNFR1 revealed reduced transloca- inflammation process [27–30]. TNF-a secreted by macrophages tion of NF-jB to the nucleus upon induction with HP0986; this could have several implications during bacterial infection. TNF-a confirmed the dependency of the expression of inflammatory binding to TNFR1 triggers Fas mediated apoptosis of macrophages response genes on TNFR1 (Fig. 3C). This has thus indicated that [31–33]. Reports also indicate that TNF-a down-regulates phago- HP0986 interaction with TNFR1 leads to a downstream inflamma- cytosis of apoptotic cells by macrophages [34]. Recently, it has tory response as well as Fas mediated apoptosis. been shown in case of H. pylori infection that TNF-a accumulation To gain further insight into the signaling events, we have shown during gastric injury has negative effect on the clearance of apop- through immunoprecipitation analysis the interaction of HP0986 to totic gastric epithelial cells [35]. The apoptotic ‘cell loads’ further TNFR1 receptor and activation of downstream signaling molecules enhance inflammatory state of gastric mucosa through release of including TRADD, TRAF2 and FADD (Fig. 6). Given these results, and proinflammatory mediators from the dying cells following their on the lines of previous observations [43–46,26,16], we propose a secondary necrosis [35]. signaling cascade similar to that mediated by TNF-a. In brief, The present results reconfirmed involvement of NF-jB in induc- stimulation of TNFR1 by HP0986 triggers the recruitment of tion of MCP-1 and TNF-a which are consistent with the previous the adaptor molecule TRADD to the cytoplasmic domain of TNFR1. reports [36,37]. The activation of NF-jB as observed through trans- Further, TRAF2 binds to the N-terminal domain of TRADD leading location of p65 to the nucleus after HP0986 induction was also to the activation of NF-jB. TNFR1 activation also contributes supported by the degradation of the IjB-alpha in RAW 264.7 cells to another pathway through the recruitment of FADD to the 116 S.A. Ansari et al. / Cytokine 68 (2014) 110–117

Fig. 5. (A) AnnexinV-FITC/PI staining to analyze apoptosis in murine macrophages cells by flow cytometry using (i) HP0986 induced cells, and (ii) cells treated with neutralizing antiTNFR1 antibody and HP0986; controls comprised of (iii) cells without protein stimulus (negative control), (iv) cells treated with recombinant His-tagged ICD protein (unrelated protein control) and (v) cells treated with Staurosporin (positive control). (B) Immunoblotting showing procaspase-8 and procaspase-3 cleavage analysis as a marker for apoptosis in murine macrophage cells in a dose dependent manner. Cells without any stimulus were taken as negative control. Staurosporin treated cells were used as positive control. (C) DNA ladder assay to assess the proapoptotic nature of HP0986 in RAW 264.7 cells. Cells without protein stimulation were taken as negative control. DNA isolated from Staurosporin induced cells was taken as positive control for apoptosis.

of various possible cellular pathways that conjoin TNFR1, Fas and NF-jB. These new data unleashing in a mouse macrophage back- ground would definitely underpin future efforts directed at under- standing the complexity and nature of H. pylori induced cancer signaling. The immunological co-ordinates so established would also facilitate in vivo validation of HP0986 function in a mouse sys- tem in order to study the secretion, localization and regulation of this important virulence factor.

Acknowledgements

We would like to thank P. Sandhya Rani for inputs during man- uscript editing. We acknowledge funding support from the Depart- ment of Biotechnology, Government of India (BT/PR6921/MED/29/ 699/2013). SAA would like to acknowledge University Grants Com- mission, India for the award of Senior Research Fellowship. SD would like to acknowledge partial support from the HIR project of the University of Malaya (UM.C/625/1/HIR/MOHE/CHAN-02 – Fig. 6. Co-precipitation of TRADD, TRAF2 and FADD recruited to activated TNFR1 Molecular Genetics). NA is an Academy Professor of the Academy bound with HP0986. These adaptor molecules were pulled down together with HP0986-TNFR1 complex and detected by western blotting. Protein-G beads of Scientific and Innovative Research, India and Visiting Professor incubated with cell lysate, and protein-G beads alone were used as negative at the Institute of Biological Sciences, University of Malaya, Kuala controls for this analysis. Lumpur.

References C-terminal domain of TRADD and subsequent activation of caspase- 8 leading to apoptosis. However, since FADD also connects Fas to [1] Suerbaum S, Michetti P. Helicobacter pylori infection. N Engl J Med 2002;347:1175–86. caspase-8 activation [26], there could be a possibility of involve- [2] Day AS, Jones NL, Policova Z, Jennings HA, Yau EK, Shannon P, et al. ment of an additional death pathway (other than that mediated Characterization of virulence factors of mouse-adapted Helicobacter pylori through TNFR1) in HP0986 treated mouse macrophages. strain SS1 and effects on gastric hydrophobicity. Dig Dis Sci 2001;46:1943–51. [3] Sugimoto M, Watada M, Jung SW, Graham DY, Yamaoka Y. Role of Helicobacter In view of our observations, it is now possible to clearly portray pylori plasticity region genes in development of gastroduodenal diseases. J Clin the role of HP0986 in inflammation and apoptosis at the interface Microbiol 2012;50:441–8. S.A. Ansari et al. / Cytokine 68 (2014) 110–117 117

[4] Ding SZ, Zheng PY. Helicobacter pylori infection induced gastric cancer; advance [27] Villiger PM, Terkeltaub R, Lotz M. Production of monocyte chemoattractant in gastric stem cell research and the remaining challenges. Gut Pathog protein-1 by inflamed synovial tissue and cultured synoviocytes. J Immunol 2012;4:18. 1992;149:722–7. [5] Borlace GN, Jones HF, Keep SJ, Butler RN, Brooks DA. Helicobacter pylori [28] Sierra B, Perez AB, Vogt K, Garcia G, Schmolke K, Aguirre E, et al. MCP-1 and phagosome maturation in primary human macrophages. Gut Pathog 2011;3:3. MIP-1a expression ina model resembling early immune response to dengue. [6] Montecucco C, Rappuoli R. Living dangerously: how Helicobacter pylori Cytokine 2010;52:175–83. survives in the human stomach. Nat Rev Mol Cell Biol 2001;2:457–66. [29] Deng YY, Lu J, Ling EA, Kaur C. Monocyte chemoattractant protein-1 (MCP-1) [7] Chattopadhyay S, Patra R, Chatterjee R, De R, Alam J, Ramamurthy T, et al. produced via NF kappa B signaling pathway mediates migration of amoeboid Distinct repeat motifs at the C-terminal region of CagA of Helicobacter pylori microglia in the periventricular white matter in hypoxic neonatal rats. Glia strains isolated from diseased patients and asymptomatic individuals in West 2009;57:604–21. Bengal, India. Gut Pathog 2012;4:4. [30] Koch AE, Kunkel SL, Harlow LA, Johnson B, Evanoff HL, Haines GK, et al. [8] Peek Jr RM, Blaser MJ. Helicobacter pylori and gastrointestinal tract Enhanced production of monocyte chemoattractant protein-1 in rheumatoid adenocarcinoma. Nat Rev Cancer 2002;2:28–37. arthritis. J Clin Invest 1992;90:772–9. [9] Covacci A, Telford JL, Del Giudice G, Parsonnet J, Rappuoli R. Helicobacter pylori [31] Boyle JJ, Weissberg PL, Bennet MR. Tumor necrosis factor-alpha promotes virulence and genetic geography. Science 1999;284:1328–33. macrophage-induced vascular smooth muscle cell apoptosis by direct and [10] Alm RA, Ling LS, Moir DT, King BL, Brown ED, Doig PC, et al. Genomic sequence autocrine mechanisms. Arterioscler Thromb Vasc Biol 2003;23:1553–8. comparison of two unrelated isolates of the human gastric pathogen [32] Hagimoto N, Kuwano K, Kawasaki M, Yoshimi M, Kaneko Y, Kunitake R, et al. Helicobacter pylori. Nature 1999;397:176–80. Induction of interleukin-8 secretion and apoptosis in bronchiolar epithelial [11] Oh JD, Kling-Bäckhed H, Giannakis M, Xu J, Fulton RS, Fulton LA, et al. The cells by Fas ligation. Am J Respir Cell Mol Biol 1999;21:436–45. complete genome sequence of a chronic atrophic gastritis Helicobacter pylori [33] Lee SY, Cherla RP, Tesh VL. Simultaneous induction of apoptotic and survival strain: evolution during disease progression. Proc Natl Acad Sci USA signaling pathways in macrophage like THP-1 cells by Shigla toxin1. Infect 2006;103:9999–10004. Immun 2001;75:1291–302. [12] Yamaoka Y, Ojo O, Fujimoto S, Odenbreit S, Haas R, Gutierrez O, et al. [34] McPhillips K, Janssen WJ, Ghosh M, Byrne A, Gardai S, Remigio L, et al. TNF- Helicobacter pylori outer membrane proteins and gastroduodenal disease. Gut alpha inhibits macrophage clearance of apoptotic cells via cytosolic 2006;55:775–81. phospholipase A2 and oxidant-dependent mechanisms. J Immunol [13] Fujimoto S, Olaniyi Ojo O, Arnqvist A, Wu JY, Odenbreit S, Haas R, et al. 2007;178:8117–26. Helicobacter pylori BabA expression, gastric mucosal injury, and clinical [35] Bimczok D, Smythies LE, Waites KB, Grams JM, Stahl RD, Mannon PJ, et al. outcome. Clin Gastroenterol Hepatol 2007;5:49–58. Helicobacter pylori infection inhibits phagocyte clearance of apoptotic gastric [14] Delgado-Rosado G, Dominguez-Bello MG, Massey SE. Positive selection of a epithelial cells. J Immun 2013;190:6626–34. bacterial oncoprotein associated with gastric cancer. Gut Pathog 2011;3:18. [36] Deng YY, Lu J, Ling EA, Kaur C. Monocyte chemoattractant protein-1 (MCP-1) [15] Alm RA, Trust TJ. Analysis of the genetic diversity of Helicobacter pylori: the tale produced via NF-jB signaling pathway mediates migration of amoeboid of two genomes. J Mol Med 1999;77:834–46. microglia in the periventricular white matter in hypoxic neonatal rats. Glia [16] Alvi A, Ansari SA, Ehtesham NZ, Rizwan M, Devi S, Sechi LA, et al. Concurrent 2009;57:604–21. proinflammatory and apoptotic activity of Helicobacter pylori protein (HP986) [37] Comstock KL, Krown KA, Page MT, Martin D, Ho P, Pedraza M, et al. LPS- points to its role in chronic persistence. PloS One 2011;6:e22530. induced TNF-alpha release from and apoptosis in rat cardiomyocytes: [17] Devi S, Ansari SA, Vadivelu J, Mégraud F, Tenguria S, Ahmed N. Helicobacter obligatory role for CD14 in mediating the LPS response. J Mol Cell Cardiol pylori antigen HP0986 (Tie A) interacts with cultured gastric epithelial cells 1998;30:2761–75. and induces IL-8 secretion via NF-jB mediated pathway. Helicobacter [38] Beg AA, Baldwin Jr AS. The I kappa B proteins: multifunctional regulators of 2014;19:26–36. Rel/NF-kappa B transcription factors. Genes Dev 1993;7:2064–70. [18] Bonifacino JS, Dell’Angelica EC, Springer TA. Immunoprecipitation. Curr [39] Lenardo MJ, Baltimore D. NF-kappa B: a pleiotropic mediator of inducible and Protocols Protein Sci 2001;18:9.8.1–9.8.28. tissue-specific gene control. Cell 1989;58:227–9. [19] Weih F, Carrasco D, Bravo R. Constitutive and inducible Rel/NF-jB activities in [40] Thanos D, Maniatis T. NF-kappa B: a lesson in family values. Cell mouse thymus and spleen. Oncogene 1994;9:3289–97. 1995;80:529–32. [20] Mosmann T. A rapid calorimetric assay for cellular growth and survival: [41] Rizwan M, Alvi A, Ahmed N. Novel protein antigen (JHP940) from the genomic application to proliferation and cytotoxicity assays. J Immunol 1983;65:55–63. plasticity region of Helicobacter pylori induces TNF-alpha and Interleukin- 8 [21] Herrmann M, Lorenz HM, Voll R, Grünke M, Woith W, Kalden JR. A rapid and secretion by human macrophages. J Bacteriol 2008;190:1146–51. simple method for the isolation of apoptotic DNA fragments. Nucl Acids Res [42] Kumar A, Lewin A, Rani PS, Qureshi IA, Devi S, Majid M, et al. Dormancy 1994;22:5506–7. Associated Translation Inhibitor (DATIN/Rv0079) of Mycobacterium [22] Tanaka M, Suda T, Takahashi T, Nagata S. Expression of the functional soluble tuberculosis interacts with TLR2 and induces proinflammatory cytokine form of human Fas ligand in activated lymphocytes. EMBO J 1995;14:1129–35. expression. Cytokine 2013;64:258–64. [23] Medema JP, Scaffidi C, Kischkel FC, Shevchenko A, Mann M, Krammer PH, et al. [43] Chen G, Goeddel DV. TNF-R1 signaling: a beautiful pathway. Science FLICE is activated by association with the CD95 death-inducing signaling 2002;2002(296):1634–5. complex (DISC). EMBO J 1997;16:2794–804. [44] Postal M, Appenzeller S. The role of tumor necrosis factor-alpha (TNF-a) [24] Stennicke HR, Jürgensmeier JM, Shin H, Deveraux Q, Wolf BB, Yang X, et al. Pro- in the pathogenesis of systemic lupus erythematosus. Cytokine 2011;56: caspase-3 is a major physiologic target of caspase-8. J Bio Chem 537–43. 1998;273:27084–90. [45] Wajant H, Pfizenmaier K, Scheurich P. Tumor necrosis factor signaling. Cell [25] Hsu H, Shu HB, Pan MG, Goeddel DV. TRADD-TRAF2 and TRADD-FADD Death Differ 2003;10:45–65. interactions define two distinct TNF receptor 1 signal transduction pathways. [46] Ea CK, Deng L, Xia ZP, Pineda G, Chen ZJ. Activation of IKK by TNFalpha requires Cell 1996;84:299–308. site-specific ubiquitination of RIP1 and polyubiquitin binding by NEMO. Mol [26] Micheau O, Solary E, Hammann A, Dimanche-Boitrel MT. Fas ligand- Cell 2006;22:245–57. independent, FADD-mediated activation of the Fas death pathway by anticancer drugs. J Bio Chem 1999;274:7987–92. Genomes of Two Clinical Isolates of Mycobacterium tuberculosis from Odisha, India

Mohammad Majid,a,d Narender Kumar,a Asifa Qureshi,d Priyadarshini Yerra,a Ashutosh Kumar,a Mandala Kiran Kumar,a Suma Tiruvayipati,a,c Ramani Baddam,a Sabiha Shaik,a Aparna Srikantam,b Niyaz Ahmeda,c Pathogen Biology Laboratory, Department of Biotechnology and Bioinformatics, University of Hyderabad, Hyderabad, Indiaa; Blue Peter Research Centre (Lepra Society), Cherlapally, Hyderabad, Indiab; Institute of Biological Sciences, Faculty of Science, University of Malaya, Kuala Lumpur, Malaysiac; Environmental Genomics Division, CSIR– National Environmental Engineering Research Institute (NEERI), Nagpur, Indiad

We report whole-genome sequences of two clinical isolates of Mycobacterium tuberculosis isolated from patients in Odisha, In- dia. The sequence analysis revealed that these isolates are of an ancestral type and might represent some of the “pristine” isolates in India that have not admixed with other lineages.

Received 20 February 2014 Accepted 27 February 2014 Published 20 March 2014 Citation Majid M, Kumar N, Qureshi A, Yerra P, Kumar A, Kumar MK, Tiruvayipati S, Baddam R, Shaik S, Srikantam A, Ahmed N. 2014. Genomes of two clinical isolates of Mycobacterium tuberculosis from Odisha, India. Genome Announc. 2(2):e00199-14. doi:10.1128/genomeA.00199-14. Copyright © 2014 Majid et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 Unported license. Address correspondence to Narender Kumar, [email protected], or Niyaz Ahmed, [email protected].

uberculosis caused by Mycobacterium tuberculosis is a chronic strains were high, as usually observed for the M. tuberculosis com- Tinfectious disease that is often fatal if not effectively treated. plex, 65.31% (NA-A0008) and 65.28% (NA-A0009). The two ge- Every year, 8 to 9 million new infections and a death toll of 1.5 nomes, NA-A0008 and NA-A0009, were predicted to encode million are recorded worldwide. It is estimated that about one- 4,400 and 4,453 CDSs with average lengths of 866 and 857 bp, third of the human population is infected with M. tuberculosis (1). respectively. Both of them contained a single rRNA operon and 45 Comparative genomic studies have provided deeper insights into tRNA genes. the genetic diversity and clonal architecture of M. tuberculosis (2). The availability of these genome sequences would definitely Recent studies conducted on isolates from India have shown that complement the gene pool analysis of the Indian strains from highly concentrated reservoirs of the ancestral M. tuberculosis lin- different parts of the country. Besides this, comparative genomic eages prevail in South and Central India (3–6). Only a limited analysis and phylogenetic study of these isolates with other M. tu- number of M. tuberculosis genomes from India are sequenced. The berculosis strains might give us important insights into the biology whole-genome analysis of ancestral and modern lineages would and molecular epidemiology of this organism. facilitate deciphering of the genetic variability and evolutionary Nucleotide sequence accession numbers. The M. tuberculosis mechanisms of this obligate parasite. We describe the whole- NA-A0008 and NA-A0009 whole-genome shotgun projects have genome sequences of two M. tuberculosis strains, NA-A0008 and been deposited in the GenBank database under the accession NA-A0009, isolated in 2008 from patients in rural Odisha, India. numbers ALYG00000000 and ALYH00000000, respectively. The Genomic DNAs of both strains were isolated using the Qiagen BioProject designations for these projects are PRJNA168604 and kit method. Whole-genome sequencing was carried out on an Ion PRJNA168605, respectively. Torrent sequencing platform (Life Technologies). The process generated 3 million and 2.9 million reads amounting to 89ϫ and ACKNOWLEDGMENTS 93ϫ genome coverage for NA-A0008 and NA-A0009, respec- We acknowledge the DFG program GRK1673 and financial support un- tively, with a mean read length of 250 bp. The reads after filtration der the DBT-COE grant (BT/01/COE/07/02) from the Department of were assembled into 280 and 310 contigs for NA-A0008 and NA- Biotechnology, Government of India. Mohammad Majid is grateful to A0009, respectively, using the MIRA v.2 de novo assembler. These Vijay N. Charde, M. Uma, and Mumtaz Baig for their help and support. contigs were ordered and reoriented according to the M. tubercu- losis CCDC 5180 genome using in-house written scripts. The re- REFERENCES sulting draft genomes were annotated using the RAST annotation 1. Frieden TR, Sterling TR, Munsiff SS, Watt CJ, Dye C. 2003. Tubercu- server (7), and CDSs were validated by comparing outputs from losis. Lancet 362:887–899. http://dx.doi.org/10.1016/S0140-6736(03)143 33-4. EasyGene (8) and Glimmer (9), as done previously (10–13). The 2. Ilina EN, Shitikov EA, Ikryannikova LN, Alekseev DG, Kamashev DE, number of rRNA operons were predicted in both strains using Malakhova MV, Parfenova TV, Afanas’ev MV, Ischenko DS, Bazaleev RNAmmer (14), while tRNAscan-SE (15) was used to identify NA, Smirnova TG, Larionova EE, Chernousova LN, Beletsky AV, tRNA sequences. Artemis (16) was used to glean the genome sta- Mardanov AV, Ravin NV, Skryabin KG, Govorun VM. 2013. Compar- ative genomic analysis of Mycobacterium tuberculosis drug resistant strains tistics of both the strains. The genome sizes of NA-A0008 and from Russia. PLoS One 8:e56577. http://dx.doi.org/10.1371/journal.pone NA-A0009 were 4,259,206 and 4,271,739 bp, with coding percent- .0056577. ages of 89.4% and 89.3%, respectively. The GϩC contents of both 3. Narayanan S, Gagneux S, Hari L, Tsolaki AG, Rajasekhar S, Narayanan

March/April 2014 Volume 2 Issue 2 e00199-14 Genome Announcements genomea.asm.org 1 Majid et al.

PR, Small PM, Holmes S, Deriemer K. 2008. Genomic interrogation of Ahmed N. 2011. Genome of multidrug-resistant uropathogenic Esche- ancestral Mycobacterium tuberculosis from south India. Infect. Genet. richia coli strain NA114 from India. J. Bacteriol. 193:4272–4273. http://dx Evol. 8:474–483. http://dx.doi.org/10.1016/j.meegid.2007.09.007. .doi.org/10.1128/JB.05413-11. 4. Rao KR, Kauser F, Srinivas S, Zanetti S, Sechi LA, Ahmed N, Hasnain 11. Avasthi TS, Devi SH, Taylor TD, Kumar N, Baddam R, Kondo S, SE. 2005. Analysis of genomic downsizing on the basis of region-of- Suzuki Y, Lamouliatte H, Mégraud F, Ahmed N. 2011. Genomes of two difference polymorphism profiling of Mycobacterium tuberculosis patient chronological isolates (Helicobacter pylori 2017 and 2018) of the West isolates reveals geographic partitioning. J. Clin. Microbiol. 43:5978–5982. African Helicobacter pylori strain 908 obtained from a single patient. J. http://dx.doi.org/10.1128/JCM.43.12.5978-5982.2005. Bacteriol. 193:3385–3386. http://dx.doi.org/10.1128/JB.05006-11. 5. Ahmed N, Saini V, Raghuvanshi S, Khurana JP, Tyagi AK, Tyagi AK, 12. Baddam R, Thong KL, Avasthi TS, Shaik S, Yap KP, Teh CS, Chai LC, Hasnain SE. 2007. Molecular analysis of a leprosy immunotherapeutic Kumar N, Ahmed N. 2012. Whole-genome sequences and comparative bacillus provides insights into Mycobacterium evolution. PLoS One genomics of Salmonella enterica serovar. Typhi isolates from patients with 2:e968. http://dx.doi.org/10.1371/journal.pone.0000968. fatal and nonfatal typhoid fever in Papua New Guinea. J. Bacteriol. 194: 6. Thomas SK, Iravatham CC, Moni BH, Kumar A, Archana BV, Majid M, 5122–5123. http://dx.doi.org/10.1128/JB.01051-12. Priyadarshini Y, Rani PS, Valluri V, Hasnain SE, Ahmed N. 2011. 13. Kumar N, Mukhopadhyay AK, Patra R, De R, Baddam R, Shaik S, Alam Modern and ancestral genotypes of Mycobacterium tuberculosis from J, Tiruvayipati S, Ahmed N. 2012. Next-generation sequencing and de Andhra Pradesh, India. PLoS One 6:e27584. http://dx.doi.org/10.1371 novo assembly, genome organization, and comparative genomic analyses /journal.pone.0027584. 7. Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, of the genomes of two Helicobacter pylori isolates from duodenal ulcer Formsma K, Gerdes S, Glass EM, Kubal M, Meyer F, Olsen GJ, Olson patients in India. J. Bacteriol. 194:5963–5964. http://dx.doi.org/10.1128/J R, Osterman AL, Overbeek RA, McNeil LK, Paarmann D, Paczian T, B.01371-12. Parrello B, Pusch GD, Reich C, Stevens R, Vassieva O, Vonstein V, 14. Lagesen K, Hallin P, Rødland EA, Staerfeldt HH, Rognes T, Ussery Wilke A, Zagnitko O. 2008. The RAST server: rapid annotations using DW. 2007. RNAmmer: consistent and rapid annotation of ribosomal subsystems technology. BMC Genomics 9:75. http://dx.doi.org/10.1186/ RNA genes. Nucleic Acids Res. 35:3100–3108. http://dx.doi.org/10.1093 1471-2164-9-75. /nar/gkm160. 8. Larsen TS, Krogh A. 2003. EasyGene—a prokaryotic gene finder that 15. Lowe TM, Eddy SR. 1997. tRNAscan-SE: a program for improved detec- ranks ORFs by statistical significance. BMC Bioinformatics 4:21. http://dx tion of transfer RNA genes in genomic sequence. Nucleic Acids Res. 25: .doi.org/10.1186/1471-2105-4-21. 955–964. http://dx.doi.org/10.1093/nar/25.5.0955. 9. Delcher AL, Harmon D, Kasif S, White O, Salzberg SL. 1999. Improved 16. Rutherford K, Parkhill J, Crook J, Horsnell T, Rice P, Rajandream microbial gene identification with GLIMMER. Nucleic Acids Res. 27: MA, Barrell B. 2000. Artemis: sequence visualization and annotation. 4636–4641. http://dx.doi.org/10.1093/nar/27.23.4636. Bioinformatics 16:944–945. http://dx.doi.org/10.1093/bioinformatics 10. Avasthi TS, Kumar N, Baddam R, Hussain A, Nandanwar N, Jadhav S, /16.10.944.

2 genomea.asm.org Genome Announcements March/April 2014 Volume 2 Issue 2 e00199-14

201 3

Cytokine 64 (2013) 258–264

Contents lists available at SciVerse ScienceDirect

Cytokine

journal homepage: www.journals.elsevier.com/cytokine

Dormancy Associated Translation Inhibitor (DATIN/Rv0079) of Mycobacterium tuberculosis interacts with TLR2 and induces proinflammatory cytokine expression

Ashutosh Kumar a, Astrid Lewin b, Pittu Sandhya Rani a, Insaf A. Qureshi c, Savita Devi a, ⇑ Mohammad Majid a, Elisabeth Kamal b, Stefanie Marek b, Seyed E. Hasnain d, Niyaz Ahmed a,e, a Pathogen Biology Laboratory, Department of Biotechnology and Bioinformatics, University of Hyderabad, Hyderabad, India b Division 16 Mycotic and Parasitic Agents and Mycobacteria, Robert Koch Institute, Nordufer 20, 13353 Berlin, Germany c Department of Biotechnology and Bioinformatics, University of Hyderabad, Hyderabad, India d Kusuma School of Biological Sciences, Indian Institute of Technology Delhi, New Delhi, India e Institute of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia article info abstract

Article history: Mycobacterium tuberculosis, the cause of tuberculosis in humans, is present approximately in one third of Received 23 March 2013 the world’s population, mostly in a dormant state. The proteins encoded by the dormancy survival reg- Received in revised form 28 May 2013 ulon (DosR regulon) are mainly responsible for survival of the bacilli in a latent form. To maintain latency, Accepted 3 June 2013 mycobacteria orchestrate a balanced interplay of different cytokines secreted by immune cells during the Available online 29 June 2013 granulomatous stage. The function of most of the DosR regulon proteins of M. tuberculosis is unknown. In this study, we have shown that one of the DosR regulon proteins, DATIN, encoded by the gene Rv0079, can Keywords: stimulate macrophages and peripheral blood mononuclear cells (PBMC) to secrete important cytokines Mycobacterium tuberculosis that may be significant in granuloma formation and its maintenance. The expression level of DATIN in Dormancy antigens DosR regulon proteins Mycobacterium bovis BCG was found to be upregulated in pH stress and microaerobic conditions. Compu- Proinflammatory cytokines tational modeling, docking and simulation study suggested that DATIN might interact with TLR2. This TLR2 was further confirmed through the interaction of recombinant DATIN with TLR2 expressed by HEK293 cells. When in vitro differentiated THP-1 cells were treated with recombinant DATIN, increased secretion of TNF-a, IL-1b and IL-8 was observed in a dose dependent manner. When differentiated THP-1 cells were infected with a modified BCG strain that overexpressed DATIN, augmented secretions of TNF-a, IL-1b and IL-8 were observed as compared to a reference BCG strain containing empty vector. Similarly, human PBMCs when infected with M. bovis BCG that overexpressed DATIN, upregulated secretion of proinflam- matory cytokines IFN-c, TNF-a, IL-1b and IL-8. The cytokine profiles dissected herein point to a possible role of DATIN in maintenance of latency with the help of the proinflammatory responses. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction granuloma, under latency. The expression of these genes is con- trolled by a regulated gene cluster consisting of three genes Tuberculosis is one of the deadly diseases that result in heavy (Rv3134c/3133c/3132c), among them, DosR (Rv3133c) functions morbidity and mortality. M. tuberculosis persists in a sizeable por- as a cognate response regulator [1–8]. It has been shown that the tion of the world’s population, probably in a latent form. It infects cellular composition of granuloma is in a dynamic state with re- alveolar macrophages and multiplies therein. During latency, these spect to oxygen tension [9]. M. tuberculosis should be able to switch infected macrophages are arranged in an ordered structure in lungs its metabolic activity from aerobic respiration to nitrate respiration and form the granuloma that signifies a microenvironment charac- in time with change in the oxygen concentration within the gran- terized by low oxygen and high nitric oxide. A set of 48 genes of uloma. DosR regulon plays an important role in recovery from DosR regulon is up-regulated under stress conditions and presum- anaerobiosis [10] and therefore it is essential for survival. It has ably plays important roles in survival of mycobacteria within the been shown that DosR regulon antigens induce stronger humoral immune response in individuals with latent infection when com- pared to acutely infected individuals [11]. This suggests stronger ⇑ Corresponding author. Address: Dept. of Biotechnology and Bioinformatics, University of Hyderabad, School of Life Sciences, HCU Post, Hyderabad 500046, expression of DosR regulon genes during latency. Apart from this, India. Tel./fax: +91 40 66794585. not many studies are carried out to identify the functions and E-mail address: [email protected] (N. Ahmed). interactions of the proteins encoded by DosR regulated genes.

1043-4666/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.cyto.2013.06.310 A. Kumar et al. / Cytokine 64 (2013) 258–264 259

During infection, M. tuberculosis is recognized via TLRs and gets 2.3. Differentiated macrophages and human blood cells phagocytized by alveolar macrophages. TLR2 in combination with TLR1/TLR6 and TLR4 [12,13] together recognize mycobacterial The human macrophage cell line THP-1 (DSMZ GmbH, Braun- components [14,15]. It has been shown that inflammatory re- schweig, Germany) and recombinant HEK293 cell line (HEK- sponses are necessary for granuloma formation and long term sur- Blue™-hTLR2, InvivoGen, San Diego, US) were used for treatment vival of M. tuberculosis; TNF-a, IFN-c, IL-1b, IL-12, reactive oxygen with DATIN to measure cytokines in the culture supernatants (ROIs) and nitrogen intermediates (RNIs) are important for the and TLR2 activation. THP-1 and PBMCs were used for infection maintenance of the granuloma structure [16,17]. In addition, the with BCG. The THP-1 was maintained in RPMI 1640 (GIBCOÒ Invit- chemokine IL-8 attracts circulating leukocytes to infiltrate in- rogen, Darmstadt, Germany) supplemented with 10% fetal bovine flamed tissues [18]. serum (BioWhittaker, Walkersville, MD, USA) by passaging twice DATIN encoded by ORF Rv0079 is one of the strongly up-regu- weekly. lated genes of the DosR regulon [8]. It has been demonstrated that Aliquots of human blood were obtained from German Red Cross DATIN may significantly act as regulator of dormancy functions and PBMCs were isolated by Ficoll-Paque™ Plus (GE Healthcare, such as translation control and stabilization of translational appa- USA) gradient centrifugation according to the manufacturer’s rec- ratus of mycobacteria [19,20]. DATIN was found to be up-regulated ommendations and as described in Sharbati et al. [26]. Cells were under different in vitro and in vivo induced stress conditions such resuspended in ACK lysis buffer (0.15 M NH4Cl, 1.0 mM KHCO3 as hypoxia [21] and treatment with nitric oxide [7]. It was also and 0.1 mM Na2EDTA, pH 7.3) to remove remaining erythrocytes shown to be recognized by PBMCs from M. tuberculosis-infected from the cell pellets and then incubated for 5 min on ice. PBS individuals, as well as by most of the M. tuberculosis specific T-cell was added to the lysis buffer, the cells were washed with PBS twice lines which were generated against lysate or culture filtrate of M. and then resuspended in IMDM medium with L-Glutamine (PAA, tuberculosis grown under low oxygen conditions or standard aer- Pasching, Austria) and 3% human AB serum (PAA, Pasching, ated condition. These observations support the contention that DA- Austria). TIN could express during infection and under stress conditions HEK-Blue™-hTLR2 cells were grown in HEK-Blue™ selection [11]. In this study, we show the up-regulation of DATIN under medium that contained several selective antibiotics to ensure per- pH stress and microaerobic conditions and we demonstrate the sistent expression of various transgenes introduced in HEK-Blue™- interaction of DATIN with TLR2 leading to the secretion of proin- hTLR2 cells. Normocin was added to protect HEK-Blue™-hTLR2 flammatory cytokines such as TNF-a, IL-1b and IL-8 by the cultured cells from any possible microbial contamination. The cells were macrophages and IFN-c, TNF-a, IL-1b and IL-8 by human PBMCs. passaged (when 75–85% confluent) using PBS by gentle pipetting. These results, as presented, might be significant in understanding Growth media and detection media for engineered HEK293 cell the role of dormancy related proteins in the maintenance of latent line were prepared according to the manufacturer’s (InvivoGen, state of tuberculosis. San Diego, USA) protocol. All cell lines and human PBMCs were

maintained at 37 °C with 5% CO2. 2. Materials and methods 2.4. Cytokine analysis of culture supernatants of THP-1 cells and PBMCs 2.1. Recombinant strain generation and purification of DATIN In 24 well plates (TPP, Trasadingen, Switzerland), approxi- The detailed methodology of the construction of recombinant mately 0.2 million THP-1 cells per well were differentiated into strains [E. coli BL21 (pRSETA + Rv0079) and BCG (pMV261 + adherent macrophage-like phenotype with an overnight treatment Rv0079)] and purification of recombinant DATIN has been de- of Phorbol 12-myristate 13-acetate (PMA) at a concentration of scribed previously [19,22–24]. 10 ng/ml. The cells were washed twice with RPMI1640. THP-1 cells were then either treated with different concentrations of DATIN 2.2. qRT-PCR to quantify the relative expression of DATIN (1.0 lg, 2.5 lg and 5.0 lg) or infected at a multiplicity of infection (MOI) of 50 with BCG (pMV261) or BCG (pMV261 + Rv0079) for M. bovis BCG (pMV261) and M. bovis BCG (pMV261 + Rv0079) 24 h. THP-1 cells treated with 100 U/ml of human interferon-c were grown up to cell density (OD600) of 2.0 in Middlebrook broth (hu-INF-c) (Invitrogen, Darmstadt, Germany) and 10 ng/ml of LPS with 10% OADC containing 25 lg/ml kanamycin at 37 °C. They were used as positive control. Untreated cells were used as nega- were then kept for 1 week at 37 °C in Middlebrook 7H9 broth of tive control. One million isolated PBMCs were seeded in 24 well pH 5.3 supplemented with 10% OADC containing 25 lg/ml kana- plates and after 24 h, cells were infected by a MOI of 2 with BCG mycin for pH stress or they were kept in microaerobic condition (pMV261) or BCG (pMV261 + Rv0079). Uninfected cells were used for oxygen stress. Microaerobic condition (6.2–13.2% oxygen) as negative control while cells used as positive control were trea- was achieved using the GENbox system from BioMérieux. RNA ted with10 ng/ml of LPS. All culture plates were incubated at and cDNA were prepared from the BCG cultures grown in stress- 37 °C with 5% CO2. Culture supernatants were collected after 24 h free condition (neutral pH and atmospheric oxygen concentration), and were stored at 80 °C until assayed. The cytokine levels of pH stress and microaerobic condition according to the protocol de- IFN-c, TNF-a, IL-1b and IL-8 were measured with ELISA Kits scribed by Khattak et al. [25]. Primers used for amplifications were (Ready-SET-Go) from eBioscience (Natutec, Frankfurt, Germany) (i) for gene Rv0079:50-GTA TTT CCC CCA GGA TGG TCA-30 and 50- by following the manufacturer’s instructions. GTC GTA GCG GGA GTA GAG CAG-30 and (ii) for 16S rRNA gene: 50-GAC CCC GTG AAG TCG GAG TCG-30 and 50-TTC ATG ACG TGA 2.5. DATIN–TLR-2 docking CGG GCG GT-30. Relative transcription level of DATIN in pH stress and microaerobic conditions was calculated in comparison to Prediction of protein–protein interaction site was done using stress-free condition by the DDCT method. Relative expression of MetaPPISP [27] while protein–protein docking was performed with DATIN in the overexpressing strain [BCG (pMV261 + Rv0079)] the web version of PatchDock [28] as described earlier [29]. The was also calculated in comparison to the BCG strain containing crystal structure of TLR1-TLR2 heterodimer (PDB ID: 2Z82) the empty vector [BCG (pMV261)]. Expression of 16S rRNA gene retrieved from Protein Data Bank (receptor) was used for docking was used as reference gene for normalization. study with DATIN model (ligand) under default complex-type 260 A. Kumar et al. / Cytokine 64 (2013) 258–264 settings. Molecular visualization and general analysis were done stress conditions in comparison to stress-free conditions using using the program PyMOL [30]. HBOND program was used to iden- DDCT method and the expression of the 16S rDNA for normaliza- tify hydrogen bonds at the molecular interface [31]. tion. As compared to stress-free growth conditions, Rv0079 was ex- pressed approximately 22-fold in pH stress and approximately 2.5- 2.6. DATIN–TLR2 interaction assay fold in microaerobic conditions, respectively (Fig. 2).

In order to identify the interaction of DATIN with TLR2, a cell- 3.2. DATIN induces proinflammatory cytokines based colorimetric assay was performed with engineered HEK293 cells containing multiple genes from the TLR2 pathway (that in- Recombinant purified DATIN induced proinflammatory re- clude TLR2 and genes participating in the signaling cascade). sponse in PMA differentiated THP-1 cells. A significant increase HEK-Blue™-hTLR2 cells also express an optimized alkaline phos- in secretion of TNF-a, IL-1b and IL-8 was observed as compared phatase gene engineered to be secreted (sAP) and controlled by to untreated cells (Fig. 3). Further, cultured THP-1 cells were in- several transcription factors such NF-jB and activator protein-1. fected with BCG strains. The amount of TNF-a, IL-1b and IL-8 se- To measure the interaction of DATIN with TLR2, different concen- creted by the THP-1 varied as a function of the type of BCG trations of DATIN (8 ng, 40 ng, 0.2 lg, and 1 lg) per aliquot were strain used [BCG (pMV261) or BCG (p MV261 + Rv0079)]. The over- used to treat approximately 50,000 HEK-Blue™-hTLR2 cells. Trea- expressing strain induced higher secretion of TNF-a, IL-1b and IL-8 ted cells were incubated at 37 °CinaCO2 incubator overnight in (Fig. 4). IFN-c is the most important cytokine to inhibit intracellu- HEK-Blue™ detection medium. In the presence of any ligand inter- lar mycobacteria and we therefore wanted to include this cytokine acting with TLR2, HEK-Blue™-hTLR2 cells would secrete sAP in the in our studies. So, we treated human PBMCs with BCG strains for HEK-Blue™ detection medium and result in a color change from measuring IFN-c together with TNF-a, IL-1b and IL-8. The amount pink to purple/blue. Detection of color change was documented of IFN-c, TNF-a, IL-1b and IL-8 secreted by the PBMCs also varied with a spectrophotometer at 620 nm. We used another His-tagged as a function of the BCG strains [BCG (pMV261) or BCG recombinant protein rADIC424A as negative control. rADIC424A is a (pMV261 + Rv0079)] used for infection. The BCG strain over- mutant arginine deiminase (ADI) of Giardia duodenalis, which expressing DATIN induced higher secretion of IFN-c, TNF-a, IL-1b was expressed and purified from the similar expression system and IL-8 (Fig. 5). In conclusion, our experiments with differentiated as used for DATIN. TLR2 deficient HEK-Blue cells [(which express THP-1 cells and human PBMCs indicate that DATIN activates im- TLR4 instead of TLR2 on the cell surface), namely the HEK- mune cells to secrete cytokines relevant to formation and mainte- Blue™-hTLR4 (InvivoGen, USA)] were used as negative control. nance of granuloma.

3. Results 3.3. DATIN interacts with TLR2

3.1. DATIN is up-regulated under pH stress and microaerobic Our data suggest that DATIN induces the secretion of pro- conditions inflammatory cytokines. Further, we were interested to identify the receptors on the cells through which DATIN triggers and up- We wanted to determine the effect of stress such as low oxygen regulates cytokine expression. In order to identify possible inter- or low pH on the expression of DATIN. Furthermore, we were inter- acting domains of DATIN as juxtaposed to TLR2, several programs ested in knowing the level of up-regulation of DATIN in a BCG such as HEX [32], GRAMM-X [33] and PatchDock were employed derivative overexpressing Rv0079 [strain BCG (pMV261 + for unbound protein–protein docking with TLR2 as a receptor Rv0079)]. The strain M. bovis BCG with the empty vector and DATIN as a ligand. Approximately 100 predictions were gener- pMV261 displays a physiologic expression of Rv0079 that was rep- ated using PatchDock. Ten best solutions were selected on the basis resented as relative expression level 1 in Fig. 1 and served as refer- of geometric shape and complementarity score and were further ence to determine the relative expression in the over-expressing analyzed to identify one that buried maximum surface area upon derivative. We observed a15 fold expression of Rv0079 in BCG (p complex formation. Since there were no biological data available MV261 + Rv0079) as compared to the control BCG strain [BCG to identify the binding interface, MetaPPISP [27] was used to gen- (pMV261)] (Fig. 1). It has been reported that Rv0079 is upregulated erate possible interacting residues with respect to both DATIN and in different stress conditions as compared to stress-free conditions TLR2. This profile was harnessed to analyze the 10 docked com- [7,8]. We also determined relative expression of Rv0079 in different plexes for the presence of such residues in the interface. Several

Fig. 1. DATIN expression in BCG strain overexpressing Rv0079. RNA was isolated from M. bovis BCG (pMV261) and M. bovis BCG (pMV261 + Rv0079) grown under Fig. 2. Impact of stress on DATIN expression. RNA was isolated from M. bovis BCG normal conditions. Reverse transcription followed by real time PCR was carried out grown under different conditions. Reverse transcription followed by real time PCR to quantify the expression levels of DATIN. Up-regulation of DATIN was observed in was done to quantify the expression levels of DATIN. Up-regulation of DATIN was M. bovis BCG (pMV261 + Rv0079) as compared to M. bovis BCG (pMV261). observed under pH stress and microaerobic conditions. A. Kumar et al. / Cytokine 64 (2013) 258–264 261

Fig. 3. DATIN stimulates the secretion of proinflammatory cytokines by THP-1 cells. Bar diagrams portraying the dose dependent release of proinflammatory cytokines (TNF-a, IL-1b and IL-8) consequent to treatment by recombinant DATIN (labeled as rRv0079 or rDATIN) (1 lg, 2.5 lg and 5 lg) for 24 h. Data represent the means ± SDs of three technical replicates; cells treated with human interferon-c+LPS were used as positive control (hu-INF-c+LPS).

Fig. 4. Overexpression of DATIN by M. bovis BCG stimulates the secretion of proinflammatory cytokines by THP-1 cells. Secretion of the cytokines TNF-a, IL-1b and IL-8 by THP-1 cells infected with BCG (pMV261) and BCG (pRv0079) for 24 h. Data represent the means ± SDs of the results of three technical replicates; cells treated with human interferon-c+LPS were used as positive control (hu-INF-c+LPS).

Fig. 5. M. bovis BCG overexpressing DATIN stimulates the secretion of proinflammatory cytokines by PBMCs. Secretion of the cytokines IFN-c, TNF-a, IL-1b and IL-8 by human PBMCs infected by BCG (pMV261) and BCG (pRv0079) for 24 h. Data represent the means ± SDs of the results of three technical replicates.

of the lowest energy docking models emerging from this exercise the positive control reading. When we applied rADIC424A protein showed DATIN on the side of the TLR2. Among ten docked com- (instead of DATIN) as a negative control, no TLR2 activation was plexes, complex 9 was identified as the most plausible one on observed. We also used the HEK-TLR4 cells which express TLR4 the basis of minimum energy score and binding interface residues. on the cell surface to see if there is any interaction with TLR4. A docking model of TLR2-DATIN is shown in Fig. 6A, in which loop These cells were the same as the HEK-TLR2 cells (only the ex- 8 region of DATIN is docked into a domain formed by loop 2 region pressed TLR type was different). There was no reaction with the of TLR2 (Fig. 6A). No covalent bonding was observed in docking HEK-TLR4 cells confirming that the observed signaling was studies while the proposed interaction is dependent on electro- through TLR2. The negative reaction with TLR4 also ruled out the static forces. To confirm DATIN–TLR2 interaction, the recombinant possibility of the presence of any lipopolysaccharide fractions in HEK-Blue™-2 cell line was used which expressed TLR2 on the cell our protein preparation. Apart from these critical controls, other surface. Generally, if any ligand interacts with TLR2, it should acti- experiment level controls such as selection media control, positive vate downstream signaling through NFkB activation resulting in control, growth media control, and fetal bovine serum control were transcription and secretion of alkaline phosphatase (sAP) engi- included in the test. In sum, these observations clearly support the neered in the cell line; it should then react with detection media interaction of DATIN with TLR2. and result in color change. Accordingly, we compared the interac- tion of TLR2 and DATIN with its interaction with the positive con- 4. Discussion trol provided in the detection kit (InvivoGen). DATIN showed significant interaction with TLR2 (Fig. 6B). The activation of TLR2 One of the most baffling aspects of M. tuberculosis infection is induced by the application of 1 lg of DATIN achieved 76.5% of that only a small fraction of the infected or exposed individuals 262 A. Kumar et al. / Cytokine 64 (2013) 258–264

Fig. 6. DATIN interacts with TLR2. A. Interaction of DATIN with TLR2 using PatchDock: the residues of DATIN and TLR2 are portrayed in cyan and green respectively. The residues showing interaction among both the proteins are labeled and shown as stick model in element colors (carbon colored green/pink, nitrogen colored blue, and oxygen red). Hydrogen bonds are annotated by black dashed lines. B. Comparative activation of TLR2 by DATIN with respect to positive control: HEK-Blue™-2 cells expressing TLR2 were treated with different concentrations (1 lg and 0.2 lg) of DATIN and interaction of TLR2 with DATIN was quantified by measuring the activity of the secreted alkaline phosphatase (sAP) read as color change which was documented as absorption measurement at 620 nm. The sAP activity caused by the positive control was set to 100%. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) develop active disease. In the remaining cases, M. tuberculosis recombinant BCG strain induced higher levels of IFN-c, TNF-a, IL- might persist in a dormant, non-replicative form and a state of 1b and IL-8 secretion (Figs. 4 and 5) supportive of a role for DATIN drug indifference for a prolonged time with an unbattered poten- in mounting a proinflammatory response. It is noteworthy that tial to resuscitate whenever a conducive milieu is available. Under- TNF-a plays a significant role in granuloma formation, macrophage standing the latent or dormant form of M. tuberculosis has been a activation and also immune regulatory functions [16,17]. The stud- tough endeavor because no significant animal models are available ies in mouse models suggested that IFN-c and TNF-a were pro- and the in vitro models are not quite successful or remain imprac- duced in the lungs of mice with a persistent M. tuberculosis tical. Further, the micro-environment of the granuloma, where the infection [37]. We interpret that immune regulation by the secre- dormant M. tuberculosis lodges, is impermeable to the present day tion of proinflammatory cytokines IFN-c, TNF-a, IL-1b and IL-8 anti-TB drugs or there is less effect of antibiotics on dormant tuber- may be significant in the formation of granuloma, in maintenance cle bacilli due to their non-replicating state [34]. Given this, the of its microenvironment and also for the persistence of latent fundamental challenge in the control of tuberculosis could be mycobacteria. due to the frustrating lack of state of art concerning molecular M. tuberculosis is specifically recognized by toll like receptors mechanisms triggering the onset of latency, the maintenance of such as TLR2, whose activation triggers transcription of proinflam- the latent state and the reactivation of dormant bacilli. In this matory cytokines such as IL1b, TNFa, and IL8, which are all essen- study, we attempted to explore this problem at the level of DosR tial for the recruitment of the cells of inflammation to the site of regulon gene functions by a combination of predicted protein–pro- infection and also to contain the infection. It is known that ligands tein interaction followed by validation through cell signaling and from mycobacteria induce secretion of pro-inflammatory [38–40] in vitro cytokine profiling using recombinant protein as well as and anti-inflammatory [41] molecules via TLR2 signaling. Our data transformed BCG strains. The use of recombinant BCG strains for specifically suggest the interaction of DATIN with TLR2 which we the analysis of genes from the M. tuberculosis complex is a main- see as an interesting possibility that has the potential to enrich stream and accepted approach and it is especially suited because the library of bacterial factors that influence host responses. a majority of BCG and M. tuberculosis genes are 100% homologous Our computational modeling could generate with a high degree and many of the DosR regulon genes have been already analyzed in of confidence a model with binding sites on DATIN corresponding BCG. to ribosomal 30S [19] and the TLR2 (this study). In silico docking It has been shown that DosR regulon genes are up-regulated analyses indicated that DATIN specifically interacts with the loop during dormancy and that they have important immunological 2 region of TLR2 (Fig. 6). Although technically highly demanding, roles [11,35] to play including their newly identified dormancy it will be possible in future to further confirm the uniqueness of functions such as stabilization of ribosomes [19,36]. In order to the binding pockets, or otherwise, on both the ligand and recep- find a new way to cure tuberculosis it is important to know the ef- tor(s) sites through mutagenesis. Nevertheless, the present set of fect of virulence associated genes of the DosR regulon and their experiments and the data generated were of sufficiently high qual- interaction with the host immune system. DATIN is one of the ity and were robust enough to approach the hypothesis of TLR2 strongly up-regulated genes in the DosR regulon [8]; the latter interaction. We could demonstrate that DATIN interacts with encompasses an essential set of genes needed for adaptation to TLR2 in a dose dependent manner. Another protein (rADIC424A) stress conditions. While DATIN has an established role in transla- used as control and applied in the same concentrations did not tion control [19,36], its amino acid sequence presented with high bind to TLR2. The DATIN concentrations reached within a granu- antigenic index prompted us to look into its possible immunolog- loma are not known and as a consequence we do not know the ical/antigenic potential. Further, DATIN was earlier listed as a can- physiological relevance of the quantum of DATIN used in our didate to induce human T-cell responses [11]. While pursuing experiments. But it is an often applied and accepted practice to ex- these upstream observations, and when we treated differentiated press proteins and then add them to cell cultures to monitor cellu- human macrophages (THP-1) with DATIN, an increased secretion lar response towards the protein. In all these studies the relevance of proinflammatory cytokines was observed (Fig. 3) revealing that with respect to physiological concentrations is difficult to espouse DATIN activates macrophages and could likely play a significant or predict. It is also to be considered that the local concentration of role in innate immune functions that govern the microenviron- molecules within a tissue or aggregation of cells is not uniform, ment of tuberculous lesions. As observed, the over-expressing leading to the possibility of relatively high local concentrations of A. Kumar et al. / Cytokine 64 (2013) 258–264 263 ligands. Having this said, our results strongly support a role of DA- [7] Voskuil MI, Schnappinger D, Visconti KC, Harrell MI, Dolganov GM, Sherman TIN in cell signaling via TLR2. DR, et al. Inhibition of respiration by nitric oxide induces a Mycobacterium tuberculosis dormancy program. J Exp Med 2003;198:705–13. Given these observations interpreted in the light of our previous [8] Voskuil MI, Visconti KC, Schoolnik GK. Mycobacterium tuberculosis gene study [19] that described DATIN to be a translation inhibitor, it is expression during adaptation to stationary phase and low-oxygen dormancy. tempting to espouse that the two phenomena, namely the ribo- Tuberculosis 2004;84:218–27. [9] Cosma CL, Humbert O, Sherman DR, Ramakrishnan L. Trafficking of some binding/stabilization and interaction with TLR2 to trigger superinfecting Mycobacterium organisms into established granulomas occurs proinflammatory signaling could be interrelated as they both favor in mammals and is independent of the Erp and ESX-1 mycobacterial virulence latency. However, it is not plausible from our data that both the loci. J Infect Dis 2008;198:1851–5. [10] Leistikow RL, Morton RA, Bartek IL, Frimpong I, Wagner K, Voskuil MI. The processes are interdependent although this possibility cannot be Mycobacterium tuberculosis DosR regulon assists in metabolic homeostasis negated. Nonetheless, in the event of DATIN limiting its own trans- and enables rapid recovery from nonrespiring dormancy. J Bacteriol lation by ribosome binding, as previously suggested [19], reduced 2010;192:1662–70. [11] Leyten EM, Lin MY, Franken KL, Friggen AH, Prins C, van Meijgaarden KE, et al. DATIN availability might affect proinflammatory activity needed Human T-cell responses to 25 novel antigens encoded by genes of the to maintain granuloma. But, the granuloma microenvironment dormancy regulon of Mycobacterium tuberculosis. Microb Infect (cytokine mediated oxidative stress) could again serve as a trigger 2006;8:2052–60. for DosR and DATIN expression; given this, the existence of a finely [12] Ozinsky A, Underhill DM, Fontenot JD, Hajjar AM, Smith KD, Wilson CB, et al. The repertoire for pattern recognition of pathogens by the innate immune negotiated homeostasis of DATIN expression and function at the le- system is defined by cooperation between toll-like receptors. Proc Natl Acad vel of granuloma cannot be ruled out. Further the exact mode of Sci USA 2000;97:13766–71. secretion of DATIN and its presentation to TLR2 would remain mat- [13] Stenger S, Modlin RL. Control of Mycobacterium tuberculosis through mammalian Toll-like receptors. Curr Opin Immunol 2002;14:452–7. ter of future research although its presence in mycobacterial cul- [14] Brightbill HD, Libraty DH, Krutzik SR, Yang RB, Belisle JT, Bleharski JR, et al. ture supernatants hints at its being in the extracellular Host defense mechanisms triggered by microbial lipoproteins through toll-like compartment (11). receptors. Science 1999;285:732–6. [15] Bulut Y, Faure E, Thomas L, Equils O, Arditi M. Cooperation of Toll-like receptor Finally, we believe that our study provides a much needed 2 and 6 for cellular activation by soluble tuberculosis factor and Borrelia snapshot of DosR regulon protein functions with DATIN being an burgdorferi outer surface protein A lipoprotein: role of Toll-interacting protein interesting candidate at the cross roads of two different latency and IL-1 receptor signaling molecules in Toll-like receptor 2 signaling. J Immunol 2001;167:987–94. functions. We believe that the combination of computational, cell [16] Cooper AM. Cell-mediated immune responses in tuberculosis. Ann Rev culture based and recombinant BCG based approaches was the Immunol 2009;27:393–422. uniqueness of the study. In view of the predicted interaction mod- [17] Sasindran SJ, Torrelles JB. Mycobacterium tuberculosis infection and inflammation: what is beneficial for the host and for the bacterium? Front el, experiments carried out using recombinant DATIN and the cor- Microbiol 2011;2:2. roboration of DATIN functions using the BCG environment made [18] Broaddus VC, Hebert CA, Vitangcol RV, Hoeffel JM, Bernstein MS, Boylan AM. the study further interesting. Future discovery and vaccine devel- Interleukin-8 is a major neutrophil chemotactic factor in pleural liquid of opment efforts would likely benefit from this knowhow. patients with empyema. Am Rev Resp Dis 1992;146:825–30. [19] Kumar A, Majid M, Kunisch R, Rani PS, Qureshi IA, Lewin A, et al. Mycobacterium tuberculosis DosR regulon gene Rv0079 encodes a putative, ‘dormancy associated translation inhibitor (DATIN)’. PloS One 2012;7:e38709. Acknowledgements [20] Mishra S. Function prediction of Rv0079, a hypothetical Mycobacterium tuberculosis DosR regulon protein. J Biomol Struct Dyn 2009;27:283–92. We thank Christoph Ufermann for his help with the TLR2 assay. [21] Sherman DR, Voskuil M, Schnappinger D, Liao R, Harrell MI, Schoolnik GK. Regulation of the Mycobacterium tuberculosis hypoxic response gene This work was supported by a Centre of Excellence Grant from the encoding alpha – crystallin. Proc Natl Acad Sci USA 2001;98:7534–9. Department of Biotechnology (DBT) of the Indian Government [22] Reichelt P, Schwarz C, Donzeau M. Single step protocol to purify recombinant entitled ‘Multidisciplinary approaches aimed at interventions proteins with low endotoxin contents. Prot Exp Purif 2006;46:483–8. against Mycobacterium tuberculosis’ (BT/01/C0E/07/02; compo- [23] Sharbati-Tehrani S, Meister B, Appel B, Lewin A. The porin MspA from Mycobacterium smegmatis improves growth of Mycobacterium bovis BCG. Int nent B2b) and another grant from the Council for Scientific and J Med Microbiol: IJMM 2004;294:235–45. Industrial Research (CSIR) India under the OSDD program to NA. [24] Stover CK, de la Cruz VF, Fuerst TR, Burlein JE, Benson LA, Bennett LT, Bansal The authors would like to also acknowledge support from the Ger- GP, Young JF, Lee MH, Hatfull GF, Snapper SB, Barletta RG, Jacobs JR WR, Bloom BR. New use of BCG for recombinant vaccines. Nature 1991;351:456–60. man Research Foundation (DFG) sponsored international research [25] Khattak FA, Kumar A, Kamal E, Kunisch R, Lewin A. Illegitimate recombination: training group (IRTG) entitled ‘Internationales Graduiertenkolleg an efficient method for random mutagenesis in Mycobacterium avium subsp. – functional molecular infection epidemiology – GRK1673 (Ber- Hominissuis. BMC Microbiol 2012;12:204. [26] Sharbati J, Lewin A, Kutz-Lohroff B, Kamal E, Einspanier R, Sharbati S. lin–Hyderabad)’. SEH is a J.C. Bose National Fellow (India) and a Integrated microRNA-mRNA-analysis of human monocyte derived Robert Koch Fellow of the Robert Koch Institute, Berlin, Germany. macrophages upon Mycobacterium avium subsp. hominissuis infection. PloS P.S.R. and A.K. would like to acknowledge DBT and the CSIR, One 2011;6:e20258. respectively, for the award of Senior Research Fellowships. [27] Qin S, Zhou HX. Meta-PPISP: a meta web server for protein–protein interaction site prediction. Bioinformatics 2007;23:3386–7. [28] Schneidman-Duhovny D, Inbar Y, Nussinov R, Wolfson HJ. PatchDock and SymmDock: servers for rigid and symmetric docking. Nucl Acids Res References 2005;33:W363–7. [29] Alvi A, Ansari SA, Ehtesham NZ, Rizwan M, Devi S, Sechi LA, et al. Concurrent [1] Boon C, Dick T. Mycobacterium bovis BCG response regulator essential for proinflammatory and apoptotic activity of a Helicobacter pylori protein hypoxic dormancy. J Bacteriol 2002;184:6760–7. (HP986) points to its role in chronic persistence. PloS One 2011;6:e22530. [2] Florczyk MA, McCue LA, Purkayastha A, Currenti E, Wolin MJ, McDonough KA. [30] Delano WL. The PyMOL molecular graphics system, Delano Scientific, San A family of acr-coregulated Mycobacterium tuberculosis genes shares a Carlos (CA); 2002. . common DNA motif and requires Rv3133c (dosR or devR) for expression. [31] Mizuguchi K, Deane CM, Blundell TL, Johnson MS, Overington JP. JOY: protein Infect Immun 2003;71:5332–43. sequence-structure representation and analysis. Bioinformatics [3] Honaker RW, Leistikow RL, Bartek IL, Voskuil MI. Unique roles of DosT and 1998;14:617–23. DosS in DosR regulon induction and Mycobacterium tuberculosis dormancy. [32] Macindoe G, Mavridis L, Venkatraman V, Devignes MD, Ritchie DW. Infect Immun 2009;77:3258–63. HexServer: an FFT-based protein docking server powered by graphics [4] Kendall SL, Movahedzadeh F, Rison SC, Wernisch L, Parish T, Duncan K, Betts JC, processors. Nucl Acids Res 2010;38:W445–9. Stoker NG. The Mycobacterium tuberculosis dosRS two-component system is [33] Tovchigrechko A, Vakser IA. GRAMM-X public web server for protein–protein induced by multiple stresses. Tuberculosis 2004;84:247–55. docking. Nucl Acids Res 2006;34:W310–4. [5] Kim MJ, Park KJ, Ko IJ, Kim YM, Oh JI. Different roles of DosS and DosT in the [34] Wayne LG, Sohaskey CD. Nonreplicating persistence of mycobacterium hypoxic adaptation of Mycobacteria. J Bacteriol 2010;192:4868–75. tuberculosis. Ann Rev Microbiol 2001;55:139–63. [6] Park HD, Guinn KM, Harrell MI, Liao R, Voskuil MI, Tompa M, et al. Rv3133c/ [35] Yamamura Y, Tsujimura K, Seto S, Uchijima M, Hozumia H, Nagatac T, et al. dosR is a transcription factor that mediates the hypoxic response of Immunogenicity of latency-associated antigens of Mycobacterium Mycobacterium tuberculosis. Mol Microbiol 2003;48:833–43. tuberculosis in DNA-vaccinated mice. Proc Vaccin 2010;3:19–26. 264 A. Kumar et al. / Cytokine 64 (2013) 258–264

[36] Trauner A, Lougheed KE, Bennett MH, Hingley-Wilson SM, Williams HD. The tuberculosis is mediated by Toll-like receptor 2-dependent release of tumor dormancy regulator DosR controls ribosome stability in hypoxic mycobacteria. necrosis factor-alpha. J Biol Chem 2007;282:1039–50. J Biol Chem 2012;287:24053–63. [40] Rani PS, Tulsian NK, Sechi LA, Ahmed N. In vitro cytokine profiles and viability [37] Flynn JL, Scanga CA, Tanaka KE, Chan J. Effects of aminoguanidine on latent of different human cells treated with whole cell lysate of Mycobacterium murine tuberculosis. J Immunol 1998;160:1796–803. avium subsp. Paratuberculosis. Gut Path 2012;4:10. [38] Aravalli RN, Hu S, Rowen TN, Palmquist JM, Lokensgard JR. Cutting edge: TLR2- [41] Nair S, Ramaswamy PA, Ghosh S, Joshi DC, Pathak N, Siddiqui I, et al. The PPE18 mediated proinflammatory cytokine and chemokine production by microglial of Mycobacterium tuberculosis interacts with TLR2 and activates IL-10 cells in response to herpes simplex virus. J Immunol 2005;175:4189–93. induction in macrophage. J Immunol 2009;183:6269–81. [39] Basu S, Pathak SK, Banerjee A, Pathak S, Bhattacharyya A, Yang Z, et al. Execution of macrophage apoptosis by PE_PGRS33 of Mycobacterium Molecular BioSystems

View Article Online PAPER View Journal | View Issue

Mycobacterial Hsp65 potentially cross-reacts with

Cite this: Mol. BioSyst., 2013, autoantibodies of diabetes sera and also induces 9, 2932 (in vitro) cytokine responses relevant to diabetes mellitus†

Pittu Sandhya Rani,a Banaganapalli Babajan,a Nikhil K. Tulsian,a Mahabubunnisa Begum,b Ashutosh Kumara and Niyaz Ahmed*ac

Diabetes mellitus is a multifactorial disease and its incidence is increasing worldwide. Among the two types of diabetes, type-2 accounts for about 90% of all diabetic cases, whereas type-1 or juvenile diabetes is less prevalent and presents with humoral immune responses against some of the autoantigens. We attempted to test whether the sera of type-1 diabetes patients cross-react with mycobacterial heat shock protein 65 (Hsp65) due to postulated epitope homologies between mycobacterial Hsp65 and an important autoantigen of type-1 diabetes, glutamic acid decarboxylase-65 (GAD65). In our study, we used either recombinant mycobacterial Hsp65 protein or synthetic peptides corresponding to some of the potential epitopes of mycobacterial Hsp65 that are shared with GAD65 or human Hsp60, and a control peptide sourced from mycobacterial Hsp65 which is not shared with GAD65, Hsp60 and other autoantigens of type-1 diabetes. The indirect ELISA results indicated that both type-1 diabetes and type-2 diabetes sera cross-react with conserved mycobacterial Hsp65 peptides and recombinant mycobacterial Hsp65 protein but do not do so with the control peptide. Our results suggest that cross-reactivity of mycobacterial Hsp65 with autoantibodies of diabetes sera could be due to the presence of significantly conserved peptides between mycobacterial Hsp65 and human Hsp60 rather than between mycobacterial Received 26th July 2013, Hsp65 and GAD65. The treatment of human peripheral blood mononuclear cells (PBMCs) with Accepted 29th August 2013 recombinant mycobacterial Hsp65 protein or the synthetic peptides resulted in a significant increase in the secretion of cytokines such as IL-1b, IL-8, IL-6, TNF-a and IL-10. Taken together, these findings point

Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM. DOI: 10.1039/c3mb70307j towards a dual role for mycobacterial Hsp65: in inducing autoimmunity and in inflammation, the two www.rsc.org/molecularbiosystems cardinal features of diabetes mellitus.

1. Introduction the possibility that early exposure to cow’s milk might lead to increased risk of type-1 diabetes mellitus (T1DM).5,6 As the Mycobacterium avium subsp. paratuberculosis (MAP) is an obligate number of diabetes cases around the world shoot up, India is pathogen linked to Johne’s disease in ruminants and is char- fast emerging as the ‘diabetes capital of the world’.7 The acterised by chronic inflammation of intestines.1 This zoonotic International Diabetes Federation in 2006 has projected around bacterium is also described as a probable causative agent of a 40.9 million people to be diabetic and this figure is expected to similar inflammatory condition of the gut in humans known as rise to 69.9 million by 2025.8 T1DM/juvenile diabetes is char- Crohn’s disease.2 MAP exists in two forms, bacillary and sphaero- acterized by the humoral response directed against human plast; the latter facilitates its survival during pasteurization and autoantigens such as GAD65, insulin, insulinoma associated chlorination.1,3,4 Furthermore, infant nutrition studies point to protein, Znt8 and Hsp60.7,9 Furthermore, it has been indicated that multiple factors such as genetic susceptibility, microbial a Pathogen Biology Laboratory, Department of Biotechnology and Bioinformatics, infections, lifestyle related factors and environmental toxins act University of Hyderabad, Hyderabad, India. E-mail: [email protected] cumulatively/individually as triggers of T1DM or type-2 diabetes b Department of Biochemistry, Deccan College of Medical Sciences and mellitus (T2DM).9,10 Recent studies have indicated that T1DM Allied Hospitals, Hyderabad, India c Institute of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia patient’s sera can cross-react with mycobacterial Hsp65, possibly † Electronic supplementary information (ESI) available: Fig. S1–S3 and Tables S1 because of epitope homology between the mycobacterial Hsp65 9,11 and S2. See DOI: 10.1039/c3mb70307j and human GAD65. Epitope homology between mycobacterial

2932 Mol. BioSyst., 2013, 9, 2932--2941 This journal is c The Royal Society of Chemistry 2013 View Article Online Paper Molecular BioSystems

Hsp65 and GAD65 (molecular mimicry) is considered as one of (gi|190358866|) using the online T-coffee multiple sequence the mechanistic processes that lead to T1DM in genetically alignment tool (EMBL-EBI). Five peptides were identified and susceptible individuals.11,12 Hsps are highly conserved and selected for the study. The synthetic linear peptides were constitutively expressed proteins that are found in all prokaryotes synthesized at Vimta Labs Limited, India. The HPLC analyses and eukaryotes and function as molecular chaperones to prevent of peptides indicated a purity of 96–99%. The amino acid the aggregation and misfolding of partially folded protein inter- sequences of MAP Hsp65, GAD65 and human Hsp60 were mediates.13,14 It has been indicated that under pathological submitted to I-TASSER (iterative threading assembly refine- conditions such as necrotic cell death, Hsps are released to the ment algorithm), a 3D protein structure prediction tool,20 in extracellular environment. They then act as pro-inflammatory order to predict the full length 3D structure of the protein. The mediators and/or antigenic carriers and induce cross-presentation rough models generated from I-TASSER were subjected to of antigens to sensitized Th and CTL cells.15 Autoantibodies energy minimization with the help of the steepest descent and T-cells reactive to Hsps have been identified in patients technique using the GROMOS96 force field21 for the elimination with numerous autoimmune diseases such as rheumatoid of bad contacts in the protein. The energy minimized models were arthritis, systemic lupus erythematosus (SLE), T1DM and multiple further evaluated by checking their stereo-chemical quality using sclerosis.16–18 The role of mycobacterial Hsp70 in the stimulation PROCHECK server.22 Molecular visualization and superpose of autoantibody production in SLE has been described.19 Given predications were carried out using PyMOL.23 To analyse the these observations, it is important to dissect further the mecha- solvent accessibility surface area (SAS) of 14 peptides (repre- nisms of cross-reactivity and to identifyspecificmotifsinvolvedin senting MAP Hsp65, GAD65 and human Hsp60), we used the T1DM. In this study, we attempted to unravel the role of epitope NetSurfP web based tool.24 homology between mycobacterial Hsp65, GAD65 and human Hsp60, and to determine the innate immune response of myco- 2.3 Purification of recombinant mycobacterial Hsp65 bacterial Hsp65. We used the peptide stretches of MAP Hsp65 that The protein sequences of Hsp65 from the Mtb strain H37Rv and are shared with human GAD65 and/or Hsp60 together with a non- MAP Hsp65 revealed 98% similarity (data not shown). The conserved peptide (without any homology with human GAD65 plasmid construct for the expression of Mtb Hsp65 was obtained or other human antigens of T1DM, such as human Hsp60). from Shekhar Mande (CDFD, Hyderabad). Luria-Bertani broth Our observations explain that epitope homology (molecular was inoculated with overnight grown cultures of BL-21 cells mimicry) between the mycobacterial Hsp65 and the host expressing recombinant Mtb Hsp65 and placed in a shaking

autoantigen, human Hsp60, might constitute an important incubator at 37 1C until an OD600 of 0.5 was reached. The culture mechanism behind the cross-reactivity of diabetes sera with was then induced with 1 mM IPTG (isopropyl-b-D-1-thiogalacto- mycobacterial Hsp65. Further, we found mycobacterial Hsp65 pyranoside, Sigma) and incubated at 37 1C for 4 h in a shaking and its synthetic peptides to be involved in the stimulation of incubator. The culture was then pelleted at 3200 Â g for 10 min. proinflammatory and anti-inflammatory cytokine responses The pellet was suspended in re-suspension buffer (50 mM Tris, relevant to diabetes mellitus. 150 mM NaCl, 15 mM imidazole, 0.3% sarcosyl and 1 mM PMSF). After sonication, the cell lysate was centrifuged at 1 Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM. 2. Materials and methods 10 000 Â g for 30 min at 4 C. The supernatant was collected and loaded on Ni2+-NTA beads in the column. After passage of 2.1 Study population and clinical presentation the flowthrough, the column was washed with wash buffer Our patient group comprised a total of 66 individuals of either (50 mM Tris, 500 mM NaCl, 30 mM imidazole). His-tagged sex (22 healthy, 22 T1DM and 22 T2DM). The diabetic indivi- protein was eluted using elution buffer (50 mM Tris, 500 mM duals were classified and identified as T1DM or T2DM based on NaCl, 100 mM imidazole). The purified protein was then dialysed established clinical criteria and were negative for Mycobacterium in 1Â PBS buffer. After concentrating the protein using Amicons tuberculosis (Mtb) infection by routine diagnosis. 5 ml of blood ultracentrifugation filters (Millipore), the protein concentration was collected from each subject and the sera obtained were was determined by Bradford’s method. The protein aliquots were stored at À80 1C for ELISA. For isolation of human peripheral then stored at À80 1C. The purified protein was treated with blood mononuclear cells (PBMCs), approximately 10 ml of blood polymyxin B (Sigma) to neutralize the effect of any endotoxin was taken from three healthy individuals. Written informed contamination as described previously.25 consent was obtained from all subjects and the study was approved by ethics committees of the participating hospital(s) 2.4 Indirect ELISA using recombinant mycobacterial Hsp65 and institutes. protein and MAP Hsp65 peptides The wells of the ELISA plates (Axygen, India) were individually 2.2 T-coffee analysis, homology modeling and SAS analysis coated with the purified recombinant Mtb Hsp65 protein (Fig. S1, The protein sequence of MAP Hsp65 (gi|438181|) was compared ESI†) and MAP Hsp65 peptides (conserved and non-conserved) of with the sequence of human autoproteins (against which the approximately 500 ng concentration in bicarbonate buffer. The humoral response was observed in T1DM sera) namely GAD65 plates were incubated at 4 1C overnight. The plates were then (gi|1352216|), human Hsp60 (gi|129379|), insulin (gi|386828|), washed thrice with 1Â PBST (1Â PBS with 0.05% Tween 20). The insulinoma associated protein-2 (gi|145309322|) and Znt8 uncoated sites of the wells were blocked with 1% bovine serum

This journal is c The Royal Society of Chemistry 2013 Mol. BioSyst., 2013, 9, 2932--2941 2933 View Article Online Molecular BioSystems Paper

albumin (BSA, Sigma) in 1Â PBS for 2 h by incubating the plates The mean and standard deviation were calculated for each at 37 1C. Each plate was washed thrice with 1Â PBST. The 1 : 50 experiment conducted in triplicate. Two tailed Student’s t-test diluted sera samples (healthy, T1DM, T2DM) were added, as was considered significant if p values were o0.05; the p value appropriate, to each corresponding well and incubated for 1 h was calculated using online GraphPad Prism software. at 37 1C. After incubation, the plates were washed thrice with 1Â PBST and 1 : 15 000 diluted anti-IgG human conjugated 3. Results and discussion peroxidase antibody (Sigma) was added and incubated for 1 h at 37 1C. The plates were then washed thrice with 1Â PBST and 3.1 Molecular structure prediction and selection of once with 1Â PBS. The plates were developed with OPD immunogenic peptides representing MAP Hsp65

(o-phenylenediamine, Sigma, USA) and H2O2 in citrate buffer. In T1DM patients, a greater level of antibodies (humoral

The reaction was stopped with 2 N H2SO4. The intensity of the response) was observed against human autoprotein GAD65. colour so developed was read at 490 nm using an ELISA reader T-coffee analysis of MAP Hsp65 and human GAD65 revealed (Infinite M200, TECAN). 45% identity and the analysis led to the identification of four conserved peptides from MAP Hsp65 (amino acid stretches that 2.5 Cytokine analysis of culture supernatants of human are highlighted in red in Fig. S2, ESI†): PBMCs peptide1 (ELEDPYEKIGAELVKEVAKK), Human PBMCs were isolated from heparinised blood of healthy peptide2 (DQIAATAAISAGDQS), individuals using the Ficoll-Hypaque (HiSepTM LSM, Himedia, peptide3 (LAKLAGGVAVIKAGAATEVELKERKHRI), India) method of Boyum.26 A total of about 0.5 million PBMCs peptide4 (AVEEGIVAGGGVALLHAIPALD). per ml of complete RPMI1640 (Hyclone, USA) media supplemen- To unravel the role of epitope homology between MAP Hsp65 ted with 10% foetal bovine serum (FBS, Hyclone) were treated and GAD65, it was significant to identify a non-conserved peptide with different concentrations (1 mg, 5 mg, 10 mg) of recombinant of MAP Hsp65 that doesn’t cross-react with GAD65 and with any Mtb Hsp65 protein and 0.5 mgofLPS(E. coli, Sigma). The culture other autoproteins of T1DM. The T-coffee analysis was performed supernatants were collected at 24 and 48 h respectively and for MAP Hsp65 as compared to other autoproteins of T1DM such stored at À80 1C until the cytokine analysis was performed. as human Hsp60, insulin, insulinoma associated protein-2 and Separate aliquots of about 0.5 million PBMCs per ml of complete Znt-8. The T-coffee analysis of MAP Hsp65 and human Hsp60 RPMI1640 media supplemented with 10% FBS were treated indicated 97% identity; MAP Hsp65 and insulin indicated 62% individually with four conserved peptides (each separately) identity; MAP Hsp65 and insulinoma associated protein indicated and a non-conserved peptide, all representing MAP Hsp65 52% identity; and MAP Hsp65 and Znt-8 indicated 51% identity. As (at different concentrations: 1 mg, 5 mg, 10 mg), and 0.1 mgof MAP Hsp65 and human Hsp60 have greater % identity, a non- LPS, in separate wells. The culture supernatants were collected conserved peptide of MAP Hsp65, from T-coffee analysis of human at 24 h and stored at À80 1C until cytokine analysis was carried Hsp60 and MAP Hsp65, the peptide5 (VGLSLESADI), was identi- out. The concentrations of cytokines IL-1b, IL-6, IL-8, IL-10 and fied (Fig. S3, ESI†). The peptide5 was further checked for over- TNF-a in culture supernatants were measured using the BD lapping amino acid stretches in the protein sequence of GAD65,

Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM. CBA Flex kit (BD Biosciences, USA) on a BD FACS Canto II flow Znt-8, insulin and insulinoma associated protein and it was cytometer by plotting the standard curve for each cytokine confirmed that peptide5 was indeed non-conserved with respect using FCAP array software (Soft Flow/BD Biosciences). to the autoantigens of T1DM (Table S2, ESI†). The PDB structure search analysis of MAP Hsp65, GAD65 and human Hsp60 revealed 2.6 Statistical analysis that these sequences have corresponding templates with more The ELISA data were analyzed by Mann–Whitney’s U test or the than 92% sequence identity. Even though the sequences had Wilcoxon rank sum test. We preferred Mann–Whitney’s U test enough identities to build a homology model, some of the when compared to Student’s t-test because of its robustness important protein segments in MAP Hsp65, GAD65 and human and also because the samples in our study were of the same size Hsp60 were missing. For this reason, we reconstructed 3D models in all the three categories. The level of significance was deter- of MAP Hsp65, GAD65 and human Hsp60 by using I-TASSER web mined by p value. The plotting of ELISA graphs and calculation server. A total of five structures each were predicted for each of the of median values, values of U and level of significance were all above proteins. Only best structures were selected (Fig. S4, ESI†) carried out using online GraphPad Prism software. The U based on the maximum C-score, TM-score and RMSD value values of each test are presented in a table (Table S1, ESI†). (Table 1) and these values were found to be in correct topology The p value o 0.05 was considered significant, indicating that with respect to reference values.27 After energy minimization of the the median value of the two groups significantly differs from 3D models of MAP Hsp65, GAD65 and human Hsp60, the stereo- each other. Statistical analysis and plotting of graphs for the chemical quality of the structures was validated by submitting the level of cytokine secretion by human PBMCs on treatment with PDB files to PROCHECK server. The results of PROCHECK analysis Mtb Hsp65 protein in the culture supernatants was carried out indicated that a relatively low percentage of residues have phi/psi by using GraphPad Prism software. The plotting of graphs for angles in the disallowed regions suggesting the acceptability of the level of cytokine secretion by human PBMCs on treatment Ramachandran plots for the proteins.28 The percentage of residues with MAP Hsp65 peptides was performed by using Sigma plot. in the allowed/core region were found to be 98.9, 99.2 and 98.8%

2934 Mol. BioSyst., 2013, 9, 2932--2941 This journal is c The Royal Society of Chemistry 2013 View Article Online Paper Molecular BioSystems

Table 1 Evaluation data of 3D protein models as gleaned from I-TASSER Though Hsp60 is an intracellular protein, its elevated levels have been found in the systemic circulation of T2DM patients. The No. of No. of S. no. Protein C-scorea TM-scoreb RMSDc decoys clustersd exact mechanism of secretion of Hsp60 into the biological fluids of T2DM individuals is not yet known.29 1 MAP Hsp65 0.99 0.85 Æ 0.08 5.3 Æ 3.4 Å 2261 0.3925 2 GAD65 1.16 0.57 Æ 0.15 10.4 Æ 4.6 Å 1047 0.0706 3 Human 0.85 0.83 Æ 0.08 5.8 Æ 3.6 Å 2237 0.3436 3.2 Cross-reactivity of diabetes sera with mycobacterial Hsp65 Hsp60 protein and MAP Hsp65 peptides a Confidence score (C-score) for estimating the quality of predicted The indirect ELISA results indicated that recombinant Mtb models typically in the range of [À5, 2]. b Template Modeling score (TM-score) measures the structural similarity between two structures; Hsp65 protein and conserved MAP Hsp65 peptides can cross- TM-score > 0.5 indicates a model of correct topology and a TM-score o react with both T1DM and T2DM sera when compared to sera c 0.17 conveys random similarity. Heavy atoms root-mean-square devia- from healthy controls. T1DM is an autoimmune disease with tion (RMSD) with respect to the experimental structure. d Number of structure decoys (low temperature replicas) at a unit of space in the antibodies produced against autoantigens such as GAD65, SPICKER cluster. Znt-8, insulin, insulinoma associatedproteinandhumanHsp60. Although T2DM is not an autoimmune disease, antibody response was previously reported against human Hsp60 in T2DM as well.29 and those in the disallowed region were found to be 1.1, 0.8 and Given that MAP Hsp65 is 97% identical to human Hsp60, we 1.2% for MAP Hsp65, GAD65 and human Hsp60, respectively could observe T2DM sera reacting with Mtb Hsp65 protein and (Fig. S5 and Table S4, ESI†). conserved peptides of the same. Localization of peptides on the 3D structures of MAP Hsp65 Further, assay with the non-conserved peptide of MAP and GAD65 is indicated in Fig. 1. Further, localization of peptides Hsp65 (peptide5) indicated no significant median difference after superimposition of 3D structures of MAP Hsp65 and human between the healthy control, T1DM and T2DM sera (Fig. 3). Hsp60 is presented in Fig. 2. The SAS was calculated using the It appears that the mycobacterial Hsp65 cross-reacts with NetSurfP web tool for individual residues of the 14 test peptides autoantibodies in T1DM and T2DM sera due to overlapping representing MAP Hsp65, GAD65 and human Hsp60. NetSurfP peptides that mimic the epitopes of the host proteins, GAD65 Z-scores enabled identification of the most reliable/unreliable and/or human Hsp60. predictions for both buried and exposed amino acids (Table S3, In a related study,30 the peptide3 sequence (LAKLAGGVAVI- ESI†). As shown in Table 2, MAP Hsp65 peptide1 showed 42.1% KAGAATEVELKERKHRI) was observed to bind to I-Ag7 (mouse of exposed residues and its overlapping peptide in GAD65 MHC) with high affinity and induce enhanced proliferation of showed more than 89% of exposed residues in its structure. T-cells. However, the peptide4 sequence (AVEEGIVAGGGVALL- Conserved peptides 2, 3 and 4 of MAP Hsp65–GAD65 revealed HAIPALD) was indicated to bind to I-Ag7 with less affinity and approximately equal ratios of buried and exposed residues. The induce lower proliferation of T-cells.30 Our results corroborate non-conserved peptide of MAP Hsp65–human Hsp60 showed with this report and suggest that the above two peptides might be 70% and 80% of exposed residues respectively. Studies have acting as T-cell and B-cell epitopes. The peptide5 (VGLSLESADI), indicated humoral immune response to human Hsp60 in both although having surface probability, could not cross-react with

Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM. T1DM and T2DM individuals. Human Hsp60 is a mitochondrial T1DM or T2DM sera due to the lack of conserved/overlapping stress protein that is induced during mitochondrial impairment. motifs in comparison to autoantigens of T1DM and T2DM.

Fig. 1 Identification of conserved peptides of MAP Hsp65 and GAD65. The conserved peptides are represented in different colours: peptides of MAP Hsp65 are shown in yellow and peptides entailing GAD65 are shown in cyan.

This journal is c The Royal Society of Chemistry 2013 Mol. BioSyst., 2013, 9, 2932--2941 2935 View Article Online Molecular BioSystems Paper

Fig. 2 Superimposition of the predicted structures of MAP Hsp65 (cyan) and human Hsp60 (green). The structures are completely aligned at their identical sequences except at non-conserved regions as shown in the zoom view. The conserved MAP Hsp65 peptides are represented in yellow, conserved human Hsp60 peptides in brown, non-conserved MAP Hsp65 peptides in blue and non-conserved human Hsp60 peptides in red.

Table 2 NetSurfP prediction results for % solvent accessibility of the residues of MAP Hsp65, GAD65 and human Hsp60 peptides

Peptide Buried Exposed Protein Domain number Peptide sequence residues (%) residues (%) MAP Hsp65 Conserved domains 1 ELEDPYEKIGAELVKEVAKK 57.9 42.1 2 DQIAATAAISAGDQS 46.6 53.4 Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM. 3 LAKLAGGVAVIKAGAATEVELKERKHRI 71.5 28.5 4 AVEEGIVAGGGVALLHAIPALD 72.7 27.3 Non-conserved domain 5 VGLSLESADI 30 70

GAD65 Conserved domains 1 DAEKPAESGGSQPPRAAARK 10.5 89.5 2 CQTTLKYAIKTGHPR 66.6 33.4 3 LMWRAKGTTGFEAHVDKCLELAEYLYNI 67.8 32.2 4 MVFDGKPQHTNVCFWYIPPSLR 72.7 27.3

Human Hsp60 Conserved domains 1 DLKDKYKNIGAKLVQDVANN 55 45 2 EEIAQVATISANGDKE 43.75 56.25 3 LAKLSDGVAVLKVGGTSDVEVNEKKDRV 67.8 32.2 4 AVEEGIVLGGGCALLRCIPALD 77.3 22.7 Non-conserved domain 5 LTNLEDVQP 20 80

A previous study indicated association of mycobacterial Hsp65 with Mtb Hsp65 protein and MAP Hsp65 conserved peptides with rheumatoid arthritis.31 It was shown that the peptide could be mostly due to the presence of autoantibodies rather sequence of M. bovis Hsp65 spanning amino acids 180–188 acts than antibodies produced against any other bacterial Hsps. This is as a major T-cell epitope for adjuvant arthritis; however, this T-cell because no such cross-reactivity was observed in healthy controls. epitope was unable to cross-react with sera of rheumatoid arthritis patients suggesting that this peptide is not recognized as a B-cell 3.3 Pro- and anti-inflammatory cytokine profiles of epitope.31 Further, Hsps are conserved proteins and Hsps homo- human PBMCs logous to mycobacterial Hsp65 might as well occur in many The cytokine analysis of culture supernatants of human PBMCs bacteria. The observed cross-reactivity of T1DM and T2DM sera upon treatment with recombinant Mtb Hsp65 protein indicated

2936 Mol. BioSyst., 2013, 9, 2932--2941 This journal is c The Royal Society of Chemistry 2013 View Article Online Paper Molecular BioSystems Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM.

Fig. 3 Indirect ELISA assay with recombinant Mtb Hsp65 protein and MAP Hsp65 peptides. The ELISA analysis of healthy, T1DM and T2DM sera against the recombinant Mtb Hsp65 protein and conserved peptides of MAP Hsp65 revealed significant serum antibody profiles suggestive of cross-reactivity of mycobacterial Hsp65 with GAD65 and/or human Hsp60 (autoproteins), but no significant antibody response with the non-conserved peptide (peptide5) of MAP Hsp65 was observed, indicating the absence of cross-reactivity. Data of the assay are indicated as OD at 490 nm for each group and the median value of each group is represented by a bold horizontal line.

increased secretion of cytokines IL-1b, IL-8, IL-6, IL-10 and suggested by another study that the pro- and anti-inflammatory TNF-a in a dose and time dependent manner, and in a dose cytokines and their signalling pathways play a key role in dependent manner, when treated with MAP Hsp65 peptides protection during mycobacterial pathogenesis and their balance (Fig. 4 and 5). likely determines the clinical outcome.32 The non-conserved peptide (peptide5), although not cross- In our previous study,33 we analyzed the cytokines secreted reacting with T1DM or T2DM sera samples as compared to the by human PBMCs upon treatment with whole cell lysates of conserved peptides, could significantly induce cytokine responses. MAP; the results indicated increased secretion of IL-1b, TNF-a, This suggests that this non-conserved peptide could be a potential IL-8, IL-6 and IL-10. In this study, we sought to analyze the epitope involved in innate immune response even if it is not profiles of cytokines secreted upon treatment with Mtb Hsp65 conserved with respect to human Hsp60 or to any other autoantigen protein and peptides from MAP Hsp65. When we consider the of T1DM. IL-1b, IL-6, IL-8, and TNF-a are pro-inflammatory two studies together, we observe an increased level of secretion cytokines and IL-10 is an anti-inflammatory cytokine. It has been of IL-6, IL-8 and IL-10 by human PBMCs on treatment with

This journal is c The Royal Society of Chemistry 2013 Mol. BioSyst., 2013, 9, 2932--2941 2937 View Article Online Molecular BioSystems Paper Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM.

Fig. 4 Cytokine analysis of culture supernatants of human PBMCs upon treatment with recombinant Mtb Hsp65. Multiplex cytokine bead array analysis was carried out for culture supernatants of human PBMCs treated with different concentrations of recombinant Mtb Hsp65 protein. The data indicated increased secretion of IL-1b, IL-6, IL-8, IL-10 and TNF-a in a dose and time dependent manner. The Student’s t-test of the cytokine concentrations as observed in three individual experiments entailing untreated (added 1Â PBS; the solvent used for dialyzing the protein) control in comparison with the supernatants of the cells treated with different concentrations of recombinant Mtb Hsp65 and LPS (positive control) revealed p o 0.05 indicating statistically significant difference between the untreated and treatment groups when recorded at 24 and 48 h.

Mtb Hsp65 protein when compared to treatment with MAP Furthermore, studies have also indicated increased levels of lysate. Further, the level of cytokines secreted by human PBMCs IL-8 in both T1DM and T2DM individuals.35 The in vivo studies on treatment with MAP Hsp65 peptides was low when com- using non-obese diabetic (NOD) mice have shown that pancreatic pared to treatment with recombinant Mtb Hsp65; this may be IL-10 induces autoimmune diabetes via an ICAM dependent because the peptides are less immunogenic when compared to pathway.36 Another study reported induction of insulin resistance whole protein. Moreover, studies from another group identified by TNF-a and IL-6.37 Taken together, our cytokine profiling of that monocytes isolated directly from the blood of T1DM indivi- peptides/epitopes points to the understanding that mycobacterial duals spontaneously secrete pro-inflammatory cytokines such as Hsp65 has antigenic peptides that could bind/interact with IL-1b and IL-6 thereby inducing the expansion of Th17 cells that immune cells and induce secretion of cytokines. Given this, the are mostly involved in the development of autoimmune diseases.34 MAP infection appears to be an important, putative aetiological

2938 Mol. BioSyst., 2013, 9, 2932--2941 This journal is c The Royal Society of Chemistry 2013 View Article Online Paper Molecular BioSystems Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM.

Fig. 5 Cytokine analysis of culture supernatants of human PBMCs upon treatment with MAP Hsp65 peptides. The culture supernatants of human PBMCs treated with different concentrations of MAP Hsp65 peptides and LPS were assayed using the multiplex cytokine bead array kit. The assay indicated increased secretion of IL-1b, IL-6, IL-8, IL-10 and TNF-a in a dose dependent manner at 24 h. Student’s t test analysis of the cytokine concentrations observed in three independent experiments revealed p o 0.05; this indicated a significant increase in the secretion of cytokines upon treatment with different concentrations of individual peptides of MAP Hsp65 and LPS (positive control) in comparison to untreated cells (added 1Â PBS; the solvent used for solubilizing the peptides).

factor in the pathogenesis of diabetes mellitus wherein the triggers of type-1 diabetes9,11,38 that operate through elusive MAP antigens such as Hsp65 can accelerate autoimmune mechanisms. Advancing this hypothesis further on, we demon- destruction or cytotoxicity of pancreatic islet cells11 and thereby strate that the obviously conserved regions of some of the human hasten the clinical progression of T1DM. In T2DM, however, autoproteins (Hsp60/GAD65) and mycobacterial Hsp65 could con- the exact role of MAP antigens will be discerned only based on stitute critical hotspots of cross-reactivity of mycobacterial antigens future mechanistic evidence. with diabetes sera. Taking the example of above proteins, we espoused a detailed scenario entailing all such components that participate in the initiation and progression of autoimmune 4. Epitope homology from a system’s signalling leading to diabetes mellitus(Fig.6).Thisscenariomight perspective as well hold true for all other possible cross-reacting microbial proteins described earlier and their human homologues (such as Recent studies have suggested the role of mycobacterial pathogens MAP3865c cross-reacting with ZnT8 antigen).38 Hsps are conserved in diabetes mellitus, wherein epitope homologies of pancreatic intracellular proteins that are released to the extracellular space antigens and mycobacterial proteins are proposed as underlying during mycobacterial infections. Further, human Hsp60 is also

This journal is c The Royal Society of Chemistry 2013 Mol. BioSyst., 2013, 9, 2932--2941 2939 View Article Online Molecular BioSystems Paper Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM. Fig. 6 Overview of the proposed molecular pathology of T1DM with special reference to epitope homology and consequent molecular mimicry: during infection, MAP is endocytosed by the intestinal macrophages and the antigenic peptides presented to the immune cells via class I and class II MHC molecules. The conserved and

non-conserved peptides originating from MAP Hsp65 could be presented to immune cells (TH and Tc cells). The activated immune cells secrete cytokines leading to an immune response and inducing differentiation of B-cells to plasma cells that secrete antibodies (Ab). The immune cells and antibodies specific to the conserved peptides of Hsp65 cross-react with the autoantigens of pancreas due to shared epitope homologies with human autoproteins (GAD65 and human Hsp60) possibly leading to cytotoxicity (right lower panel). On the other hand, the immune cells and antibodies triggered or activated by the non-conserved peptide will not be able to cross-react with the autoproteins due to the lack of shared sequences/epitope homology (left lower panel). All the conserved peptides and the corresponding immune cells and antibodies triggered by them are colored in teal while the non-conserved peptides and the corresponding elements are depicted in red.

released due to unknown mechanisms in both diseased and nature and the molecular basis of diabetes mellitus as a multi- healthy individuals. When these cryptic proteins are accumulated, factorial disease with several interacting factors cannot be they could be cross-presented to the immune cells resulting in explained from the analysis of any single component. There- autoimmunity. Our study based on patient sera samples partly fore, autoimmunity as a putative mechanism of diabetes from explainstheaboveespousal.Moreconcreteproofofmolecular the perspective of mycobacterial triggers has to be studied mimicry at the base of diabetes could emerge from future studies through a holistic approach that combines the systems level by developing monoclonal antibodies against the cross-reacting understanding of mycobacterial infections and diabetes while peptides and harnessing suitable animal models. Until then, also integrating their systems epidemiology.39 Given this, we molecular cross-reactivity as deduced from our work could be strongly believe that our observations in identifying the putative considered as a plausible mechanism of (autoimmune) diabetes mechanism of molecular mimicry (epitope homology) of Hsp65 in genetically susceptible hosts. Having this said, future efforts are (which is conserved in most organisms) and human Hsp60, indeed required because autoimmune diseases are complex in based on computational, clinical and immunological data,

2940 Mol. BioSyst., 2013, 9, 2932--2941 This journal is c The Royal Society of Chemistry 2013 View Article Online Paper Molecular BioSystems

constitute some of the baseline components of the ‘system’ 16 J. Panchapakesan, M. Daglis and P. Gatenby, Immunol. Cell comprising molecular pathways that lead to autoimmunity. Out Biol., 1992, 70(pt 5), 295–300. of these components and pathways, a few would potentially 17 W. N. Jarjour, Arthritis Rheum., 1991, 34, 1133–1138. serve as interventional targets to achieve reduction of circulating 18 C. Georgopoulos and H. McFarland, Immunol. Today, 1993, autoantibodies thereby mitigating the impact of diabetes in a 14, 373–375. given patient. 19 S. Tasneem, N. Islam and R. Ali, Microbiol. Immunol., 2001, 45, 841–846. Acknowledgements 20 A. Roy, A. Kucukural and Y. Zhang, Nat. Protocols, 2010, 5, 725–738. Our profuse thanks are due to Prof. Leonardo Sechi for his help 21 M. Delarue and P. Dumas, Proc. Natl. Acad. Sci. U. S. A., and discussion during the planning and progress of this study. 2004, 101, 6957–6962. We would like to acknowledge help from Vynika Goud and 22 R. A. Laskowski, J. A. Rullmannn, M. W. MacArthur, M. Abid Hussain during the collection of sera samples. Work R. Kaptein and J. M. Thornton, J. Biomol. NMR, 1996, 8, in Ahmed lab is funded by grants from the Department of 477–486. Biotechnology (DBT) and the Council for Scientific and Indus- 23 The PyMOL Molecular Graphics System, Version 1.2r3pre, trial Research (OSDD program) of the Indian Government. Schro¨dinger, LLC. NA is an Academy Professor of the Academy of Scientific and 24 B. Petersen, T. N. Petersen, P. Andersen, M. Nielsen and Innovative Research (AcSIR), New Delhi, India. C. Lundegaard, BMC Struct. Biol., 2009, 9, 51. 25 H. Bausinger, D. Lipsker, U. Ziylan, S. Manie, J. P. Briand, Notes and references J. P. Cazenave, S. Muller, J. F. Haeuw, C. Ravanat, H. de la Salle and D. Hanau, Eur. J. Immunol., 2002, 32, 3708–3713. 1 M. T. Rowe and I. R. Grant, Lett. Appl. Microbiol., 2006, 42, 26 A. Boyum, Tissue Antigens, 1974, 4, 269–274. 305–311. 27 Y. Zhang, BMC Bioinf., 2008, 9, 40. 2 S. A. Naser, G. Ghobrial, C. Romero and J. F. Valentine, 28 G. N. Ramachandran, C. Ramakrishnan and V. Sasisekharan, Lancet, 2004, 364, 1039–1044. J. Mol. Biol., 1963, 7,95–99. 3 E. Pierce, Gut Pathog., 2009, 1, 17. 29 J. Yuan, P. Dunn and R. D. Martinus, Cell Stress Chaperones, 4 J. L. Ellingson, J. L. Anderson, J. J. Koziczkowski, R. P. 2011, 16, 689–693. Radcliff, S. J. Sloan, S. E. Allen and N. M. Sullivan, J. Food 30 A. G. van Halteren, B. O. Roep, S. Gregori, A. Cooke, Prot., 2005, 68, 966–972. W. van Eden, G. Kraal and M. H. Wauben, J. Autoimmun., 5 The TRIGR study [http://www.trigr.org]. 2002, 18, 139–147. 6 H. C. Gerstein, Diabetes Care, 1994, 17, 13–19. 31 C. Karopoulos, M. J. Rowley, C. J. Handley and 7 V. Mohan, S. Sandeep, R. Deepa, B. Shah and C. Varghese, R. A. Strugnell, J. Autoimmun., 1995, 8, 235–248. Indian J. Med. Res., 2007, 125, 217–230. 32 E. Sahiratmadja, B. Alisjahbana, T. de Boer, I. Adnan, 8 R. Sicree, J. Shaw and P. Zimmet, Diabetes Atlas Intl Diabetes A. Maya, H. Danusantoso, R. H. H. Nelwan, S. Marzuki, Fed., 2006, 3e, 15–103. J. W. M. vander Meer, R. van Crevel, E. van de Vosse and Published on 30 August 2013. Downloaded by Tripura University 1/10/2020 9:17:48 AM. 9 P. S. Rani, L. A. Sechi and N. Ahmed, Gut Pathog., 2010, 2,1. T. H. M. Ottenhoff, Infect. Immun., 2007, 75, 820–829. 10 J. Tuomilehto, J. Lindstrom, J. G. Eriksson, T. T. Valle, 33 P. S. Rani, N. K. Tulsian, L. A. Sechi and N. Ahmed, Gut H. Hamalainen, P. Ilanne-Parikka, S. Keinanen-Kiukaanniemi, Pathog., 2012, 4, 10. M. Laakso, A. Louheranta, M. Rastas, V. Salminen, 34 E. M. Bradshaw, K. Raddassi, W. Elyaman, T. Orban, M. Uusitupa, S. Aunola, Z. Cepaitis, V. Moltchanov, P. A. Gottlieb, S. C. Kent and D. A. Hafler, J. Immunol., M. Hakumaki, M. Mannelin, V. Martikkala, J. Sundvall 2009, 183, 4432–4439. and F. D. P. S. Gr, N. Engl. J. Med., 2001, 344, 1343–1350. 35 D. Zozulin˜ska, A. Majchrzak, M. Sobieska, K. Wiktorowicz 11 S. A. Naser, S. Thanigachalam, C. T. Dow and M. T. Collins, and B. Wierusz-Wysocka, Diabetologia, 1999, 42, 117–118. Gut Pathog., 2013, 5, 14. 36 B. Balasa, A. La Cava, K. Van Gunst, L. Mocnik, 12 D. F. Child, C. P. Williams, R. P. Jones, P. R. Hudson, D. Balakrishna, N. Nguyen, L. Tucker and N. Sarvetnick, M. Jones and C. J. Smith, Diabetic Med., 1995, 12, 595–599. J. Immunol., 2000, 165, 7330–7337. 13 R. J. Ellis and S. M. van der Vies, Annu. Rev. Biochem., 1991, 37 S. E. Borst, Endocrine, 2004, 23, 177–182. 60, 321–347. 38 S. Masala, D. Paccagnini, D. Cossu, V. Brezar, A. Pacifico, 14 J. C. Borges and C. H. Ramos, Protein Pept. Lett., 2005, 12, N. Ahmed, R. Mallone and L. A. Sechi, PLoS One, 2011, 257–261. 6, e26931. 15 J. G. Routsias and A. G. Tzioufas, Ann. N. Y. Acad. Sci., 2006, 39 N. Ahmed and S. E. Hasnain, Tuberculosis, 2011, 91, 1088, 52–64. 407–413.

This journal is c The Royal Society of Chemistry 2013 Mol. BioSyst., 2013, 9, 2932--2941 2941

201 2

Mycobacterium tuberculosis DosR Regulon Gene Rv0079 Encodes a Putative, ‘Dormancy Associated Translation Inhibitor (DATIN)’

Ashutosh Kumar1, Mohammad Majid1, Ralph Kunisch2, Pittu Sandhya Rani1, Insaf A. Qureshi3, Astrid Lewin2, Seyed E. Hasnain4,5,6, Niyaz Ahmed1,7* 1 Pathogen Biology Laboratory, Department of Biotechnology, School of Life Sciences, University of Hyderabad, Hyderabad, India, 2 Robert Koch Institute, Berlin, Germany, 3 Department of Biotechnology, School of Life Sciences, University of Hyderabad, Hyderabad, India, 4 School of Biological Sciences, Indian Institute of Technology, New Delhi, India, 5 Institute of Life Sciences, University of Hyderabad Campus, Hyderabad, India, 6 King Saud University, Riyadh, Saudi Arabia, 7 Institute of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia

Abstract Mycobacterium tuberculosis is a major human pathogen that has evolved survival mechanisms to persist in an immune- competent host under a dormant condition. The regulation of M. tuberculosis metabolism during latent infection is not clearly known. The dormancy survival regulon (DosR regulon) is chiefly responsible for encoding dormancy related functions of M. tuberculosis. We describe functional characterization of an important gene of DosR regulon, Rv0079, which appears to be involved in the regulation of translation through the interaction of its product with bacterial ribosomal subunits. The protein encoded by Rv0079, possibly, has an inhibitory role with respect to protein synthesis, as revealed by our experiments. We performed computational modelling and docking simulation studies involving the protein encoded by Rv0079 followed by in vitro translation and growth curve analysis experiments, involving recombinant E. coli and Bacille Calmette Gue´rin (BCG) strains that overexpressed Rv0079. Our observations concerning the interaction of the protein with the ribosomes are supportive of its role in regulation/inhibition of translation. We propose that the protein encoded by locus Rv0079 is a ‘dormancy associated translation inhibitor’ or DATIN.

Citation: Kumar A, Majid M, Kunisch R, Rani PS, Qureshi IA, et al. (2012) Mycobacterium tuberculosis DosR Regulon Gene Rv0079 Encodes a Putative, ‘Dormancy Associated Translation Inhibitor (DATIN)’. PLoS ONE 7(6): e38709. doi:10.1371/journal.pone.0038709 Editor: Javed N. Agrewala, Institute of Microbial Technology, India Received February 7, 2012; Accepted May 11, 2012; Published June 13, 2012 Copyright: ß 2012 Kumar et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was primarily supported by a Centre of Excellence Grant from the Department of Biotechnology of the Indian Government entitled ‘Multidisciplinary approaches aimed at interventions against Mycobacterium tuberculosis’ [BT/01/C0E/07/02] to NA (component B2b). The authors would like to also acknowledge support from the German Research Foundation (DFG) sponsored international research training group (IRTG) entitled ‘Internationales Graduiertenkolleg - functional molecular infection epidemiology - GRK1673 (Berlin-Hyderabad)’ of which NA is a speaker, and partial support through the University of Malaya High Impact Research (HIR) grant (Ref. UM.C/625/1HIR/MOHE/CHAN-02). SEH is a J. C. Bose National Fellow (India) and a Robert Koch Fellow of the Robert Koch Institute, Berlin. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have read the journal’s policy and have the following conflicts: Niyaz Ahmed is a Section Editor of PLoS ONE. This does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials. * E-mail: [email protected]

Introduction transcription factor, DosR, pointing out that the control of the regulon is related to physiology of respiration in M. tuberculosis [6]. Tuberculosis (TB) is a chronic infectious disease caused by Although many of the constituent genes of DosR regulon encode Mycobacterium tuberculosis which is linked to high morbidity and hypothetical proteins, growing knowledge of the conditions under mortality worldwide. M. tuberculosis can exist in active form or can which these genes are likely up-regulated could lead to their role in remain alive in a dormant state in lungs after forming granuloma adaptation of M. tuberculosis to the host environment [6]. The where it can prolong its persistence without replication [1]. The humoral immune response for DosR regulon encoded antigens is tubercle bacilli in the dormant or latent state may not be affected stronger in latently infected individuals when compared to by antibiotics or the host immune system due to their bare individuals with active infection [7] suggesting that DosR regulon minimum growth [2,3]. Several studies have indicated that, under genes are more likely expressed during latency. Many proteins latency, the granuloma offers a niche with increased concentration encoded by the genes of this regulon are thought to be helpful to of nitric oxide, low oxygen and absence of nutrients. To survive obtain energy from alternative sources of carbon such as under such unfavourable conditions [2–6], M. tuberculosis might glyoxylate metabolism, nitrate reduction and fatty acid metabolism have evolved mechanism(s) whereby it decreases the rate of protein [6,8]. Several of the members of DosR regulon which might be synthesis to conserve its cellular resources. Such mechanisms have significant in understanding dormancy regulation could not be not been clearly deciphered. functionally characterized as yet. Rv0079 is the first member, by M. tuberculosis genome encodes a regulon of 48 constituent genes order [6], in the DosR regulon that has not been functionally called the dormancy survival regulon (DosR regulon) [6]. In- characterized, although it has been computationally shown to hibition of aerobic respiration causes up-regulation of the encode a predicted translation factor [9].

PLoS ONE | www.plosone.org 1 June 2012 | Volume 7 | Issue 6 | e38709 Ribosome Binding Protein Encoded by Rv0079

Here we describe functional characterization of the protein primer. The construct of Rv0079 was then transformed into E.coli encoded by Rv0079. The gene was found to be conserved in all the BL-21 (DE3). The gene locus Mb0082 of M. bovis BCG displays clinical isolates analysed by us, and also in almost all the publicly 100% sequence identity with the gene Rv0079 of M. tuberculosis.An available M. tuberculosis genomes. In silico structural analyses Rv0079 or Mb0082-overexpressing recombinant BCG [BCG predicted the ability of Rv0079 encoded protein (henceforth called (pMV261+Rv0079 or Mb0082)] was generated by amplifying the DATIN) to dock at the 30S ribosomal subunit. Furthermore, we coding sequence (using M. bovis BCG DNA as template) with the identified that DATIN inhibits in vitro protein synthesis from help of primers [Rv0079_FP_HindIII (CCAAGCTTGTG- mRNA templates. Its own synthesis was arrested when over- GAACCGAAACGCAGTCG) & Rv0079_RP_HpaI (CC expressed in E. coli BL21 (DE3) cells, and it decreased the growth GTTAACTCATGCCAGACCGTCGGC)] and inserting the rate of E. coli BL21 (DE3) and BCG upon over-expression. PCR product downstream of the hsp60 promoter in the HindIII/ HpaI sites of the vector pMV261 [19]. After confirmation of the Materials and Methods absence of any mismatches by sequencing, the recombinant plasmid as well as empty vector, pMV261 were introduced by Ethics Statement electroporation into BCG as described before [20]. This work did not involve any animal experiments or the use of patient material or human biological samples. All recombinant Expression and Purification of His- Tagged Recombinant DNA work reported in the manuscript was approved by the institutional biosafety committee of the School of Life Sciences, Protein University of Hyderabad. DATIN was produced by transforming E. coli BL21 (DE3) cells with the construct generated for recombinant protein expression Distribution of the Locus Rv0079 in M. tuberculosis and the transformed colonies were picked up using ampicillin selection. Recombinant E. coli BL21 (DE3) cells were grown in Distribution of Rv0079 gene was analyzed in 10 randomly taken m clinical isolates of different geographical origins. PCR amplifica- Luria-Bertani broth (containing 100 g/ml ampicillin) up to a cell- tion of Rv0079 was carried out using 100 ng of genomic DNA, density (OD600) of 0.4–0.6 and then induced by IPTG (Sigma, 10 pMol of each primers (see sequences in the next sections), USA) followed by incubation at 37uC for 4 hr. The culture was forward and reverse (designed to target the locus), 200 mMol of centrifuged at 6000 rpm and the cell-pellet was lysed in 20 mM each deoxynucleoside triphosphate, and 1 unit of Pfu DNA Tris-HCl and 200 mM NaCl of pH 8.0 (lysis buffer) by sonication. polymerase (Fermentas Inc., USA) in a standard PCR buffer The resultant lysate was centrifuged at 12000 rpm for 45 min at supplied by the manufacturer. Amplification was performed in 4uC and its supernatant was loaded on cobalt based resin a Master Cycler gradient PCR machine (Eppendorf, Germany) (TALONH Metal Affinity Resins, Clontech) to purify His-tagged under the following conditions: an initial denaturation at 95uC for DATIN. The column was washed extensively with washing buffer 10 min was followed by 30 cycles of 94uC for 30 sec, 60uC for (lysis buffer with 10 mM imidazole; pH 8.0) and the over- 45 sec, 72uC for 1 min 30 sec, and a final extension at 72uC for expressed His-tagged protein was eluted using elution buffer (lysis 15 min. PCR products were separated by electrophoresis in 1% buffer with 250 mM imidazole). Further, size exclusion chroma- agarose gel and visualized under UV light. tography was performed using Superose12 10/300 GL column (GE Healthcare Ltd.) in a buffer containing 20 mM Tris-Cl and Computational Analysis and Structure Prediction 300 mM NaCl (pH 8.0) and the protein profile was compared In silico sequence analysis and computational modelling of with protein molecular size standards. The recombinant protein DATIN (Figure 1) was performed using different tools such as was quantified using Bradford’s reagent [21] and the purified DAS [10], ConPred II [11], TOPCONS [12], and PSORTb v3.0 protein was stored at 220uC until further use. [13] to specifically analyze for the hydropathy index, membrane topology and surface probability. Amino acid sequences of In Vivo Translation Inhibition Assay DATIN were submitted for 3-dimensional structure predictions The Rv0079 and HP0023 (encoding isocitrate dehydrogenase at I-TASSER server [14]. Structure was selected on the basis of of Helicobacter pylori) genes were cloned separately into pRSET-A RMSD values and agreement with Ramachandran Plot, available vector for high-level protein expression of cloned genes in E. with PDBsum program [15]. Molecular visualization and general coli. Later, the constructs were used to transform E. coli BL21 analysis were carried out using the program PyMOL [16]. In silico (DE3) cells and the recombinant colonies were selected using docking experiments were performed using PatchDock [17] and ampicillin. The recombinant E. coli grown up to a cell-density then further refined and ranked with FireDock [18]. Crystal (OD 600) of 0.4–0.6 were induced with 0.25 mM IPTG (Sigma, structure of 30S ribosome (PDB ID: 2AVY) consists of RNA USA) and the cell pellet was collected at every 2 hr interval for molecule and several protein molecules, and therefore only the 24 hr. The cell pellet was lysed in 20 mM Tris-HCl and protein molecules (‘receptor’) were taken for docking study with 200 mM NaCl (lysis buffer, pH 8.0) using sonication. The DATIN (‘ligand’) under default complex-type settings. resultant lysate was centrifuged at 12,000 rpm for 45 min at 4uC. Equal amount of protein was loaded on 12.5% SDS- Molecular Cloning and Engineering of Recombinant PAGE gel and western blotting was performed to detect the Bacterial Strains expression level of the desired protein(s). The probing antibodies The construct for recombinant protein expression was gener- comprised of anti-His mouse IgG (primary) and HRP conju- ated by amplifying the Rv0079 gene using forward gated goat anti mouse IgG (secondary) (Santa Cruz Bio- GCCCTCGAGGTGGAACCGAAAC (XhoI) and reverse technology Inc, USA). The signal was detected using VersaDoc CGAAAGCTTGCTCATGCCAGAC (HindIII) primers. Ampli- imaging system (Bio-Rad). fied product and pRSET-A (Invitrogen, CA, USA) vector were double digested by HindIII and XhoI enzymes and kept for In Vitro Translation Inhibition Assay overnight ligation at 16uC. The clone was confirmed by releasing To identify the role of DATIN in protein synthesis inhibition, the insert after restriction digestion and sequencing with T7 coupled in vitro transcription/translation of circular DNA tem-

PLoS ONE | www.plosone.org 2 June 2012 | Volume 7 | Issue 6 | e38709 Ribosome Binding Protein Encoded by Rv0079

Figure 1. Predicted 3D-structure of DATIN and its docking simulation with 30S ribosomal subunit. A. Predicted 3D-structure of DATIN. The protein secondary structure elements were labelled and colored (helices and sheets displayed in red and yellow, respectively). B. Docking study of DATIN with 30S ribosomal proteins using PatchDock and FireDock. The protein chains of 30S ribosome and DATIN presented as electrostatic surface and colored in yellow and green, respectively. doi:10.1371/journal.pone.0038709.g001 plates was carried out by pBESTlucTM containing the firefly product. Electrophoresis was carried out at a constant voltage of luciferase gene and E. coli S30 Extract System (Promega 50 volt in stacking gel and 100 volt in separating gel. Following Corporation, USA). Three sets of coupled in vitro transcription/ electrophoresis, gels were dried, exposed overnight to a phos- translation were carried out. The reaction mixtures contained 2 ml phorscreen (Amersham Biosciences), and scanned with a Typhoon of pBESTlucTM DNA (1 mg/ml), 5 ml of amino acid mixture minus 9410 variable mode imager. methionine, 20 ml of S30 premix without amino acids, 15 mlof S30 extract circular and 1 ml of (35S) methionine (1,200 Ci/mmol Growth Experiments in Broth Culture at 15 m Ci/ml). Two mg of DATIN and 2 mg of BSA were added To see the effect of DATIN on the growth of the bacteria, in the second and third reaction mixtures, respectively. In the first comparison of the growth rates of the strain overexpressing set of reaction mixture, Milli-Q water was added instead of Rv0079 [E. coli BL21(pRSETA+Rv0079)] and the strain which DATIN or BSA. The reaction mixtures were incubated at 37uC does not express Rv0079 [E. coli BL21(pRSETA)] was carried out. for 90 min, then the tubes containing reaction mixtures were The recombinant E. coli BL21(pRSETA+Rv0079) and E. coli placed in ice to stop the reaction. Aliquots (5 ml) of each reaction BL21(pRSETA) were generated by transforming the E. coli were loaded onto 12.5% SDS-PAGE gel to analyze the translation BL21(DE3) cells with a recombinant construct of pRSETA vector

PLoS ONE | www.plosone.org 3 June 2012 | Volume 7 | Issue 6 | e38709 Ribosome Binding Protein Encoded by Rv0079 containing Rv0079 and, an empty pRSETA, respectively, and the types of plates nullifies the probability of any toxicity [22] of recombinant colonies were picked up using ampicillin selection. DATIN towards E. coli BL21 cells. This confirms that decrease in Both the strains were inoculated in Luria Bertani broth containing the growth rate of E. coli expressing Rv0079, that we recorded (see 100 mg/ml ampicillin with an initial optical density (OD600)of later) in broth cultures, was not due to any toxic effect of 0.04 and ODs of the cultures were measured at hourly intervals recombinant heterologous protein, but may be due to its after inoculation. IPTG (Sigma, USA) was added in both the interaction with or impact on the ribosomes (Figure 1) (that cultures when recombinant E. coli were grown up to a cell-density possibly slows down translation rate). To determine/confirm the (OD600) of 0.4–0.6. putative biological function, Rv0079 was over-expressed in E. coli Further, to confirm the role of DATIN in a mycobacterial and purified to homogeneity under native conditions as His-tagged background, an Rv0079-overexpressing recombinant BCG protein. The elution profile of the recombinant protein confirmed [BCG(pMV261+Rv0079)] and BCG containing empty vector monomeric nature of DATIN in solution. The purified protein on BCG(pMV261) were inoculated in Middlebrook 7H9 media a 12.5% polyacrylamide gel showed a single band corresponding supplemented with oleic albumin dextrose catalase and containing to ,29.5 kDa after staining with Coomassie Brilliant Blue dye. 0.05% Tween 80 and 25 mg/ml kanamycin. OD values of the cultures were measured at 4 days intervals after inoculation. DATIN Impacts Protein Synthesis and Bacterial Growth Both DATIN and rHP0023 (an unrelated negative control) Toxicity Assay were produced in E. coli BL21 expression system separately. The To assess the toxic effect, if any, of DATIN on E.coli, a single western blot analysis of rHP0023 and DATIN indicated that colony of E. coli BL21(pRSETA+Rv0079) was inoculated in Luria synthesis of DATIN gradually decreased at durations from 10 hr Bertani broth containing 100 mg/ml ampicillin and kept at 37uC to 24 hr, whereas, synthesis of rHP0023 was consistent even after in shaking incubator for overnight. Diluted overnight grown 24 hr (Figure 2, panel A). This result suggests inhibitory nature of culture was plated on two sets of LB–ampicillin agar plates, one set DATIN towards protein synthesis. In vitro protein synthesis was containing 0.25 mM IPTG and the second one without this carried out by pBESTlucTM containing the firefly luciferase gene as inducer. a template. After completion of the reaction, it was observed that protein synthesis was diminished when 2 mg of DATIN was added (In Vivo) DATIN-ribosome Interaction to the reaction mixture. However, protein synthesis was not The ribosomes were isolated from E. coli BL21(transformed with affected when similar reaction was performed using bovine serum pRSETA+Rv0079) and the E. coli BL21(pRSETA+vector alone). albumin instead of DATIN (Figure 2, panel B). To isolate the ribosomes, sonication of bacterial cells was The influence of DATIN on the growth rates of E. coli was performed and the cells were resuspended in a buffer containing determined by comparing growth curves of broth cultures from E. 20 mM Hepes-KOH of pH 7.8, 10 mM MgCl2,60mMNH4Cl, coli BL21(pRSETA+Rv0079) and E. coli BL21(pRSETA) (Figure 3). 4 mM 2-mercaptoethanol, 1 mg/ml lysozyme and 0.2 mM DATIN decreased the growth rate of the bacilli and also prevented phenylmethylsulfonyl fluoride. Lysates were clarified by centrifu- them from achieving a higher cell density in the stationary growth gation at 10000 rpm for 30 minutes; supernatant was then phase. The OD values obtained with stationary phase cultures centrifuged at 20000 rpm for 1 hr followed by centrifugation from E. coli BL21(pRSETA+Rv0079) were less than the OD values again at 40000 rpm for 4 hr. The pellet thus obtained was obtained with cultures from E. coli BL21(pRSETA) roughly by dissolved in the same buffer (as above) and centrifugation was done a factor of 2 (Figure 3). Also, the E. coli expressing Rv0079 entered for 2 hr at 40000 rpm to achieve pure ribosome fraction. All the into the stationary phase much earlier (at 6 hr) than the E. coli not steps of ribosome isolation were performed at 4uC and the purified expressing Rv0079 (at 10 hr) (Figure 3). This result was further ribosomes were kept at 280uC for further use. Equal amounts of confirmed by comparing growth curves obtained from the broth purified ribosomes were loaded on a 15% SDS-PAGE gel and cultures of recombinant BCG [BCG(pMV261+Rv0079)] over- western blotting was done by probing with antibodies against 6X expressing Rv0079 and the BCG containing empty vector His tag to detect the presence of DATIN bound with the [BCG(pMV261)]. As expected, BCG expressing Rv0079 showed ribosome(s). delayed growth rate compared to the control. The difference was statistically significant when the OD values after mid log phase Results were compared (Figure 4). Characterization of Rv0079 and the Recombinant Protein DATIN Interacts with E. coli Ribosomes (DATIN) The analysis for cytoplasmic or membrane localization using Rv0079 was found intact in all the 10 clinical isolates that we transmembrane prediction tools did not suggest any transmem- tested and in the sequenced genomes of M. tuberculosis F11, M. brane segment in DATIN. Due to unavailability of crystallo- tuberculosis W-148, M. tuberculosis KZN 4207 (DS), M. tuberculosis graphic/solution structure of DATIN, the search for possible KZN 605 (XDR), M. tuberculosis 98-R604 INH-RIF-EM, M. homologs was carried out using several programs. Sequence-based tuberculosis Haarlem, M. tuberculosis C, M. tuberculosis KZN 1435 search methods (BLASTp) did not provide any significant hit but (MDR), M. tuberculosis H37Rv, M. tuberculosis H37Ra and M. bovis sequence search in PDB identified a template with 27% sequence BCG. The gene was found intact and normally located in both the identity. The template used in the analysis of DATIN was ancestral and modern lineages of M. tuberculosis. This tight a bacterial ribosome binding protein (HI0257, PDB ID: 1IMU) conservation of the locus in all the strains representing different from Haemophilus influenzae which showed similar fold at C- geographical origins and clinical backgrounds points to the terminus in spite of low sequence identity between the two functional importance of DATIN for mycobacterial lifestyle. proteins. A total of 5 models were obtained using I-TASSER In the toxicity assay, the average number of colonies of E. coli server and model number 1 was considered the best among them, BL21 (pRSETA+Rv0079) grown on the LB –ampicillin agar plates based on energy considerations. The quality of the structure was were 40.33 and on LB–ampicillin agar plates containing IPTG assessed using Procheck and this displayed less than 2% were 39. Presence of an almost equal number of colonies on both discrepancy from the Ramachandran plot. Secondary structure

PLoS ONE | www.plosone.org 4 June 2012 | Volume 7 | Issue 6 | e38709 Ribosome Binding Protein Encoded by Rv0079

Figure 2. Protein synthesis inhibition by DATIN. A: Inhibition of protein synthesis in vivo. There was a decrease in DATIN production in recombinant E. coli over-expressing Rv0079 gene after 10 hr of incubation. E. coli expressing H. pylori isocitrate dehydrogenase (HP0023) was taken as a control. B: Coupled in vitro transcription/translation of circular DNA templates using E. coli S30 Extract System for Circular DNA. Reaction mixtures contained Milli-Q water (lane 1), DATIN (lane 2) and BSA (lane 3) (an unrelated control). Full-length luciferase migrated at 61 kDa. An apparent internal translation start resulted in a second major gene product of 48 kDa. doi:10.1371/journal.pone.0038709.g002 analysis showed six alpha helices and four beta sheets in the Western blotting to detect any His-tagged DATIN fraction modelled structure as shown in Figure 1(panel A). bound to purified E. coli ribosomes indicated that DATIN also In order to identify possible interacting domains of DATIN with migrated with the ribosomes. This means that DATIN remained 30S ribosomal subunit, PatchDock was employed for unbound bound to extracted ribosomes from E. coli overexpressing Rv0079. protein-protein docking with 30S ribosome as a receptor and The negative control did not reveal any recombinant protein DATIN as a ligand. Approximately 100 predictions were bound to ribosomal fractions (Figure 5). generated using PatchDock and were submitted to FireDock to refine 10 best solutions on the basis of global energy. Possible Discussion binding interface residues were identified using 3D2GO binding site prediction server. Several of the lowest energy docking models As per the World Health Organisation (WHO) estimates, emerging from this exercise placed DATIN on the side of the 30S approximately 1.8 billion people around the world are infected ribosomal subunit. Among ten docked complexes, complex 1 was with M. tuberculosis and a majority of the cases are clinically latent, identified as the plausible one on the basis of minimum energy waiting only to get activated upon immune-suppression [23]. score and binding interface residues. A docking model of 30S Attempts to completely understand physiology and survival ribosomal subunit with DATIN is shown in Figure 1(panel B). We mechanisms of tubercle bacilli entailing latent TB have largely found that helical regions 5 and 6 of DATIN have moved towards remained inconclusive. DosR regulon is one of the potential 30S ribosomal subunit and docked into a groove formed by the genomic co-ordinates in M. tuberculosis that are being aggressively chains G, I and M (Figure 1). There was no indication of covalent studied to understand latency survival. The DosR regulon has bond involvement during docking, however, electrostatic forces been studied with respect to its regulation and expression but little were found to play major role in interaction. is known about specific functions for DosR regulon proteins. Gene/protein sequence analysis and domain comparisons in-

Figure 3. Growth (in broth) of recombinant E. coli expressing DATIN. Broth cultures from E. coli BL21(pRSETA+Rv0079) and E. coli Figure 4. Growth (in broth) of recombinant M. bovis BCG over- BL21(pRSETA) were inoculated into Luria Bertani broth to give an initial expressing DATIN. Broth cultures from BCG (pMV261+Rv0079) and OD (600 nm) of 0.04 and the cultures were incubated for 15 or 24 hr at BCG (pMV261) were inoculated into Middlebrook 7H9 broth to give an 37uC under IPTG induction. The growth curves of the strains were initial OD (600 nm) of 0.02 and the cultures were incubated at 37uCas generated by measurement of the OD at 600 nm. The values represent static cultures. The growth curves of the strains were generated by the mean of three independent shaking cultures with the standard measurement of the OD at 600 nm. The values represent mean of the deviation. three independent ODs/static cultures with standard deviation. doi:10.1371/journal.pone.0038709.g003 doi:10.1371/journal.pone.0038709.g004

PLoS ONE | www.plosone.org 5 June 2012 | Volume 7 | Issue 6 | e38709 Ribosome Binding Protein Encoded by Rv0079

Figure 5. Binding of DATIN to E. coli ribosomes: Western blot detects purified ribosome preparation from E. coli cells (bound to overexpressed, His-tagged DATIN) probed with anti-His antibody (lane 2); E. coli cells containing empty vector (lane 3) served as negative control and purified DATIN (lane 1) was used as a positive control. Protein molecular weight marker (M) to indicate different size standards was run alongside. doi:10.1371/journal.pone.0038709.g005 dicated that DATIN is likely involved in translation [9]. We, in extending this thinking, we generated a further refined and this dossier, provide several pieces of evidence that support a role resolved computational model with full protein sequence, and of DATIN in translation and ribosome regulation. We believe that carried out a docking simulation followed by functional analysis of these data are compelling, accurate and timely even though only DATIN to elucidate its role in latency. We believe that the current a part of the work was conducted in a mycobacterial system. model and, the envisaged interaction of DATIN with the Adaptation to the host niches during latency requires sensing of ribosome(s) (Figure 1) actually strengthen the conceptualization the stress factors and slowing of the metabolic activity in order to of its role in translation and arrest of protein synthesis. Figure 5 save energy. Sensing of the stress factors present in the granuloma reveals experimental evidence strongly suggestive of this possibil- is ensured by the sensor kinases which can autophosphorylate at ity. a conserved histidine [24]. Transfer of the phosphate to the A role as translation factor or an inhibitor would require strong transcription factor DosR activates DosR regulon via transcription interaction with the translation machinery. The computational of 48 constituent genes of the regulon; these are up-regulated analysis of DATIN indicated that it has sequence homology to the during oxygen stress [6,24]. Slowing down of metabolic activity ribosome binding protein of Haemophilus influenzae. Computational and saving of energy can be achieved by reducing the cell division docking studies indicated that DATIN interacts with the domains rate and by storing macromolecular structures and preventing of 30S ribosomal subunit. ProFunc [27], a computational tool to their degradation [25]. An effective reduction/inhibition of identify likely biochemical functions, based on a 3D protein protein synthesis would likely arrest cell division and, the storage structure, revealed that DATIN has similar folds like ribosome of ribosomes in an inactive state would save energy needed to binding or ribosome associated proteins (for example, PDB codes synthesize ribosomal proteins and rRNA, which makes up major 1IMU, 2YWQ, 3KA5 and 3LYV). In view of this, we carried out metabolic activity of bacteria. Storage of ribosomes during latency docking studies predicting DATIN’s interaction with 30S would also ensure their availability for protein synthesis if ribosomal subunit. This prediction points to a role for DATIN reactivation occurs. Both these aims might be achieved by the in stabilisation of 30S and 70S ribosomal subunits/ribosome and/ function of a factor that interacts with ribosomes to control or translation inhibition. To stabilize the translation machinery translation. during stress/unfavourable conditions, certain dormancy related Regulation of DosR regulon genes including DATIN is proteins might be expressed to possibly prevent the dissociation of controlled by the cognate response regulator/transcription factor, the 50S and 30S. Upon attainment of favourable conditions the DosR [26]. The latter binds to conserved consensus binding same ribosomal machinery can be used for protein synthesis. sequences (the dev boxes or dos boxes) present in the upstream Further, there is also the possibility that this protein can work as promoter region of the DosR regulon genes, and thereby activates a translation factor by interacting with 30S ribosomal subunit as gene expression. Dev boxes have also been identified upstream of has been hinted at by our growth experiments performed with Rv0079 [26]. Given this, one of the significant proteins that could E. coli and BCG and the in vitro translation assay (see later). Given potentially express under the conditions present in the granuloma this, we believe that the interaction apparently supports a more would be DATIN. generalized mechanism of inhibition or arrest of translation. In In silico analysis of Rv0079 encoded protein has previously order to investigate this role of DATIN, we tested its influence on shown that it might potentially act as a translation factor [9]; protein synthesis in vivo and in vitro. We overexpressed Rv0079 in E.

PLoS ONE | www.plosone.org 6 June 2012 | Volume 7 | Issue 6 | e38709 Ribosome Binding Protein Encoded by Rv0079 coli and observed that the amount of DATIN produced therein fold up-regulation of the product of Rv0079 [28]. However, it is declined after 10 hr of IPTG induction but the synthesis of an not clear as to which gene/protein is responsible for the delayed unrelated protein, rHP0023 was consistent even after 24 hr growth or the maintenance of a non-replicating bacteriostasis; (Figure 2, panel A). This observation suggests that DATIN is given our findings, it appears quite probable that DATIN could be potentially involved in translation. The function of DATIN in potentially responsible for slowing the bacterial growth. general translation inhibition was also supported by our in vitro Collectively, these results lead to the thinking that DATIN protein synthesis experiments (Figure 2, panel B). While these might play a significant role in latent phase of M. tuberculosis in vitro translation data suggest that DATIN inhibits protein infection by its ribosome binding function to retard bacterial translation, it will be interesting to carry out similar translation multiplication. We hope that this study would form the basis for inhibition studies in a mycobacterial background. future experiments (involving M. tuberculosis and many other host As mentioned above, the adaptation to the conditions in the bacteria) required to unequivocally present a specific or general granuloma also requires a reduction of the bacterial replication inhibitory role of DATIN linked to latency or otherwise. Further, rate. To confirm this, we have grown recombinant E. coli and mechanistic elucidation of ribosome binding that provides more BCG in a series of experiments and it was observed that after mid- direct evidence of protein synthesis arrest, would be necessary. log phase the growth rates of the overexpressing strains were lower than those of the normal strains/controls (Figures 3 and 4). Acknowledgments Furthermore, DATIN expression resulted in an early entry to the stationary phase of E. coli which in addition was characterized by We would like to thank Elisabeth Kamal for excellent technical assistance a much lower cell density. Figure 3 shows in E. coli that induction and Sven Maurischat for providing materials and help. We acknowledge of Rv0079 expression by IPTG stops growth after 6 hr while this is support from University of Malaya (HIRG, reference UM.C/625/1/HIR/ not the case with a vector that does not contain Rv0079. This is an MOHE/CHAN-02 [A000002-5000 1]) – Molecular Genetics. We would interesting and suggestive experiment that has become significant- like to convey our acknowledgement of help received from Syed Asif Hasan and Priyadarshini Yerra in carrying out additional confirmatory growth ly stronger when at least one other protein (mycobacterial Hsp65) curve experiments involving DATIN and Hsp65. was expressed in the system to show that hyper-expression of any protein does not result in stasis (data not shown). Also, the Author Contributions observed growth inhibition was surely not caused by toxicity of the protein, as we have shown by a toxicity plate assay. Conceived and designed the experiments: NA. Performed the experiments: Recently, it has been reported that overexpression of hypoxia AK MM RK IAQ PSR. Analyzed the data: NA AK AL SEH. Contributed response regulator, DosR, leads to about 2 fold or greater reagents/materials/analysis tools: AL SEH IAQ. Wrote the paper: NA AK transcription of 38 out of the 48 genes of DosR regulon, with a 10 AL.

References 1. Wayne LG (1994) Dormancy of Mycobacterium tuberculosis and latency of 16. Delano, W.L. 2002. The PyMOL molecular graphics system. Delano Scientific, disease. Eur J Clin Microbiol Infect Dis 13: 908–914. San Carlos, CA. http://www.pymol.org/. 2. Harries AD, Dye C (2006) Tuberculosis. Ann Trop Med Parasitol 100: 415–431. 17. Schneidman-Duhovny D, Inbar Y, Nussinov R, Wolfson HJ (2005) PatchDock 3. Wayne LG, Sohaskey CD (2001) Nonreplicating persistence of mycobacterium and SymmDock: servers for rigid and symmetric docking. Nucleic Acids Res 33: tuberculosis. Annu Rev Microbiol 55: 139–163. W363–367. 4. Flynn JL, Chan J (2001) Tuberculosis: latency and reactivation. Infect Immun 18. Mashiach E, Schneidman-Duhovny D, Andrusier N, Nussinov R, Wolfson HJ 69: 4195–4201. (2008) FireDock: a web server for fast interaction refinement in molecular 5. Betts JC, Lukey PT, Robb LC, McAdam RA, Duncan K (2002) Evaluation of docking. Nucleic Acids Res 36: W229–232. a nutrient starvation model of Mycobacterium tuberculosis persistence by gene 19. Stover CK, de la Cruz VF, Fuerst TR, Burlein JE, Benson LA, et al. (1991) New and protein expression profiling. Mol Microbiol 43: 717–731. use of BCG for recombinant vaccines. Nature 351: 456–460. 6. Voskuil MI, Schnappinger D, Visconti KC, Harrell MI, Dolganov GM, et al. 20. Sharbati-Tehrani S, Meister B, Appel B, and Lewin A (2004) The porin MspA (2003) Inhibition of respiration by nitric oxide induces a Mycobacterium from Mycobacterium smegmatis improves growth of Mycobacterium bovis tuberculosis dormancy program. J Exp Med 198: 705–713. BCG. International journal of medical microbiology 294: 235–245. 7. Leyten EM, Lin MY, Franken KL, Friggen AH, Prins C, et al. (2006) Human T- 21. Bradford MM (1976) A rapid and sensitive method for the quantitation of cell responses to 25 novel antigens encoded by genes of the dormancy regulon of microgram quantities of protein utilizing the principle of protein-dye binding. Mycobacterium tuberculosis. Microbes Infect 8: 2052–2060. Anal Biochem 72: 248–254. 8. Wayne LG, Lin KY (1982) Glyoxylate metabolism and adaptation of Mycobacterium tuberculosis to survival under anaerobic conditions. Infect 22. Dumon-Seignovert L, Cariot G, Vuillard L (2004) The toxicity of recombinant Immun 37: 1042–1049. proteins in Escherichia coli: a comparison of overexpression in BL21(DE3), 9. Mishra S (2009) Function prediction of Rv0079, a hypothetical Mycobacterium C41(DE3), and C43(DE3). Protein Expr Purif 37: 203–206. tuberculosis DosR regulon protein. J Biomol Struct Dyn 27: 283–292. 23. Frieden TR, Sterling TR, Munsiff SS, Watt CJ, Dye C (2003) Tuberculosis. 10. Cserzo M, Eisenhaber F, Eisenhaber B, Simon I (2002) On filtering false positive Lancet 362: 887–899. transmembrane protein predictions. Protein Eng 15: 745–752. 24. Roberts DM, Liao RP, Wisedchaisri G, Hol WG, Sherman DR (2004) Two 11. Arai M, Mitsuke H, Ikeda M, Xia JX, Kikuchi T, et al. (2004) ConPred II: sensor kinases contribute to the hypoxic response of Mycobacterium a consensus prediction method for obtaining transmembrane topology models tuberculosis. J Biol Chem 279: 23082–23087. with high reliability. Nucleic Acids Res 32: W390–393. 25. El-Sharoud WM (2004) Ribosome inactivation for preservation: concepts and 12. Bernsel A, Viklund H, Hennerdal A, Elofsson A (2009) TOPCONS: consensus reservations. Science Progress 87: 137–152. prediction of membrane protein topology. Nucleic Acids Res 37: W465–468. 26. Chauhan S, Sharma D, Singh A, Surolia A, Tyagi JS (2011) Comprehensive 13. Yu NY, Wagner JR, Laird MR, Melli G, Rey S, et al. (2010) PSORTb 3.0: insights into Mycobacterium tuberculosis DevR (DosR) regulon activation improved protein subcellular localization prediction with refined localization switch. Nucleic Acids Res 39: 7400–7414. subcategories and predictive capabilities for all prokaryotes. Bioinformatics 26: 27. Laskowski RA, Watson JD, Thornton JM (2005) ProFunc: a server for predicting 1608–1615. protein function from 3D structure. Nucleic Acids Res 33: W89–93. 14. Roy A, Kucukural A, Zhang Y (2010) I-TASSER: a unified platform for 28. Flores Valdez MA, Schoolnik GK (2010) DosR-regulon genes induction in automated protein structure and function prediction. Nat Protoc 5: 725–738. Mycobacterium bovis BCG under aerobic conditions. Tuberculosis (Edinb) 90: 15. Laskowski RA, Chistyakov VV, Thornton JM (2005) PDBsum more: New 197–200. summaries and analyses of the known 3D structures of proteins and nucleic acids. Nucleic Acids Res 33: D266–268.

PLoS ONE | www.plosone.org 7 June 2012 | Volume 7 | Issue 6 | e38709 Khattak et al. BMC Microbiology 2012, 12:204 http://www.biomedcentral.com/1471-2180/12/204

RESEARCH ARTICLE Open Access Illegitimate recombination: An efficient method for random mutagenesis in Mycobacterium avium subsp. hominissuis Faisal Asghar Khattak1,2, Ashutosh Kumar1,3, Elisabeth Kamal1, Ralph Kunisch1 and Astrid Lewin1*

Abstract Background: The genus Mycobacterium (M.) comprises highly pathogenic bacteria such as M. tuberculosis as well as environmental opportunistic bacteria called non-tuberculous mycobacteria (NTM). While the incidence of tuberculosis is declining in the developed world, infection rates by NTM are increasing. NTM are ubiquitous and have been isolated from soil, natural water sources, tap water, biofilms, aerosols, dust and sawdust. Lung infections as well as lymphadenitis are most often caused by M. avium subsp. hominissuis (MAH), which is considered to be among the clinically most important NTM. Only few virulence genes from M. avium have been defined among other things due to difficulties in generating M. avium mutants. More efforts in developing new methods for mutagenesis of M. avium and identification of virulence-associated genes are therefore needed. Results: We developed a random mutagenesis method based on illegitimate recombination and integration of a Hygromycin-resistance marker. Screening for mutations possibly affecting virulence was performed by monitoring of pH resistance, colony morphology, cytokine induction in infected macrophages and intracellular persistence. Out of 50 randomly chosen Hygromycin-resistant colonies, four revealed to be affected in virulence-related traits. The mutated genes were MAV_4334 (nitroreductase family protein), MAV_5106 (phosphoenolpyruvate carboxykinase), MAV_1778 (GTP-binding protein LepA) and MAV_3128 (lysyl-tRNA synthetase LysS). Conclusions: We established a random mutagenesis method for MAH that can be easily carried out and combined it with a set of phenotypic screening methods for the identification of virulence-associated mutants. By this method, four new MAH genes were identified that may be involved in virulence. Keywords: Mycobacterium, Mycobacterium avium subsp. hominissuis, Non-tuberculous mycobacteria, Virulence, Mutagenesis, Illegitimate recombination

Background lung diseases, lymphadenitis and disseminated disease The genus Mycobacterium (M.) comprises highly patho- mostly in immuno-compromised persons [5]. Lung genic bacteria such as M. tuberculosis as well as environ- infections as well as lymphadenitis are most often caused mental opportunistic bacteria called NTM. They are by M. avium [5,6], and M. avium is considered to be ubiquitous and have been isolated from soil, natural among the clinically most important NTM [7]. water sources, tap water, biofilms, aerosols, dust and M. avium can be divided into four subspecies. M. sawdust [1-3]. Remarkably, NTM are resistant to avium subsp. paratuberculosis (MAP) causes the Johne’s amoeba and protected against adverse conditions inside disease in ruminants; M. avium subsp. avium (MAA) amoebal cysts [4]. While the incidence of tuberculosis is and M. avium subsp. silvaticum infect birds; and finally declining in the developed world, infection rates by M. avium subsp. hominissuis (MAH) which causes dis- NTM are increasing [5]. NTM cause skin infections, ease in humans [8]. The main route of infection in AIDS patients is the in- * Correspondence: [email protected] vasion of mucosal epithelial cells of the gastrointestinal 1Robert Koch-Institute, Division 16 Mycology/Parasitology/Intracellular Pathogens, Nordufer 20, Berlin 13353, Germany tract, while in non-AIDS patients infections mainly Full list of author information is available at the end of the article occur through the respiratory route [9]. Recognition of

© 2012 Khattak et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Khattak et al. BMC Microbiology 2012, 12:204 Page 2 of 14 http://www.biomedcentral.com/1471-2180/12/204

M. avium by mouse macrophages involves binding of a of illegitimate recombination of a linear recombination 20 – 25 kDa lipoprotein from the cell envelope of M. substrate for random mutagenesis of MAH. avium to TLR2. This interaction leads to bacteriostasis of M. avium in a MyD88-dependent way [10]. Even Methods though the expression of TNF-α is also induced via Bacterial strains, amoeba, cell lines and growth conditions TLR2-signalling, its role in growth restriction of M. Mycobacterial strains were grown in Middlebrook (MB) avium is unclear [10]. IFN-γ is considered to be a key 7H9 broth (BD Biosciences, USA), supplemented with cytokine for killing of M. avium and its expression is either 10% ADC (BD Biosciences) or 10% OADC (BD promoted by IL-18 secreted by M. avium-infected Biosciences) and 0.05% Tween 80 without shaking, and human macrophages [11]. Phagocytosis of M. avium is on MB 7H11 agar (BD Biosciences) at 37°C. Escherichia supposed to be mediated via binding of the bacteria to a coli DH5α was used as host strain for plasmid pYUB854, variety of receptors including complement receptors a cosmid vector with a Hygromycin resistance (Hygr) CR1, CR2, CR3, CR4, the mannosyl-fucosyl-receptor, the gene [31] and was cultured in/on Luria-Bertani broth fibronectin receptor, the integrin receptor α(v)β3, and and agar at 37°C. Antibiotics when required were added the transferrin receptor [12-15]. M. avium inhibits the at the following concentrations: Kanamycin (50 μgml-1) acidification of the phagosome and the fusion of the or Hygromycin (50 μgml-1). For Congo Red plating agar phagosome with lysosomes [16,17]. media was supplemented with 100 μgml-1 Congo Red. Intracellular M. avium survives antibacterial activities The Acanthamoeba castellanii strain 1BU group II [32] such as nitric oxide and reactive oxygen species and the was cultivated in PYG medium (Proteose peptone-Yeast mechanisms leading to killing of M. avium are still un- extract-Glucose [33]) at 28°C and passaged once per known [18]. The cell wall structure is an important fac- week. The human macrophage cell line THP-1 (DSMZ- tor determining virulence of M. avium [19]. Thus, No. ACC-16, DSMZ GmbH, Braunschweig, Germany) different colony morphotypes (smooth opaque, smooth was maintained by passaging twice weekly in RPMI 1640 W transparent, rough) distinguishable on Congo Red plates (GIBCO Invitrogen, Darmstadt, Germany) supplemen- display different degrees of virulence. Smooth transpar- ted with 10% foetal bovine serum (Bio Whittaker, Walk- ent and rough colonies are considered to be more viru- ersville, MD, USA). Cells were cultured in BD FalconTM 2 lent than smooth opaque colonies [20,21]. The colony 75 cm trays (BD Biosciences) at 37°C and in 5% CO2. morphotype is associated with the glycopeptidolipid For human macrophages infection and washing, Iscove's (GPL) composition [19]. By inducing the release of vari- Modified Dulbecco's Media (IMDM) (PAA laboratories, ous pro-inflammatory cytokines such as IL-1, IL-6 or Austria) with 3% Human AB-serum (PAA laboratories) TNF-α, GPL modulate the immune response against was used. mycobacteria [22]. Only relatively few virulence genes from MAH have Molecular biology techniques been defined with respect to their role in infection. This All molecular biology techniques were carried out accord- is partly attributable to difficulties in generating MAH ing to standard protocols [34] or according to the recom- mutants. The major obstacle is the low transformation mendations of the manufacturers of kits and enzymes. frequency if MAH is used as recipient. This also limits Primers were purchased from Metabion (Martinsried, the efficiency of so far described random mutagenesis Germany). Plasmid DNA was isolated with the QIAGEN systems, such as the commercially available EZ- Plasmid Mini Prep Kit (Qiagen, Hilden, Germany). Poly- TN < KAN2 > Tnp Transposome from Epicentre. This merase chain reaction (PCR) was performed with the Tn903-based system consists of a stable complex formed DreamTaq Kit from Fermentas (St. Leon-Rot, Germany). between the EZ::TN Transposase enzyme and the EZ:: Restriction enzymes were purchased from Fermentas. For TN < KAN-2 > Transposon. It was used in MAA and elution of DNA fragments from agarose gels, the MAH to analyse mechanisms of multidrug resistance QIAquick Gel Extraction kit (Qiagen) was used. Ligation and the role of GPL [23-25]. Another system for the reactions were performed with the T4 DNA Ligase Kit generation of random mutants is based on transduction from Fermentas. Genomic DNA from mycobacteria was using temperature-sensitive phages containing a trans- isolated according to the protocol described in Sjöbring poson with a selection marker [26,27]. In other myco- et al. [35]. Sequencing reactions were performed by using bacterial species such as M. tuberculosis and M. bovis the Prism Big Dye FS Terminator Cycle Sequencing Ready BCG linear recombination substrates have been applied Reaction Kit from PE Applied Biosystems (Darmstadt, to generate random as well as site-directed mutants Germany). Nucleotide sequence analysis was performed [28-30]. This approach, however, so far has not been using the software packages MacVector™ 7.2.3 (Accelrys, published for mutagenesis of MAH or MAA. With the Cambridge, UK) and Lasergene (DNASTAR, Inc., present study we intended to explore the performance Madison, WI, USA). Khattak et al. BMC Microbiology 2012, 12:204 Page 3 of 14 http://www.biomedcentral.com/1471-2180/12/204

For Southern blotting 2 μg of genomic DNA from were used to transcribe cDNA from the RNA. The Mycobacterium were digested with ApaI or SmaI, sepa- cDNA was then used to perform real time PCR with the rated by electrophoresis in a 1% agarose gel and capillary MaximaTM SYBR Green/Rox qPCR Master Mix 2x from transferred to positively charged nylon membranes (GE Fermentas. Primers were (i) for gene MAV_1779: 5´- Healthcare, Buckinghamshire, UK) by following a stand- CTG CAG AAG AGC GTC TAC CC-3´ and 5´-CTC ard protocol [34]. An 1818 bp region of the plasmid TGT TCG GAG GTC GTC AT-3´, (ii) for gene pYUB854 carrying the Hygr gene was amplified using MAV_3129: 5´-GGT CAA GAC CAT CGA CGA CT-3´ the primer pair Hyg2K FW (5´-CAC CGT ACG TCT and 5´-AGA TCA TGA ACG GCA CAA CA-3´, (iii) for CGA GGA ATT CCT G-3´) and Hyg2K BW (5´-GCG gene MAV_4332: 5´-ATG GTC GAG CAG AGC ATC TCG TGA AGA AGG TGT TGC TGA-3´) and the TGG-3´ and 5´-ATG GCG TCC ACG AAC CG-3´, (iv) digoxigenin labeling Kit (Roche, Mannheim Germany). for gene MAV_5105: 5´-GTT GTC GAG TTC ACC The labeled PCR-product was used as a probe and de- GGT CT-3´ and 5´-ATT CAC TCG GCG AAT ACC tection was carried out using anti-digoxigenin-AP conju- TG-3´ and (v) for 16S rRNA gene: 5´-GAG TGA GAA gate and CDP-star (Roche) according to the TGC AGG CAT GA-3´ and 5´-ACA CGG GTA CGG manufacturers’ instructions. GAA TAT CA-3´. The ΔΔCT method was used to calcu- Reverse PCR was applied to exactly locate the inser- late the relative expression of the gene of interest in the tion sites of the Hygr gene in the mutants. 2 μg of DNA mutant in comparison to the mean of its expression in of each mutant was digested with the restriction enzyme the other three mutants. Normalisation was obtained by ApaI or SmaI (which do not cut in the recombination measuring the expression of 16S rRNA gene as reference substrate). The multiple sized DNA fragments were gene. ethanol precipitated and then self-ligated by T4 DNA ligase enzyme, thus resulting in different sized circular Random mutagenesis by illegitimate recombination DNA molecules. A PCR was then performed with pri- 1 μg of plasmid pYUB854 DNA was double digested mers [Hyg mut_1 (5´-AAC TGG CGC AGT TCC TCT with restriction enzymes StuI and SpeI Fast digest at G-3´) and Hyg mut_2 (5´-TCA GCA ACA CCT TCT 37°C for 30 min. The 2030 bp linear DNA fragment TCA CGA-3´)] binding within the Hygr gene and carrying the Hygr gene was gel-eluted after electrophor- oriented towards the unknown genomic MAH DNA esis and 3–6 μg linear DNA fragment was transformed located adjacent to the Hgyr gene. Sequencing of the into M. avium strains by electroporation with the PCR products using the primers Hyg mut_1 and Hyg Biorad GenePulser apparatus applying 1000 Ω,25μF mut_2 followed by BLAST analysis of the sequences and 1.25 kV in 1 mm gap cuvettes. The preparation of allowed the exact identification of the insertion sites of electrocompetent cells and electroporation were per- the recombination substrates. formed using standard protocols [36]. Plasmid pMN437 For quantitative RT-PCR the mutants were grown in was used as positive control for transformation [37]. -1 MB/ADC with 25 μgml of Hygromycin B to an OD600 Electroporated bacteria were incubated at 37°C for of 2. The pellet of 10 ml of culture was resuspended in 24 hours (h) before plating on selective plates. Potential 4 ml of protoplasting buffer (15 mM of Tris–HCl pH 8, mutants were characterised by PCR amplifying a part of 0.45 M of Sucrose, 8 mM of EDTA) with 4 mg ml-1 the Hygr gene [primers Hyg 2 K LC FW (5´-AGT TCC Lysozyme. After incubation at 37°C for 45 minutes TCC GGA TCG GTG AA-3´) and Hyg 2 K LC BW (min) the protoplasts were harvested by centrifugation (5´-AGG TCG TCC CGG AAC TGC TGC G-3´)], and the pellets were resuspended in 1050 μl of the RLT Southern blotting, reverse PCR (primers Hyg mut 1 buffer from the RNeasy Minikit (Qiagen) with 10.5 μlof and Hyg mut 2) and sequencing. ß-Mercaptoethanol. This suspension was transferred into tubes containing 25–50 mg of glass beads (0.5 mm, Construction of a complemented derivative of mutant PeqLab, Erlangen, Germany) and shaken in the MAV_3128 homogenizer Precellys 24 (PeqLab) for 45 sec at 6,500 g. Primers MAV3128_MV306_1 (5´-CGG TCT AGA CTA The tubes were chilled on ice and centrifuged at 8,000 g TGC CTA CCT GCT CTC-3´) and MAV3128_MV306_2 for 5 min at 4°C. Then, 0.7 volume of absolute Ethanol (5´-GCA GTT AAC CTA ATG CGG CTT GGC CAG- was added to the supernatant and this solution was dis- 3´) were designed to amplify the gene MAV_3128 tributed onto two columns of the RNeasy Kit. The sam- (3227 bp) plus 680 bp of upstream sequence of the wild ples were further processed as described in the RNeasy type with pfu polymerase from Fermentas. The amplified manual. Residual DNA present in the RNA preparations product was cloned into the restriction sites XbaI and was removed with the Kit Desoxyribonuclease I (DNaseI) HpaI respectively of the integrative vector pMV306 [38]. RNase free from Fermentas. The M-MLV Reverse Tran- The recombinant plasmid pFKaMAV3128 was trans- scriptase and Random primers from Promega (WI, USA) formed into E. coli DH5α by a method already described Khattak et al. BMC Microbiology 2012, 12:204 Page 4 of 14 http://www.biomedcentral.com/1471-2180/12/204

by Hanahan [39]. The plasmid pFKaMAV3128 was then totalof200,000cellsperwellofTHP-1weregrownalong introduced into competent cells of mutant MAV_3128 by with addition of phorbol-12-myristate-13-acetate (PMA, electroporation. PCR analyses with the primer pair Sigma, Taufkirchen, Germany) (10 ng ml-1) and allowed to MAV3128_MV306_1 and 2 confirmed the presence of adhere to the surface of the plate well overnight at 37°C and wild type gene in the mutant MAV_3128. in 5% CO2. Cells were then infected with mutants and WT at a multiplicity of infection (MOI) of 50 colony forming Screening for virulence-mutants units (CFU). The supernatants were removed after 24 h and Amoeba Plate Test (APT) cytokines were quantified in appropriate dilutions of the The APT was previously described [40]. In short, known supernatants by ELISA using the Human ELISA Ready to go concentrations of Acanthamoeba castellanii (1BU group Kits (Natutec, Frankfurt, Germany). II strain) diluted in PYG medium were spread on MB agar plates and these plates served as test plates. For Intracellular survival in THP-1 cells control plates only PYG medium without amoeba was THP-1 cells were seeded, treated with PMA and infected spread on MB agar plates. Plates were dried and incu- as described above. The supernatants were removed after bated at room temperature. The next day series of ten- 4 h infection period and adherent cells were washed twice fold dilution (1:10, 1:100, and 1:1000) in sterile water with RPMI 1640. The cells were then treated with were prepared from cultures of the mutants and the M. 200 μgml-1 of Amikacin (Sigma) for 2 h to kill the myco- avium 104 wild type (WT). 3 μl of undiluted culture and bacteria in the supernatant. After washing twice with PBS of each dilution were spotted onto the test and control buffer (10 mM sodium phosphate, 126 mM sodium chlor- agar plates. Plates were then incubated at 30°C for one ide, pH 7.2), 1 ml of medium supplemented with 5 μgml-1 week. Mutants showing reduced growth on test plates of Amikacin was added to each well. Samples for quantifi- compared to the control plates were selected for further cation of intracellular bacteria were taken at the end of the molecular characterisation. infection time after removal and killing of extracellular bacteria and then after 1, 2, and 4 days. For this, the cells Growth rate in broth cultures under pH stress were lysed in 1 ml of water at 37°C for 20 min and the The growth rates of mutants and WT were compared in mycobacterial DNA in the lysates was quantified by real- MB 7H9 broth with neutral pH (7) and under pH stress time PCR as described in Lewin et al. [41]. Additionally, 3 (pH 5). Cultures were inoculated to an initial OD600 of 100 μl of 1:10 dilution in sterile water of samples were 0.02 to 0.03 and allowed to grow for two weeks. Three cul- plated in triplicate on agar plates supplemented with ADC tures per strain were inoculated. Growth of cultures was for counting of CFU. determined by measurement of OD600 of cultures and also by quantification of ATP with the luminescence-based Kit Intracellular survival in human monocytes BacTiter-GloTM Microbial Cell Viability Assay (Promega). Human monocytes were isolated from buffy coats from The luminescence was recorded as relative light units healthy donors using Ficoll-PaqueTM Plus (GE Health- (RLU) with the microplate luminometer LB96V (EG & G care) and PercollTM (GE Healthcare) gradient centrifuga- Berthold). Mutants showing differences of growth pattern tion according to the manufacturer’s recommendations compared to the WT in both neutral medium and under and as described in Sharbati et al. [42]. One million pH stress conditions were considered for further molecu- macrophages were seeded per well in 24-well cell culture lar characterisation. plates, with three to five wells per sample per sampling point. Infection with mutants, complemented strain and Congo Red plating WT, Amikacin treatment and sampling were done as 100 μl of 1:105 and 1:106 dilutions in sterile water of described above for THP-1 cells infection, except that mutants, complemented strain and WT were spread in human monocytes were pre-activated with 100 U ml-1 of triplicate on MB agar plates supplemented with OADC human IFN-γ (Invitrogen, Darmstadt, Germany) and and 100 μgml-1 Congo Red. Plates were incubated for 10 ng ml-1 of LPS (Sigma), IMDM was used for washing, 2–3 weeks and observed for colony morphology. the MOI for infection was 10 and the dilution of the Mutants showing differences in colony morphology samples for plating and counting of CFU was 1:500. (white vs. red staining, transparent vs. opaque colonies, smooth vs. rough colonies) compared to the WT were Results and discussion considered for further molecular characterisation. Generation and genetic characterisation of M. avium mutants Induction of cytokine expression in THP-1 cells Our aims were the establishment of a new method to Infection of the cell line THP-1 was performed in 24-well cell mutagenise MAH and the identification of mutants po- culture plates (TPP) with three to five wells per sample. A tentially affected in virulence. The mutagenesis approach Khattak et al. BMC Microbiology 2012, 12:204 Page 5 of 14 http://www.biomedcentral.com/1471-2180/12/204

involved transformation of a recombination substrate by Hygr gene in the genome. Additionally, Southern blots electroporation into MAH, and we therefore first identi- using a PCR fragment produced with primer pair Hyg2K fied clinical and environmental MAH strains applicable FW and BW as probe were performed to verify if the to electroporation. We considered a prior investigation insertions had occurred at different genome sites in dif- of transformability to be necessary, because other ferent colonies (data not shown). Hybridising bands authors had reported some clinical M. avium strains to were obtained with the DNA from 20 colonies and con- be inaccessible to electroporation [43]. As proposed by firmed independent insertion events. Lee et al. [43], we chose a gfp-containing plasmid Inverse-PCR using the primers Hyg mut 1 and Hyg (pGFP: gfp cloned in vector pMV261 [38]) for trans- mut 2 followed by sequencing of the PCR products formation assays. We tested 14 clinical isolates and two enabled us to identify the sites of insertion of the Hygr soil isolates. Strain M. avium 104 was originally isolated gene in 13 mutants. As shown in Figure 1, there were no from an HIV patient [44] and strains 2721/04, 10091/06, hot spots for integration but the insertions were distrib- 10203/06, 4557/08, 4023/08, 3646/08, 3449/08, 3269/08, uted within the whole M. avium genome. 2630/08, 2014/08, 772/08, 709/08, 528/08 were isolated The genetic characterisation of four virulence- from children with lymphadenitis. Strains 128 and 129 associated mutants is shown in Figure 2. The integration are soil isolates. Out of these 16 M. avium strains, five events were accompanied by deletions in all 13 mutants. (104, 2721/04, 2014/08, 4023/08 and 528/08) could be The smallest deletion had a size of 2 bp, the largest one transformed with pGFP. As the genome sequence from of 669 bp. All insertions were located within coding M. avium strain 104 is available in the genome data regions. Only in one mutant more than one gene was bases, simplifying a precise mutant description, we affected by the insertion. In 12 of the 13 mutants the lin- decided to concentrate on this strain for further analysis. ear recombination substrate had been completely Our mutagenesis approach took advantage of the high inserted and in one mutant the inserted fragment had rate of illegitimate recombination in slow growing myco- been shortened at both ends. The sequences next to the bacteria [28,45] and their ability to take up linear DNA inserted fragment showed no special structure or nu- [29]. For selection purposes we chose the Hygr gene in- cleotide sequences. stead of also often used Kanamycin resistance gene Both, the random distribution of insertion sites and (Kmr), because the Hygr gene had been shown before to the low rate of large deletions affecting more than one be superior to the Kmr gene especially for the transform- gene are benefits of our method. Contrary to our experi- ation of other than laboratory strains [46]. The Hygr ence with MAH, Collins and colleagues [49] observed gene used for electroporation was flanked by plasmid more clustered insertions and deletions of up to 12 DNA of 793 bp on one side and 238 bp on the other side. These flanking regions served as substrates for the illegitimate recombination. After electroporation of 3–6 μg of restriction fragment and selection on plates containing Hygromycin, about 1000 colonies could be obtained. Around ten percent of this number appeared on the negative control plates (no DNA added) and had to be considered spontaneously resistant. The positive controls (with 1–2 μg plasmid DNA) gener- ated around 5–6 times more colonies than could be observed on the test plates. Transposon/transduction mutagenesis procedures have been reported to deliver around 1,000 to 3,500 mutants per mutagenesis pro- cedure [19,23,24,27,47,48] which means that the effi- ciency or our method was below the efficiency of transposon/transduction systems. Taking into account the simple handling of our method we consider it nevertheless to be a good alternative to the currently applied methods for mutagenesis of MAH. Fifty randomly chosen colonies from the sample plates were tested for insertion of the Hygr gene by performing a PCR using the primers Hyg 2 K LC FW and Hyg 2 K Figure 1 Sketch showing randomly mutated genes distributed LC BW (data not shown). By this PCR 49 of the 50 col- within the M. avium genome. Genes location mapped on the genome after sequencing. onies could be confirmed to carry an insertion of the Khattak et al. BMC Microbiology 2012, 12:204 Page 6 of 14 http://www.biomedcentral.com/1471-2180/12/204

Figure 2 Sketch illustrating the genetic characterisation of the mutants MAV_1778, MAV_3128, MAV_4334, and MAV_5106. The sites of the insertion of the marker (Hygr gene) were identified by inverse PCR followed by sequencing of the eluted PCR products. The figure shows for four mutants the mutated gene (dark blue) with the site of insertion of the fragment (grey) carrying the Hygr gene (red) and the four genes located upstream and downstream of the mutated gene (light blue). Numbers in the arrows indicate the gene names. The direction of the arrows stands for gene direction. Gene sizes and distances between genes are approximations. Below the map of each mutant the size of the deletion generated as result of insertion of the marker is indicated. genes by mutagenising M. bovis with a DNA fragment macrophages and (v) intracellular survival and growth in carrying a Kanamycin resistance gene by illegitimate re- human macrophages. combination. It would be interesting to find out the rea- sons for these differing outcomes. Are the specific Colony morphology and Congo Red staining characteristics parameters of the illegitimate recombination events The occurrence of different colony morphotypes is an species-specific or does the composition of the recom- eye-catching feature of M. avium including MAH and bination substrate play a more important role? has attracted attention also because it is associated to In favor of a straight forward procedure, we concen- virulence [19,24,50,51]. The colony morphology is influ- trated our further efforts on those mutants, which ful- enced by the composition of the cell wall, which is a filled the following requirements: - an insertion in the major determinant of mycobacterial virulence [52-54]. middle of the coding region of a gene, – mutation of Congo Red, a planar hydrophobic molecule can bind to only one gene and - mutation of a single copy gene. diverse lipids and lipoproteins and is thus applicable for After applying these criteria, eight mutants (see Table 1 the detection of changes in cell wall composition [54-56]. for mutated genes and their functions) were selected for Upon plating of MAH on Congo Red agar plates, smooth phenotypic analysis. Table 1 Mutated M. avium genes and their functions Phenotypic characterisation of MAH mutants Mutated Gene Function of the gene Since virulence is regulated on many different levels we MAV_2555 Short-chain dehydrogenase/reductase SDR applied more than one screening test (as for example MAV_1888 Hypothetical protein intracellular multiplication) to identify a greater spectrum of relevant virulence-associated genes. We MAV_4334 Nitroreductase family protein searched for phenotypic assays allowing a fast screening MAV_5106 Phosphoenolpyruvate carboxykinase of our mutants and not requiring special and expensive MAV_1778 GTP-Binding protein LepA equipment. The selected tests should monitor changes MAV_3128 Lysl-tRNA synthetase (LysS) in (i) cell wall composition (plating on Congo Red Agar), MAV_3625 Hypothetical protein (ii) resistance towards low pH, (iii) amoeba resistance, MAV_2599 Hypothetical protein (iv) induction of cytokine secretion by infected Khattak et al. BMC Microbiology 2012, 12:204 Page 7 of 14 http://www.biomedcentral.com/1471-2180/12/204

transparent, smooth opaque and rough colonies as well The colonies generated by mutant MAV_3128 (Figure 3 as red and white colonies can be distinguished. Appear- G) were in average larger in size and the smooth-opaque ance of the three morphotypes smooth opaque, smooth colonies appeared paler than in the WT. Also, the edges transparent and rough was also confirmed for strain 104 of these colonies were more irregular. Some red-rough [21]. While the opaque-transparent switch is reversible, colonies were also visible. The most multifaceted image the rough phenotype results from irreversible deletion of was displayed by mutant MAV_3625. This strain gener- cell envelope glycopeptidolipid genes and is irreversible ated smooth-domed-opaque, smooth-domed-red (sdr), [24,51]. smooth-flat-red, smooth-transparent and rough-trans- TLC (Thin Layer Chromatography) analysis of the dif- parent (rt) and red-rough colonies (Figure 3 H). A ferent morphotypes from strain 104 has been performed high proportion of red colonies (smooth-domed-red, by Torelles [21]. They also analysed the sugar compos- smooth-flat-red, red-rough) was generated by mutant ition of the glycopeptidolipids (GPL) by gas chromatog- MAV_2599 (Figure 3 I) additionally to smooth-opaque raphy–mass spectrometry (GC–MS) analysis. They and smooth-transparent colonies. This mutant pro- found that the smooth opaque and smooth transparent duced only few rough (rough-transparent, rough-red) colonies formed similar GPL and both expressed besides colonies. Altogether, we observed a high frequency and the nsGPL (ns: non-specific) the ssGPL (ss:serovar spe- intensity of morphological changes in the mutants cific) of serovar 1. However, the ssGPL was absent in the pointing to involvement of the mutated genes in the rough morphotype, which had a strong band of the composition of cell wall structure. Since studies by dif- nsGPL. A band in the lipopeptid region devoid of sugars ferent authors have related colony morphotype to viru- was present in the smooth transparent morphotype and lence it would be of interest to investigate in further the rough morphotype but lacking in the smooth opaque experiments if and to which degree the different col- morphotype. The sugar composition of all morphotypes ony types are stable and differ in their virulence. showed the typical profiles related to ns and ssGPL of serovar 1, only in the rough morphotype 6-deoxytalose pH-resistance and 3-O-methyl-6-deoxytalose were missing. The intraphagosomal pH of M. avium-containing pha- The transparent colony variant grows better in macro- gosomes decreases to pH 5.2 in activated macrophages phages and animals compared to the opaque variant. More- [59]. We therefore investigated the pH-resistance of the over, white transparent colonies survived better in mutants compared to the WT by inoculating them into macrophages than red transparent colonies [19,24,50,51,56]. MB broth at pH 5 and pH 7 and measuring the growth These differences in intracellular survival may be caused by during 11 days at 37°C by means of OD measurement variations in the cytokine response towards infection by dif- and ATP quantification. ATP measurement represents a ferent morphotypes. The smooth opaque morphotype has much more sensitive method than the OD measure- been shown to induce higher levels of secretion of IL-1α,IL- ment. Additionally, the OD of a culture not only 1β and TNF-α by human blood-derived monocytes com- depends on cell number but also on the size of the cells, pared to the smooth-transparent morphotype [57]. Variation their morphology and the degree of clumping of the in cytokine response upon infection with either smooth- cells. For these reasons, ATP measurement was reported opaque or smooth-transparent M. avium was also reported to be a more reliable method for quantification of myco- upon infection of human microglia cultures [58]. bacteria in broth culture [41]. As shown in Figure 4, the The colony morphology of the WT and the mutants WT grew better at neutral pH than at low pH. After upon plating on Congo Red Agar is shown in Figure 3. 11 days of growth in neutral medium, it generated The WT (Figure 3 A) mainly formed smooth-domed- 722,491 RLU (relative light units), while in medium with opaque (sdo) colonies along with smooth-transparent acidic pH only 143,082 RLU were achieved. The mutants (st) colonies. Mutant MAV_2555 showed the same MAV_2555, MAV_1888, MAV_4334 and MAV_5106 morphologies, but additionally smooth-flat-red (sfr) col- showed a similar growth pattern as the WT, both in onies were visible (Figure 3 B). Relatively few smooth- neutral and acidic pH (data not shown). The mutants transparent and rough colonies occurred in mutant MAV_1778 and MAV_3128 grew similar as the WT at MAV_1888 (Figure 3 C), MAV_4334 (Figure 3 D) and neutral pH; however, at low pH these strains enhanced MAV_5106 (Figure 3 E). Mutant MAV_4334 (Figure 3 their growth rate even above the level reached at neutral D) showed a higher variation with respect to the inten- pH (Figure 4 A and B). While the mutant MAV_3128 sity of red color of smooth-domed-opaque colonies. Mu- showed enhanced growth in comparison to the WT at tant MAV_1778 showed a very high degree of variability low pH already at day 1, the mutant MAV_1778 showed displaying red-rough (rr) and smooth-flat-red colonies an identical growth rate as the WT at low pH until day 5 additionally to the smooth-domed-opaque, smooth- and then showed strongly enhanced growth. The transparent and rough-white (rw) colonies (Figure 3 F). mutants MAV_3625 and MAV2599 grew better than the Khattak et al. BMC Microbiology 2012, 12:204 Page 8 of 14 http://www.biomedcentral.com/1471-2180/12/204

Figure 3 Colony morphology upon plating on Congo Red agar plates. Well-grown broth cultures of all strains were diluted 1:106 and 100 μl plated in triplicate onto Middlebrook agar with OADC containing 100 μgml-1 Congo Red. Plates were incubated on average for three weeks. The arrows point to smooth-domed-opaque (sdo), smooth-flat-red (sfr), smooth transparent (st), rough red (rr) and rough transparent (rt) colonies. A: WT; B: mutant MAV_2555; C: mutant MAV_1888; D: mutant MAV_4334; E: mutant MAV_5106; F: mutant MAV_1778; G: mutant MAV_3128; H: mutant MAV_3625; I: mutant MAV_2599.

Figure 4 Resistance towards pH stress. The bacteria were grown in Middlebrook 7H9 broth with OADC at pH 7 and pH 5 during 11 days; the ATP content was recorded by quantification of the amount of ATP in the cultures. The amount of ATP is represented as RLU (relative light units). A: WT and mutant MAV_1778; B: WT and mutant MAV_3128; C: WT and mutant MAV_3625; D: WT and mutant MAV_2599. Khattak et al. BMC Microbiology 2012, 12:204 Page 9 of 14 http://www.biomedcentral.com/1471-2180/12/204

WT at pH 7 and were able to maintain this growth rate The signaling cascade is mainly initiated by binding of M. at pH 5 (Figure 4 C and D). In summary, the mutations avium components to TLR2 followed by recruitment of either had no influence on the survival under pH stress the MyD88 adaptor molecule and the activation of NFκB conditions or improved resistance towards pH stress. and MAP kinases. This chain of events ends with the in- duction of inflammatory cytokines [10] controlling macro- Amoeba plating test phage activation and granuloma formation. We monitored Free-living amoebae are known to host environmental the induction of cytokine expression of THP-1 macro- mycobacteria including M. avium, which are able to phages by the WT and the mutants in order to evaluate survive in Acanthamoeba trophozoites as well as in their ability to stimulate the immune signaling. To this the exocysts [4,60,61]. Growth in Acanthamoeba was aim we quantified the secretion of selected cytokines: the associated with subsequently enhanced virulence in in- pro-inflammatory cytokines TNF-α,IL-1β and the anti- fection experiments with mice [62]. Since some viru- inflammatory cytokine IL-10. Five independent experiments lence mechanisms are employed by amoeba-resistant were normalised for WT (expression ratio 1) to determine bacteria to survive in amoebae as well as in macro- the expression ratio for the mutants in comparison to WT. phages [4,63-65], amoebae have been used as test sys- While results for TNF-α and IL-1β were not significantly tems for determination of bacterial virulence factors different as compared to WT, IL-10 was significantly (P [40,63,66]. An Acanthamoeba castellanii agar plate <0.007) up-regulated for mutant MAV_4334 (Figure 5). IL- assay was developed and successfully employed for 10 can inhibit the production of inflammatory cytokines screening of mutants of Legionella pneumophila [40]. such as TNF-α in monocytes pre-activated by IFN-γ and We adapted this APT to fit the growth conditions LPS [67,68] and therefore plays an important role in the im- (medium, temperature, duration) of M. avium and tested mune response. the eight mutants in comparison to the WT. After incu- bation for five to seven days at 28°C, the WT formed col- Intracellular survival onies even if the cultures were diluted 1:103 before being The ability to survive and even replicate inside the pha- dropped on the lawn of amoebae. The growth of some gosomes of macrophages is an important virulence fac- mutants was more strongly affected by the amoebae but tor of mycobacteria and was therefore included in our a differentiated evaluation of the impact of the various screening options. Infection experiments with macro- mutations on survival in the amoebae was not possible phages give information on the early host response to (data not shown). The APT thus was not sensitive mycobacterial infections [69]. Different types of macro- enough to reveal differences in the capacity of the phages or monocytic cells have been employed to assess mutants to survive within the amoebae. This was surpris- mycobacterial virulence and among these the human ing, because the APT has proven to be an efficient tool macrophage-like cell line THP-1 has proven a suitable for the identification of virulence genes in L. pneumophi- lae [40]. There are several possible explanations for this discrepancy. Amoebae are the most important habitat of Legionella, while M. avium is not dependent on the pres- ence of amoebae for survival and distribution. As a con- sequence, Legionella might have evolved more important virulence factors interacting with amoebae. Another pos- sible explanation may result from the differences in the generation times of L. pneumophilae and M. avium. L. pneumophilae is a fast-growing bacterium forming clearly visible colonies few days after plating, while the slow-growing M. avium 104 requires two weeks to gen- erate colonies of comparable size. This time span may be too long to maintain the amoebae as trophozoites ac- tively interacting with the mycobacteria. In conclusion, we estimate the APT to be of only little value for the de- tection of virulence genes of slow-growing mycobacteria. Figure 5 Induction of IL-10 cytokine secretion by infected macrophages. THP-1 cells (2.0x105) were infected (MOI 50) with Induction of cytokine secretion mutants and WT. After 24 hours cytokines from supernatants were The innate immune recognition by phagocytic cells med- measured by ELISA. When compared to WT a P value <0.01 – iates cellular activation enabling killing of the bacteria and (two-tailed, unpaired Mann Whitney test) was considered very significant (**). the production of pro- and anti-inflammatory cytokines. Khattak et al. BMC Microbiology 2012, 12:204 Page 10 of 14 http://www.biomedcentral.com/1471-2180/12/204

system for virulence testing [69,70]. It was shown that MAV_5106 to survive better in macrophages suggests THP-1 cells are similar to primary human monocyte- that it may be characterised by a higher virulence as derived macrophages with respect to their ability to take compared to the other mutants. Tateish et al. [70] com- up mycobacteria and limit their growth [71]. We pared the virulence of different M. avium isolates in infected THP-1 cells that had been differentiated by humans, immuno-competent mice and THP-1 cells. PMA with the WT and the mutants. Intracellular myco- They found that the strain causing the most serious bacteria were measured by quantitative real-time PCR disease in humans and the highest bacterial load in and CFU by plating. Survival of mutants in THP-1 cells mouse lungs also grew better in THP-1 cells than the was not consistently different if compared to the WT other strains tested. According to this, the mutants (data not shown). More significant differences were MAV_4334, MAV_1778 and MAV_3128 may display obtained when using human blood monocytes for the in- reduced virulence and the corresponding genes may rep- fection experiments. The growth of mutant MAV_4334, resent virulence-associated genes. MAV_1778 and MAV_3128 was affected the most in human monocytes (Figure 6). They were reduced signifi- Evaluation of the screening procedure cantly for the first two days (P < 0.05 to P < 0.01). Mutant We have employed five screening methods (colony morph- MAV_4334 and MAV_1778 (Figure 6 A and C) were al- ology, pH stress resistance, amoeba resistance, cytokine in- most reduced to half during the first two days. As shown duction, intracellular survival) to select mutants affected in in Figure 6 D, mutant MAV_3128 had the highest sig- virulence-related traits. Two mutants (MAV_4334 and nificant (P < 0.001) difference in growth as compared to MAV_3128) responded differently from the WT in four of WT, which had survived better during this time period. these five screening tests and two mutants (MAV_5106 and The mutant MAV_5106 largely differed from other MAV_1778) reacted differently in three screening tests. mutants and during four days of infection had shown The most prominent differences were exhibited by mutant constant survival (Figure 6 B). The capacity of mutant MAV_3128. The other mutants either did not show any

Figure 6 Intracellular survival of mutants compared to WT in human monocytes. Human blood monocytes (1.0x106) from healthy volunteers were infected (MOI 10) with mutants and WT. Intracellular bacteria were quantified after 4 hour of infection, and after 1, 2, & 4 days. The monocytes were lysed in 1 ml of sterile water and 100 μl of 1:500 dilution in sterile water of sample were plated on Middlebrook agar plates supplemented with ADC for CFU counting. A: WT and mutant MAV_4334; B: WT and mutant MAV_5106; C: WT and mutant MAV_1778; D:WT and mutant MAV_3128. Statistical analysis was done using a two tailed, paired Student’s t test. When compared to wild-type a P < 0.05 was considered significant (*) and a P < 0.01 very significant (**). Khattak et al. BMC Microbiology 2012, 12:204 Page 11 of 14 http://www.biomedcentral.com/1471-2180/12/204

differences compared to the WT or reacted differently in The lysS gene from M. avium is 81% homologous to only one or two tests. the lysX gene from M. tuberculosis. LysX from M. tuber- The insertions in mutants MAV_4334, MAV_5106, culosis is required for synthesis of lysinylated phosphati- MAV_1778 and MAV_3128 have been mapped and the dylglycerol. A LysX mutant was shown to be sensitive to structure of the mutated regions has been analyzed on cationic antibiotics and peptides, to be more lysosome- nucleotide level. In all cases only one gene has been associated and to display defective growth in mouse and mutagenised. The insertions are located in the genes guinea pig lungs [76]. MAV_4334 (nitrogenase reductase family), MAV_5106 So far, nothing is known about the role of the nitro- (phosphoenolpyruvate carboxykinase), MAV_1778 (GTP- genase reductase family protein for growth and patho- binding protein LepA) and MAV_3128 (lysyl-tRNA genicity of mycobacteria and answering this question synthestase LysS). will be one of our future aims. Phosphoenolpyruvate carboxykinases (PEPCK) catalyse In summary, by analysing 50 random mutants, we the reversible decarboxylation and phosphorylation of uncovered four genes from MAH to play a role in the oxaloacetate to form phosphoenolpyruvate. Mutations of interaction with host cells and thus in virulence. The the PEPCK gene from M. bovis BCG are characterised homologues of three of the four genes were shown to by attenuated virulence and reduced survival in macro- contribute to virulence in other bacterial species, which phages [72]. The PEPCK gene from M. tuberculosis was supports the significance of our screening procedure. shown to be required for replication in murine bone marrow macrophages and mice [73]. Mutant complementation and evaluation of polar The LepA protein from M. tuberculosis possess down-stream effects GTPase activity. Bacterial GTP-binding proteins play a To prove that the phenotypes of the mutants were in- role in regulation of ribosomal function and cell cycle, deed a cause of the inactivation of the mutated genes, modulation of DNA partitioning and DNA segregation we aimed at complementing the mutants by introducing [74]. In Helicobacter pylori LepA is important for growth the intact genes by electroporation. Only the transfer of at low pH and may play a role in infection [75]. gene MAV_3128 into the respective mutant was

Figure 7 Phenotype of the complemented strain MAV3128Comp compared to mutant MAV_3128 and WT. A: Colony morphology on Congo Red plates. B: Intracellular survival in human blood monocytes. Khattak et al. BMC Microbiology 2012, 12:204 Page 12 of 14 http://www.biomedcentral.com/1471-2180/12/204

successful. Mutant MAV_3128 had shown the strongest same direction was not or only slightly changed. Only in and most different phenotypic changes in comparison to mutant MAV_1778 a two-fold expression of gene wild-type among the eight tested mutants in almost all MAV_1779 was observed. We conclude that with one the phenotypic tests. A complementation is best per- exception no relevant polar effects could be observed. formed if the copy number of gene transcripts generated by the complementing gene narrows the copy number in Conclusions the wild-type. We therefore used a plasmid for cloning Our study proposes a well-functioning method to ran- (pMV306) that integrates once in the genome of the mu- domly mutagenise MAH, by illegitimate recombination, tant and included the upstream region of MAV_3128 to genetically characterise the mutations to the nucleotide most likely cover the promoter of the gene. This up- level and screen the mutants with simple phenotypic stream region had a size of about 680 bp and the gene tests providing information about virulence-associated MAV_3127, which is located upstream of MAV_3128, features. has an orientation in opposite direction of MAV_3128 (see Figure 2). Therefore it was expected that the up- Competing interests The authors declare that they have no competing interests. stream region will contain the promoter sequence of the MAV_3128 gene. Thus a 3907 bp DNA fragment was Authors’ contributions cloned into the integrative vector pMV306. The result- Conceived and designed the study: FAK and AL. Carried out the Laboratory ing recombinant plasmid pFKaMAV3128 was success- work: FAK, AK, EK and RK. Manuscript drafted: FAK and AL. All authors read and approved the final manuscript. fully transformed into the mutant MAV_3128 to generate the complemented strain MAV3128Comp. Acknowledgements Selected phenotypic tests (plating on Congo Red Agar We thank Dr. Elvira Richter, from National Reference Center for Mycobacteria, Borstel, Germany for generously providing 14 M. avium clinical isolates and and intracellular survival) were repeated with the comple- Dr. Petra Möbius, from Friedrich Löffler Institute, Jena, Germany for giving mented strain. Upon plating on Congo Red agar (Figure 7 2 M. avium environmental strains. We also thank Prof. Dr. Michael Niederweis, A), the pale colour of mutant MAV_3128 could no longer University of Alabama, Birmingham, USA for donating plasmid pMN437. be seen in MAV3128Comp,exceptsomepalecornersin Author details colonies. This may indicate the loss of the plasmid in ab- 1Robert Koch-Institute, Division 16 Mycology/Parasitology/Intracellular sence of selection pressure. The intracellular survival ex- Pathogens, Nordufer 20, Berlin 13353, Germany. 2Faculty of Biological Sciences, Islamia College Peshawar (a public sector University), periment has also conclusively indicated a reversal of the KhyberPakhtunkhwa, Pakistan. 3Pathogen Biology Laboratory, Department of mutation. The complemented strain showed more similar Biotechnology, School of Life Sciences, Unversity of Hyderabad, Hyderabad, growth tendency towards wild-type strain than towards India. the mutant (Figure 7 B). In conclusion we successfully Received: 17 August 2012 Accepted: 7 September 2012 complemented the mutant MAV_3128 by introducing the Published: 11 September 2012 intact gene proving that the phenotype of mutant MAV_3128 was indeed caused by the inactivation of gene References 1. Kirschner RA Jr, Parker BC, Falkinham III JO: Epidemiology of infection by MAV_3128 and not by a second line mutation. nontuberculous mycobacteria: Mycobacterium avium, Mycobacterium Since introduction of the intact genes into the other intracellulare, and Mycobacterium scrofulaceum in acid, brown-water three mutants failed we additionally investigated the oc- swamps of the Southeastern United States and their association with environmental variables. Am Rev Respir Dis 1992, 145:271–275. currence of polar effects in the four mutants by quanti- 2. Matlova L, Dvorska L, Palecek K, Maurenc L, Bartos M, Pavlik I: Impact of tative RT-PCR. As polar effects most probably will have sawdust and wood shavings in bedding on pig tuberculous lesions in an impact on genes which are located downstream of lymph nodes, and IS1245 RFLP analysis of Mycobacterium avium subsp. hominissuis of serotypes 6 and 8 isolated from pigs and environment. the mutated gene and exhibit the same orientation, we Vet Microbiol 2004, 102:227–236. quantified expression of genes MAV_1779 (in mutant 3. van Ingen J, Boeree MJ, Dekhuijzen PNR, van Soolingen D: Environmental MAV_1778), MAV_3129 (in mutant MAV_3128), sources of rapid growing nontuberculous mycobacteria causing disease in humans. Clin Microbiol Infect 2009, 15:888–893. MAV_4332 (in mutant MAV_4334) and MAV_5105 (in 4. Salah IB, Ghigo E, Drancourt M: Free-living amoebae, a training field for mutant MAV_5106) by qRT-PCR. The 16S rRNA gene macrophage resistance of mycobacteria. Clin Microbiol Infect 2009, 15:894–905. was used as reference gene. The ΔΔCT method was used 5. McGrath EE, McCabe J, Anderson PB: Guidelines on the diagnosis and to calculate expression of the gene in the corresponding treatment of pulmonary non-tuberculous mycobacteria infection. Int J mutant compared to the mean expression in the other Clin Pract 2008, 62:1947–1955. three mutants. The expression levels measured were: 6. Cassidy PM, Hedberg K, Saulson A, McNelly E, Winthrop KL: Nontuberculous mycobacterial disease prevalence and risk factors: A changing MAV_1779 (in mutant MAV_1778): 2.1 fold, MAV_3129 epidemiology. Clin Infect Dis 2009, 49:e124–e129. (in mutant MAV_3128): 1.1 fold, MAV_4332 (in mutant 7. Alvarez-Uria G: Lung disease caused by nontuberculous mycobacteria. MAV_4334): 1.0 fold and MAV_5105 (in mutant Current Opinion in Pulmonary Medicine 2010, 16:251–256. 8. Mijs W, de Haas P, Rossau R, Van Der Laan T, Rigouts L, Portaels F, van MAV_5106): 1.4 fold. In three of the four mutants, the Soolingen D: Molecular evidence to support a proposal to reserve the expression of the down-stream genes transcribed in the designation Mycobacterium avium subsp. avium for bird-type isolates Khattak et al. BMC Microbiology 2012, 12:204 Page 13 of 14 http://www.biomedcentral.com/1471-2180/12/204

and 'M. avium subsp. hominissuis' for the human/porcine type of M. 29. Balasubramanian V, Pavelka MS Jr, Bardarov SS, Martin J, Weisbrod TR, avium. Int J Syst Evol Microbiol 2002, 52:1505–1518. McAdam RA, Bloom BR, Jacobs WR Jr: Allelic exchange in Mycobacterium 9. Harriff MJ, Danelishvili L, Wu M, Wilder C, McNamara M, Kent ML, Bermudez tuberculosis with long linear recombination substrates. J Bacteriol 1996, LE: Mycobacterium avium genes MAV-5138 and MAV-3679 are 178:273–279. transcriptional regulators that play a role in invasion of epithelial cells, in 30. Armitige LY, Jagannath C, Wanger AR, Norris SJ: Disruption of the genes part by their regulation of CipA, a putative surface protein interacting encoding antigen 85A and antigen 85B of Mycobacterium tuberculosis with host cell signaling pathways. J Bacteriol 2009, 191:1132–1142. H37Rv: Effect on growth in culture and in macrophages. Infect Immun 10. Salomé Gomes M, Fernandes SS, Cordeiro JV, Gomes SS, Vieira A, Appelberg 2000, 68:767–778. R: Engagement of Toll-like receptor 2 in mouse macrophages infected 31. Bardarov S, Bardarov S Jr, Pavelka MS Jr, Sambandamurthy V, Larsen M, with Mycobacterium avium induces non-oxidative and TNF-independent Tufariello J, Chan J, Hatfull G, Jacobs WR Jr: Specialized transduction: An anti-mycobacterial activity. Eur J Immunol 2008, 38:2180–2189. efficient method for generating marked and unmarked targeted gene 11. Shiratsuchi H, Ellner JJ: Expression of IL-18 by Mycobacterium avium- disruptions in Mycobacterium tuberculosis, M. bovis BCG and M. infected human monocytes; association with M. avium virulence. Clin Exp smegmatis. Microbiology 2002, 148:3007–3017. Immunol 2001, 123:203–209. 32. Walochnik J, Obwaller A, Aspock H: Correlations between morphological, 12. Bermudez LE, Young LS, Enkel H: Interaction of Mycobacterium avium molecular biological, and physiological characteristics in clinical and complex with human macrophages: Roles of membrane receptors and nonclinical isolates of Acanthamoeba spp. Appl Environ Microbiol 2000, serum proteins. Infect Immun 1991, 59:1697–1702. 66:4408–4413. 13. Rao SP, Ogata K, Catanzaro A: Mycobacterium avium-M. intracellulare 33. Visvesvara GS, Balamuth W: Comparative studies on related free-living and binds to the integrin receptor alpha v beta 3 on human monocytes and pathogenic amebae with special reference to Acanthamoeba. JProtozool monocyte-derived macrophages. Infect Immun 1993, 61:663–670. 1975, 22:245–256. 14. Roecklein JA, Swartz RP, Yeager H Jr: Nonopsonic uptake of Mycobacterium 34. Sambrook J: FE, Maniatis T: Molecular Cloning - A Laboratory Manual. 2nd avium complex by human monocytes and alveolar macrophages. Journal edition. New York: Cold Spring Harbor Laboratory Press; 1989. of Laboratory and Clinical Medicine 1992, 119:772–781. 35. Sjobring U, Mecklenburg M, Andersen AB, Miorner H: Polymerase chain 15. Jha S, Danelishvili L, Wagner D, Maser J, Li Y, Moric I, Vogt S, Yamazaki Y, Lai reaction for detection of Mycobacterium tuberculosis. J Clin Microbiol B, Bermudez L: Virulence-related Mycobacterium avium MAV_2928 gene 1990, 28:2200–2204. is associated with vacuole remodeling in macrophages. BMC Microbiol 36. Krzywinska E, Schorey JS: Characterization of genetic differences between 2010, 10:100. Mycobacterium avium subsp. avium strains of diverse virulence with a focus 16. De Chastellier C, Lang T, Thilo L: Phagocytic processing of the on the glycopeptidolipid biosynthesis cluster. Vet Microbiol 2003, 91:249–264. macrophage endoparasite, Mycobacterium avium, in comparison to 37. Steinhauer K, Eschenbacher I, Radischat N, Detsch C, Niederweis M, Goroncy- phagosomes which contain Bacillus subtilis or latex beads. European Bermes P: Rapid evaluation of the Mycobactericidal efficacy of disinfectants Journal of Cell Biology 1995, 68:167–182. in the quantitative carrier test EN 14563 by using fluorescent 17. Oh YK, Straubinger RM: Intracellular fate of Mycobacterium avium: Use of Mycobacterium terrae. Appl Environ Microbiol 2010, 76:546–554. dual-label spectrofluorometry to investigate the influence of bacterial 38. Stover CK, De La Cruz VF, Fuerst TR, Burlein JE, Benson LA, Bennett LT, viability opsonization on phagosomal pH phagosome-lysosome Bansal GP, Young JF, Lee MH, Hatfull GF, et al: New use of BCG for interaction. Infect Immun 1996, 64:319–325. recombinant vaccines. Nature 1991, 351:456–460. 18. Li YJ, Danelishvili L, Wagner D, Petrofsky M, Bermudez LE: Identification of 39. Hanahan D: Studies on transformation of Escherichia coli with plasmids. virulence determinants of Mycobacterium avium that impact on the J Mol Biol 1983, 166:557–580. ability to resist host killing mechanisms. J Med Microbiol 2010, 59:8–16. 40. Albers U, Reus K, Shuman HA, Hilbi H: The amoebae plate test implicates a 19. Laurent JP, Hauge K, Burnside K, Cangelosi G: Mutational analysis of paralogue of lpxB in the interaction of Legionella pneumophila with cell wall biosynthesis in Mycobacterium avium. J Bacteriol 2003, Acanthamoeba castellanii. Microbiology 2005, 151:167–182. 185:5003–5006. 41. Lewin A, Freytag B, Meister B, Sharbati-Tehrani S, Schäfer H, Appel B: Use of 20. Meylan PR, Richman DD, Kornbluth RS: Characterization and growth in a Quantitative TaqMan-PCR for the Fast Quantification of Mycobacteria human macrophages of Mycobacterium avium complex strains isolated in Broth Culture, Eukaryotic Cell Culture and Tissue. Journal of Veterinary from the blood of patients with acquired immunodeficiency syndrome. Medicine Series B: Infectious Diseases and Veterinary Public Health 2003, Infect Immun 1990, 58:2564–2568. 50:505–509. 21. Torrelles JB, Ellis D, Osborne T, Hoefer A, Orme IM, Chatterjee D, Brennan PJ, 42. Sharbati J, Lewin A, Kutz-Lohroff B, Kamal E, Einspanier R, Sharbati S: Cooper AM: Characterization of virulence, colony morphotype and the Integrated microRNA-mRNA-analysis of human monocyte derived glycopeptidolipid of Mycobacterium avium strain 104. Tuberculosis 2002, macrophages upon Mycobacterium avium subsp. hominissuis infection. 82:293–300. PLoS One 2011, 6:e20258. 22. Schorey JS, Sweet L: The mycobacterial glycopeptidolipids: Structure, 43. Lee SH, Cheung M, Irani V, Carroll JD, Inamine JM, Howe WR, Maslow JN: function, and their role in pathogenesis. Glycobiology 2008, 18:832–841. Optimization of electroporation conditions for Mycobacterium avium. 23. Philalay JS, Palermo CO, Hauge KA, Rustad TR, Cangelosi GA: Genes Tuberculosis 2002, 82:167–174. required for intrinsic multidrug resistance in Mycobacterium avium. 44. Horan KL, Freeman R, Weigel K, Semret M, Pfaller S, Covert TC, Van Antimicrob Agents Chemother 2004, 48:3412–3418. Soolingen D, Leão SC, Behr MA, Cangelosi GA: Isolation of the genome 24. Cangelosi GA, Do JS, Freeman R, Bennett JG, Semret M, Behr MA: The two- sequence strain Mycobacterium avium 104 from multiple patients over a component regulatory system mtrAB is required for morphotypic 17-year period. J Clin Microbiol 2006, 44:783–789. multidrug resistance in Mycobacterium avium. Antimicrob Agents 45. Niranjala Muttucumaru DG, Parish T: The molecular biology of Chemother 2006, 50:461–468. recombination in Mycobacteria: What do we know and how can we use 25. Freeman R, Geier H, Weigel KM, Do J, Ford TE, Cangelosi GA: Roles for cell it? Current Issues in Molecular Biology 2004, 6:145–158. wall glycopeptidolipid in surface adherence and planktonic dispersal of 46. Garbe TR, Barathi J, Barnini S, Zhang Y, Abou-Zeid C, Tang D, Mukherjee R, Mycobacterium avium. Appl Environ Microbiol 2006, 72:7554–7558. Young DB: Transformation of mycobacterial species using hygromycin 26. Otero J, Jacobs WR Jr, Glickman MS: Efficient allelic exchange and resistance as selectable marker. Microbiology 1994, 140:133–138. transposon mutagenesis in Mycobacterium avium by specialized 47. Scandurra GM, Young M, de Lisle GW, Collins DM: A bovine transduction. Appl Environ Microbiol 2003, 69:5039–5044. macrophage screening system for identifying attenuated transposon 27. Li Y, Miltner E, Wu M, Petrofsky M, Bermudez LE: A Mycobacterium mutants of Mycobacterium avium subsp. paratuberculosis with vaccine avium PPE gene is associated with the ability of the bacterium to potential. J Microbiol Methods 2009, 77:58–62. grow in macrophages and virulence in mice. Cell Microbiol 2005, 48. Cavaignac SM, White SJ, De Lisle GW, Collins DM: Construction and 7:539–548. screening of Mycobacterium paratuberculosis insertional mutant 28. Kalpana GV, Bloom BR, Jacobs WR Jr: Insertional mutagenesis and libraries. Arch Microbiol 2000, 173:229–231. illegitimate recombination in mycobacteria. Proc Natl Acad Sci U S A 1991, 49. Collins DM, Wilson T, Campbell S, Buddle BM, Wards BJ, Hotter G, de Lisle 88:5433–5437. GW: Production of avirulent mutants of Mycobacterium bovis with Khattak et al. BMC Microbiology 2012, 12:204 Page 14 of 14 http://www.biomedcentral.com/1471-2180/12/204

vaccine properties by the use of illegitimate recombination and macrophage-like cell line THP-1: A comparison with human monocyte- screening of stationary-phase cultures. Microbiology 2002, 148:3019–3027. derived macrophages. Cell Immunol 1999, 197:1–9. 50. Mukherjee S, Petrofsky M, Yaraei K, Bermudez LE, Cangelosi GA: The white 72. Liu K, Yu J, Russell DG: pckA-deficient Mycobacterium bovis BCG shows morphotype of Mycobacterium avium-intracellulare is common in attenuated virulence in mice and in macrophages. Microbiology 2003, infected humans and virulent in infection models. J Infect Dis 2001, 149:1829–1835. 184:1480–1484. 73. Marrero J, Rhee KY, Schnappinger D, Pethe K, Ehrt S: Gluconeogenic 51. Cangelosi GA, Palermo CO, Bermudez LE: Phenotypic consequences of carbon flow of tricarboxylic acid cycle intermediates is critical for red-white colony type variation in Mycobacterium avium. Microbiology Mycobacterium tuberculosis to establish and maintain infection. Proc 2001, 147:527–533. Natl Acad Sci U S A 2010, 107:9819–9824. 52. Belisle JT, Brennan PJ: Chemical basis of rough and smooth variation in 74. Meena LS, Chopra P, Bedwal RS, Singh Y: Cloning and characterization of mycobacteria. J Bacteriol 1989, 171:3465–3470. GTP-binding proteins of Mycobacterium tuberculosis H37Rv. Enzym 53. Collins FM, Cunningham DS: Systemic Mycobacterium kansasii infection and Microb Technol 2008, 42:138–144. regulation of the alloantigenic response. Infect Immun 1981, 32:614–624. 75. Bijlsma JJE, Lie-A-Ling M, Nootenboom IC, Vandenbroucke-Grauls CMJE, 54. Parrish NM, Ko CG, Dick JD, Jones PB, Ellingson JL: Growth, Congo Red Kusters JG: Identification of loci essential for the growth of Helicobacter agar colony morphotypes and antibiotic susceptibility testing of pylori under acidic conditions. J Infect Dis 2000, 182:1566–1569. Mycobacterium avium subspecies paratuberculosis. Clin Med Res 2004, 76. Maloney E, Stankowska D, Zhang J, Fol M, Cheng QJ, Lun S, Bishai WR, 2:107–114. Rajagopalan M, Chatterjee D, Madiraju MV: The two-domain LysX protein 55. Deshayes C, Laval F, Montrozier H, Daffé M, Etienne G, Reyrat JM: A of Mycobacterium tuberculosis is required for production of lysinylated glycosyltransferase involved in biosynthesis of triglycosylated phosphatidylglycerol and resistance to cationic antimicrobial peptides. glycopeptidolipids in Mycobacterium smegmatis: Impact on surface PLoS Pathogens 2009, 5:e1000534. properties. J Bacteriol 2005, 187:7283–7291. 56. Cangelosi GA, Palermo CO, Laurent JP, Hamlin AM, Brabant WH: Colony doi:10.1186/1471-2180-12-204 morphotypes on Congo red agar segregate along species and drug Cite this article as: Khattak et al.: Illegitimate recombination: An efficient susceptibility lines in the Mycobacterium avium-intracellulare complex. method for random mutagenesis in Mycobacterium avium subsp. Microbiology 1999, 145:1317–1324. hominissuis. BMC Microbiology 2012 12:204. 57. Shiratsuchi H, Toossi Z, Mettler MA, Ellner JJ: Colonial morphotype as a determinant of cytokine expression by human monocytes infected with Mycobacterium avium. J Immunol 1993, 150:2945–2954. 58. Curto M, Reali C, Palmieri G, Scintu F, Schivo ML, Sogos V, Marcialis MA, Ennas MG, Schwarz H, Pozzi G, et al: Inhibition of cytokines expression in human microglia infected by virulent and non-virulent mycobacteria. Neurochem Int 2004, 44:381–392. 59. Schaible UE, Sturgill-Koszycki S, Schlesinger PH, Russell DG: Cytokine activation leads to acidification and increases maturation of Mycobacterium avium-containing phagosomes in murine macrophages. J Immunol 1998, 160:1290–1296. 60. Steinert M, Birkness K, White E, Fields B, Quinn F: Mycobacterium avium bacilli grow saprozoically in coculture with Acanthamoeba polyphaga and survive within cyst walls. Appl Environ Microbiol 1998, 64:2256–2261. 61. Adékambi T, Salah SB, Khlif M, Raoult D, Drancourt M: Survival of environmental mycobacteria in Acanthamoeba polyphaga. Appl Environ Microbiol 2006, 72:5974–5981. 62. Cirillo JD, Falkow S, Tompkins LS, Bermudez LE: Interaction of Mycobacterium avium with environmental amoebae enhances virulence. Infect Immun 1997, 65:3759–3767. 63. Harriff M, Bermudez LE: Environmental amoebae and mycobacterial pathogenesis. Methods in molecular biology (Clifton, NJ) 2009, 465:433–442. 64. Danelishvili L, Wu M, Stang B, Harriff M, Cirillo S, Cirillo J, Bildfell R, Arbogast B, Bermudez LE: Identification of Mycobacterium avium pathogenicity island important for macrophage and amoeba infection. Proc Natl Acad Sci U S A 2007, 104:11038–11043. 65. Tenant R, Bermudez LE: Mycobacterium avium genes upregulated upon infection of Acanthamoeba castellanii demonstrate a common response to the intracellular environment. Curr Microbiol 2006, 52:128–133. 66. Goy G, Thomas V, Rimann K, Jaton K, Prod'hom G, Greub G: The Neff strain of Acanthamoeba castellanii, a tool for testing the virulence of Mycobacterium kansasii. Res Microbiol 2007, 158:393–397. 67. De Waal Malefyt R, Abrams J, Bennett B, Figdor CG, De Vries JE: Interleukin Submit your next manuscript to BioMed Central 10 (IL-10) inhibits cytokine synthesis by human monocytes: An and take full advantage of: autoregulatory role of IL-10 produced by monocytes. Journal of Experimental Medicine 1991, 174:1209–1220. • Convenient online submission 68. Cyktor JC, Turner J: Interleukin-10 and immunity against prokaryotic and eukaryotic intracellular pathogens. Infect Immun 2011, 79:2964–2973. • Thorough peer review 69. Smith I: Mycobacterium tuberculosis pathogenesis and molecular • No space constraints or color figure charges determinants of virulence. Clin Microbiol Rev 2003, 16:463–496. 70. Tateishi Y, Hirayama Y, Ozeki Y, Nishiuchi Y, Yoshimura M, Kang J, Shibata A, • Immediate publication on acceptance Hirata K, Kitada S, Maekura R, et al: Virulence of Mycobacterium avium • Inclusion in PubMed, CAS, Scopus and Google Scholar complex strains isolated from immunocompetent patients. Microb Pathog • Research which is freely available for redistribution 2009, 46:6–12. 71. Stokes RW, Doxsee D: The receptor-mediated uptake, survival, replication, and drug sensitivity of Mycobacterium tuberculosis within the Submit your manuscript at www.biomedcentral.com/submit

201 1

Modern and Ancestral Genotypes of Mycobacterium tuberculosis from Andhra Pradesh, India

Shirly K. Thomas1., Chitra C. Iravatham2., Bottu Heleena Moni3, Ashutosh Kumar3, Bandaru V. Archana1, Mohammad Majid3, Yerra Priyadarshini3, Pittu Sandhya Rani3, Vijayalakshmi Valluri2,4, Seyed E. Hasnain1,5, Niyaz Ahmed1,3,6* 1 Institute of Life Sciences, University of Hyderabad Campus, Hyderabad, India, 2 Bhagwan Mahavir Hospital and Research Centre, AC Guards, Hyderabad, India, 3 Pathogen Biology Laboratory, Department of Biotechnology, School of Life Sciences, University of Hyderabad, Hyderabad, India, 4 Blue Peter Research Centre – Lepra India, Cherlapally, Hyderabad, India, 5 School of Biological Sciences, Indian Institute of Technology, New Delhi, India, 6 Institute of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia

Abstract Traditionally, the distribution of the Mycobacterium tuberculosis genotypes in India has been characterized by widespread prevalence of ancestral lineages (TbD1+ strains and variants) in the south and the modern forms (TbD12 CAS and variants) predominating in the north of India. The pattern was, however, not clearly known in the south-central region such as Hyderabad and the rest of the state of Andhra Pradesh where the prevalence of both tuberculosis (TB) and human immunodeficiency virus (HIV) infection is one of the highest in the country; this area has been the hotspot of TB vaccine trials. Spoligotyping of 101 clinical isolates obtained from Hyderabad and rural Andhra Pradesh confirmed the occurrence of major genogroups such as the ancestral (or the TbD1+ type or the East African Indian (EAI) type), the Central Asian (CAS) or Delhi type and the Beijing lineage in Andhra Pradesh. Sixty five different spoligotype patterns were observed for the isolates included in this study; these were further analyzed based on specific genetic signatures/mutations. It was found that the major genogroups, CAS and ‘‘ancestral,’’ were almost equally prevalent in our collection but followed a north-south compartmentalization as was also reported previously. However, we observed a significant presence of MANU lineage in south Andhra Pradesh, which was earlier reported to be overwhelmingly present in Mumbai. This study portrays genotypic diversity of M. tuberculosis from the Indian state of Andhra Pradesh and provides a much needed snapshot of the strain diversity that will be helpful in devising effective TB control programs in this part of the world.

Citation: Thomas SK, Iravatham CC, Moni BH, Kumar A, Archana BV, et al. (2011) Modern and Ancestral Genotypes of Mycobacterium tuberculosis from Andhra Pradesh, India. PLoS ONE 6(11): e27584. doi:10.1371/journal.pone.0027584 Editor: Stefan Bereswill, Charite´-University Medicine Berlin, Germany Received September 1, 2011; Accepted October 19, 2011; Published November 17, 2011 Copyright: ß 2011 Thomas et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was funded under the Centre of Excellence Program of the Department of Biotechnology of the Indian Government (Ref. BT/01/CoE/07/02). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have read the journal’s policy and have the following conflicts: Niyaz Ahmed is a Section Editor of PLoS ONE and a member of the PLoS International Advisory Group. This does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials. * E-mail: [email protected] . These authors contributed equally to this work.

Introduction CAS lineages. Genotyping studies from India revealed a CRT specific silent mutation in 65th codon of the pncA gene which is Tuberculosis (TB) today constitutes the second major cause of specific for the CAS (Central Asian) lineage [4]. A new, ancient death due to infectious diseases. India being the hotspot region for clade of strains, called as ‘MANU’ was identified in India which TB witnesses one of the highest incidence rates although the belongs to the ancestral family of principle genetic group (PGG) - 1 mortality figures are on the decline due to effective implementa- [5,6] and is heavily concentrated in Mumbai [7]. It is tentatively tion of the control programs. Nevertheless, the scientific challenge subdivided in to MANU1 (ST100; loss of spacer 34), MANU2 (ST in TB control has become complicated with the emergence of new 54; loss of spacers 33, 34) and MANU3 (ST1378; deletion of frightening forms of tuberculosis – the extensively drug resistant spacers 34–36) [5,7,8,9]. It has been suggested that spoligotypes tuberculosis (XDR-TB) and the HIV-TB co-infection. evolve through the successive loss of spacer DNA sequences. In Correct identification of the underlying strains is of paramount addition to the loss of spacer 34, MANU [this study] and importance in devising TB control strategies. Spoligotyping is one ‘ancestral’ EAI or Indo-oceanic lineages [3] are characterized by of the potentially powerful tools for simultaneous detection and the deletion of RD 239 and an intact TbD1 region; it thus appears differentiation of M. tuberculosis complex lineages [1]. Studies using that MANU and EAI lineages are closely related or were derived multiple markers revealed predominance of ancestral (TbD1+) from a last common ancestor. and modern (TbD12) genogroups of M. tuberculosis strains in India Andhra Pradesh, with a population of about 80 million is one of [2]. The signature deletions RD239 and RD 750 [3] are India’s states with high prevalence of HIV/AIDS. Patients with specifically useful to define isolates belonging to the ‘Indo- latent TB infection are at higher risk of progression if they are co- Oceanic’ (TbD1+) [3] or ‘East African/Indian’ – EAI [2] and infected with HIV. A few TB vaccine trials have been initiated in

PLoS ONE | www.plosone.org 1 November 2011 | Volume 6 | Issue 11 | e27584 M. tuberculosis Genotypes from Andhra Pradesh this region although the repertoire of circulating strains is largely ‘Spotclust’ [12,13,14] to assign strain families based on standard unknown. Thus, the aim of this study was to characterize the definition/convention [13,14]. The isolates with spoligotype predominant genotypes responsible for TB in urban and rural patterns present in the SITVIT database (SpolDB4) were Andhra Pradesh and to generate a preliminary, baseline data for automatically labeled with a ‘shared type’ number and unique further epidemiological and infection control studies. profiles were mentioned as ‘new’. The spoligotyping results were further analyzed with the help of BionumericsH software program Materials and Methods (Applied Maths, Belgium); a dendrogram was constructed (Figure 1) by un-weighted pair group method using arithmetic Ethics statement averages (UPGMA) [15]. All the mycobacterial strains analyzed in this study were available as a part of routine TB testing and surveillance programs TbD1 analysis being implemented at different centres and, therefore, human The presence (TbD1+) or absence (TbD12) of the TbD1 region ethics committee (Institutional Review Board equivalent in India) was analyzed by PCR using two primer sets complementary to the approvals were not mandatory (also since no human biological sequence of the deleted region or complementary to the internal samples were collected here). The study was part of a long term sequence of the intact TbD1 region [16]. TbD1+ isolates open ended project originally approved by the Institutional generated a PCR product primed by the internal primers Biosafety Committee (IBSC) of the School of Life Sciences, (2,153 bp), whereas the TbD12 isolates generated a PCR product University of Hyderabad. based on the flanking primers (500 bp).

Bacterial isolates RD 239 and RD 750 deletion analysis A total of 101 clinical isolates of M. tuberculosis representing TB The RD 239 and RD 750 specific primers were used to amplify patients from urban and rural Andhra Pradesh were analyzed in specific signature DNA and assign strains to either the EAI lineage this study; 59 of these originated from a collection of MDR isolates or the CAS lineage [3,17]. RD239 deletions were denoted by an cultured during 2000–2005 as a part of Hyderabad Urban DOTS amplicon of size 888 bp and intact regions were scored based on program. The other isolates represented randomly selected the presence of an amplicon of 1730 bp. Similarly, an RD750 collection at microscopy centres operating under the aegis of the specific deletion resulted in a 734 bp PCR product and an intact revised national tuberculosis control program (RNTCP) (Table 1). region generated a product of 1533 bp. Smear positive sputum specimens received for routine diagnosis were processed by modified Petroff’s method and cultured on PCR-RFLP and Sequencing Lowenstein–Jensen slants at 37 C for 6–8 weeks. Drug suscepti- th u The pncA gene was analyzed for a silent mutation in 65 codon bility testing was performed by the absolute concentration (MIC) th at the 195 bp [4]. Restriction digestion using BseLI enzyme method for the anti-TB drugs, namely, isoniazid [H], ethambutol (Fermentas) differentiated the mutated and wild type alleles. Gel [E], rifampicin [R] and by resistance ratio method (RR method) electrophoresis was performed to resolve the digested products in for other drugs such as streptomycin [S], according to the protocol 1.5% agarose gel (Figure 2). of Tuberculosis Research Centre, Chennai, India (TRC) [10]. Multi Drug Resistance (MDR) phenotype was defined as resistance to both isoniazid and rifampicin. Results and Discussion Spoligotyping of 101 clinical M. tuberculosis isolates revealed 65 DNA isolation and spoligotyping distinct spoligopatterns. Thirty nine out of 101 clinical isolates Isolation of DNA was carried out as previously described [11]. were represented by a unique pattern whereas sixty two isolates Spoligotyping was performed according to the standard method segregated into 26 different clusters or branches (Figure 1). Three [1] and with the help of commercially available line probe arrays clusters comprising of 13 isolates, 7 isolates and 7 isolates belonged (Isogen Biosciences BV, Maarssen, the Netherlands). The to ST 26, ST126 and ST 100 respectively, followed by 6 isolates hybridization pattern was visualized after incubation with belonging to ST11 and 4 isolates belonging to ST591. Two streptavidin peroxidase using an enhanced chemiluminescence clusters (ST1, ST288) were formed with only 2 isolates each. Two detection system (Amersham) followed by exposure to an X-ray other clusters (ST127 and ST357) also comprised only 2 isolates film (Hyperfilm ECL, Amersham). Results were expressed as each and seventeen spoligopatterns represented single isolates presence (n) or absence (0) of each of the 43 spacers and converted (Figure 1). Sixty three percent of all clusters/branches had only to an octal code format according to Dale et al [12]. The data were one isolate and ,39% of all the isolates were not previously then compared to the SpolDB4 or SITVIT database and reported in the SpolDB4, SITVIT or ‘Spotclust’ and were

Table 1. Details of the M. tuberculosis isolates studied herein.

Isolate/batch identity (Numbers) Geographical origin Selection criteria Source centre type Important genogroups

AP01 to AP59 (59) Hyderabad and Ranga Reddy Pulmonary TB Hyderabad urban DOTS CAS district (urban group) MDR isolates EAI Beijing AP60 to AP101 (42) Chittoor district (rural group) Pulmonary TB RNTCP rural microscopy EAI Random selection/cross sectional centres MANU (No resistotyping data) CAS

doi:10.1371/journal.pone.0027584.t001

PLoS ONE | www.plosone.org 2 November 2011 | Volume 6 | Issue 11 | e27584 M. tuberculosis Genotypes from Andhra Pradesh

Figure 1. Genetic affinities within the M. tuberculosis isolates based on spoligotyping. Different clades corresponding to prevalent genotypes are prominently highlighted. In the inset is the distance coding convention relevant to the genetic relatedness of different isolates within a clade. doi:10.1371/journal.pone.0027584.g001 identified as ‘new’. These unique patterns were variously affiliated groups together comprised 77.22% of the total isolates. Other to CAS, EAI1, EAI3, EAI4, EAI5, family33, X1, T1, T3 and families represented were Beijing (,4%), family33 (4%), Haarlem Beijing genogroups. (H) [,3%], X (1%) and Tuscany (T) [,3%]. The EAI family The predominant spoligo families corresponded to CAS comprised of 55.26% EAI5, 36.84% EAI3, 5.26% EAI4, and (,40%), EAI (,38%) and MANU (,8%). The CAS and EAI ,2.63% EAI1. While the ST26 belonged to CAS1_Delhi lineage,

PLoS ONE | www.plosone.org 3 November 2011 | Volume 6 | Issue 11 | e27584 M. tuberculosis Genotypes from Andhra Pradesh

region. Shared type ST11 which belongs to EIA3_IND clade accounted for only 6% of all the clustered isolates. For this collection, the CAS (TbD12) and EAI (TbD1+) lineages were found almost in an equal ratio (,40% CAS and ,38% EAI). However, if we take in to account the MANU types, then the incidence of ancestral type isolates (TbD1+) becomes clearly dominant. It has previously been shown that the CAS lineage predominates in north and EAI in south India [2,4,5,6]. In our earlier study carried out on a nation wide sample, EAI genotypes were clearly less predominant in north India (32%), followed by 52% in central India and 80% in south India [2]. The present data are in agreement with such findings and also with respect to the prevalence of Beijing strains [18] that accounted to be only about 3–5% [2]. We did not find any Beijing isolates in Figure 2. PCR-RFLP analysis of the pncA gene after restriction our rural group and it is possible that southern Andhra Pradesh digestion with BseLI. The pncA gene was analyzed for a silent may not have any prevalence of this exotic genotype. mutation in the 65th codon at the 195th bp. Mutated alleles correspond Both mutated and wild type forms of the pncA gene were present to lanes 1, 5 and 6 (344 bp product and one minor fragment of 81 bp). in the MANU types. Thus, the CAS - lineage specific silent (CRT) Lanes 2, 3 and 7 denote wild type pattern (280 bp product and two minor fragments of 81 and 64 bp sizes) while lane 4 represents a mutation [4] was also noticed in the MANU isolates characterized negative control. Lane M corresponds to the profiling of a 100 bp DNA in this study. Since both the EAI and MANU lineages have molecular weight marker. RD239 deleted along with an intact TbD1, it is interesting to know doi:10.1371/journal.pone.0027584.g002 whether EAI or Indo-oceanic lineage is indeed evolved from MANU1 through successive loss of spacer DNA sequences. Since our data are based on limited isolates, there is a need for an ST11 was grouped under EAI3_IND lineage. ST100 belonged to elaborate study to understand this possibility as well as the MANU clade and ST126 was grouped under EAI5 (Figure 1) significance of MANU strains with mutated pncA gene. As The drug susceptibility testing of the 32 representative isolates described above, the codon 65 mutation seems to be common obtained by random selection out of the urban MDR group among ST26 and a single ST100 genotype, however, it is difficult (Table 1) showed 25 isolates (78.12%) of the families CAS, Beijing, to relate them based solely on single gene mutations. On the other EAI5, EAI3 and EAI4, were MDR. Although a single isolate from hand, TbD1 positivity and deletion of RD239 constitute much EAI4 was found to be resistant to streptomycin and isoniazid (but stronger evidence to portray them as ancestrally very closely sensitive to ethambutol and rifampicin), 6 other isolates belonging related. to different genotypes were found sensitive. Both the modern and ancestral M. tuberculosis strains are Random samples representing the EAI/Indo-oceanic lineage prevailing in this region with a north-south compartmentalization, were reconfirmed based on 1) the presence of RD239, 2) an intact respectively, and the isolates show a high degree of spoligotype RD750 region, 3) an intact TbD1 region together with 4) non- signature diversity. The present data once again remind of the mutated codon 65 of the pncA gene. Similarly, representative possibility that the ancestral strains are somehow more adapted to isolates from the CAS lineage were reconfirmed by 1) an intact southern peninsula. Although we did not analyze our samples RD239 region, 2) a deleted RD750 region, 3) the TbD1deletion based on demographic data of patients, it is possible that the and 4) mutations in the codon 65 of the pncA gene. The seven cosmopolitan nature of Hyderabad population could have lead to isolates belonging to MANU1 lineage were analyzed for mutations more diverse spoligotype patterns and representation of different in the pncA gene. PCR-RFLP for the pncA gene (Figure 2) was lineages without clear dominance of a single lineage. Although developed and standardized in house. Six of the MANU1 isolates ‘modern’ (TbD12) M. tuberculosis strains are far more prevalent had a wild type gene and a single isolate was found to be mutated worldwide, the ancestral clones of M. tuberculosis are responsible for at codon 65. Sequencing of the amplicons from four representative a majority of TB cases in India with the exception of major MANU1 isolates confirmed the PCR-RFLP data (Gen bank metropolitan cities. Our results therefore corroborate with the Accessions for the pncA: GU817406, GU817407, GU817408, findings of Gutierrez et al. [2] who reported similar trend on the GU817409). basis of genotyping of a ‘national collection’ of isolates. Spoligotype data from this study indicate that the known To sum up, this study has proven useful in identifying the lineages ST26, ST100, ST126 and ST11 predominate in Andhra predominant spoligotypes responsible for disease transmission in Pradesh; this trend was also reported from elsewhere [2,6], Andhra Pradesh. Significant presence of the ancestral type although the most significant finding from our data was the bacteria in the TB patients from Andhra Pradesh as shown here occurrence of orphan spoligotypes (new) not identified in such high assumes importance in the light of our earlier espousal [19,20,21] proportion in any previous studies from India. and one recent finding that M. tuberculosis belonging to the According to other studies from India, ST26, ST11, ST1 and ancestral lineage (EAI) could show reduced transmission as ST126 together accounted for a major proportion of all the compared to other lineages [22]. This perhaps explains why the isolates studied from this country [2,4,5,6,7,8]. Although, Indian population has never suffered with institutionalized TB traditionally, ST100 was found to be a minor clade in different outbreaks as seen in some other parts of the world where ancestral studies from India, except its high occurrence in Mumbai [7], we type bacteria are not so prevalent. Having said this, we sincerely in our study found that the ancestral isolates of ST126 and ST100 wish our observations may form path forward to more rigorous type (MANU) could be much more concentrated in southern future studies, based on whole genome sequencing of the Andhra Pradesh (Chittoor district). Moreover, the most predom- underlying isolates, to obtain better picture of the existing inant spoliogotype, ST26 (CAS1_Delhi), was found in 12.87% of diversity, transmission patterns and the preponderance of drug all the clustered isolates and was highly prevalent in Hyderabad resistant strains in this high-incidence region.

PLoS ONE | www.plosone.org 4 November 2011 | Volume 6 | Issue 11 | e27584 M. tuberculosis Genotypes from Andhra Pradesh

Acknowledgments Aparna Srikantam for suggestions. We would like to extend special thanks to Christa Ewers for help with the use of BionumericsH program. Our sincere thanks are due to the late K. J. R. Murthy. We are thankful for the guidance and support provided by V. M. Katoch, Director General, Author Contributions Indian Council of Medical Research, as well as technical assistance provided by Pragya Sharma, National JALMA Institute of Leprosy and Conceived and designed the experiments: NA VV SEH. Performed the other Mycobacterial Diseases, Agra, India. This work represents part of the experiments: SKT CCI BHM BVA. Analyzed the data: AK MM PSR YP. doctoral research of Chitra Iravatham. We are thankful to our Contributed reagents/materials/analysis tools: SEH VV. Wrote the paper: collaborators for providing genomic DNA of mycobacterial isolates and NA. Obtained permission from biosafety committee: NA.

References 1. Kamerbeek J, Schouls L, Kolk A, van Agterveld M, van Soolingen D, et al. 12. Dale JW, Brittain D, Cataldi AA, Cousins D, Crawford JT, et al. (2001) Spacer (1997) Simultaneous detection and strain differentiation of Mycobacterium oligonucleotide typing of bacteria of the Mycobacterium tuberculosis complex: tuberculosis for diagnosis and epidemiology. J Clin Microbiol 35: 907–914. recommendations for standardised nomenclature. Int J Tuberc Lung Dis 5: 2. Gutierrez MC, Ahmed N, Willery E, Narayanan S, Hasnain SE, et al. (2006) 216–219. Predominance of ancestral lineages of Mycobacterium tuberculosis in India. 13. Brudey K, Driscoll JR, Rigouts L, Prodinger WM, Gori A, et al. (2006) Emerg Infect Dis 12: 1367–1374. Mycobacterium tuberculosis complex genetic diversity: mining the fourth 3. Gagneux S, DeRiemer K, Van T, Kato-Maeda M, de Jong BC, et al. (2006) international spoligotyping database (SpolDB4) for classification, population Variable host-pathogen compatibility in Mycobacterium tuberculosis. Proc Natl genetics and epidemiology. BMC Microbiol 6: 23. Acad Sci U S A 103: 2869–2873. 14. Vitol I, Driscoll J, Kreiswirth B, Kurepina N, Bennett KP (2006) Identifying 4. Stavrum R, Myneedu VP, Arora VK, Ahmed N, Grewal HM (2009) In-depth Mycobacterium tuberculosis complex strain families using spoligotypes. Infect molecular characterization of Mycobacterium tuberculosis from New Delhi - Genet Evol 6: 491–504. predominance of drug resistant isolates of the ‘modern’ (TbD1) type. PLoS One 15. Hirsh AE, Tsolaki AG, DeRiemer K, Feldman MW, Small PM (2004) Stable 4: e4540. association between strains of Mycobacterium tuberculosis and their human host 5. Singh UB, Suresh N, Bhanu NV, Arora J, Pant H, et al. (2004) Predominant populations. Proc Natl Acad Sci U S A 101: 4871–4876. tuberculosis spoligotypes, Delhi, India. Emerg Infect Dis 10: 1138–1142. 16. Brosch R, Gordon SV, Marmiesse M, Brodin P, Buchrieser C, et al. (2002) A 6. Singh UB, Arora J, Suresh N, Pant H, Rana T, et al. (2007) Genetic biodiversity new evolutionary scenario for the Mycobacterium tuberculosis complex. Proc of Mycobacterium tuberculosis isolates from patients with pulmonary tubercu- Natl Acad Sci U S A 99: 3684–3689. losis in India. Infect Genet Evol 7: 441–448. 17. Reed MB, Pichler VK, McIntosh F, Mattia A, Fallow A, et al. (2009) Major 7. Chatterjee A, D’Souza D, Vira T, Bamne A, Ambe GT, et al. (2010) Strains of Mycobacterium tuberculosis lineages associate with patient country of origin. Mycobacterium tuberculosis from western Maharashtra, India, exhibit a high J Clin Microbiol 47: 1119–1128. degree of diversity and strain-specific associations with drug resistance, cavitary disease, and treatment failure. J Clin Microbiol 48: 3593–3599. 18. van Soolingen D, Qian L, de Haas PE, Douglas JT, Traore H, et al. (1995) 8. Suresh N, Singh UB, Arora J, Pant H, Seth P, et al. (2006) rpoB gene sequencing Predominance of a single genotype of Mycobacterium tuberculosis in countries and spoligotyping of multidrug-resistant Mycobacterium tuberculosis isolates of east Asia. J Clin Microbiol 33: 3234–3238. from India. Infect Genet Evol 6: 474–483. 19. Ahmed N, Hasnain SE (2011) Molecular epidemiology of tuberculosis in India: 9. Helal ZH, Ashour MS, Eissa SA, Abd-Elatef G, Zozio T, et al. (2009) Moving forward with a systems biology approach. Tuberculosis (Edinb) 91: Unexpectedly high proportion of ancestral Manu genotype Mycobacterium 407–413. tuberculosis strains cultured from tuberculosis patients in Egypt. J Clin Microbiol 20. Ahmed N, Ehtesham NZ, Hasnain SE (2009) Ancestral Mycobacterium 47: 2794–2801. tuberculosis genotypes in India: implications for TB control programmes. Infect 10. Tuberculosis Research Centre, Chennai (2010) Standard Operating Procedures Genet Evol 9: 142–146. for Mycobacteriology Laboratory. Tuberculosis Research Centre (Indian 21. Ahmed N, Dobrindt U, Hacker J, Hasnain SE (2008) Genomic fluidity and Council of Medical Research), Chennai, India. pathogenic bacteria: applications in diagnostics, epidemiology and intervention. 11. van Soolingen D, de Haas PE, Hermans PW, Groenen PM, van Embden JD Nat Rev Microbiol 6: 387–394. (1993) Comparison of various repetitive DNA elements as genetic markers for 22. Albanna AS, Reed MB, Kotar KV, Fallow A, McIntosh FA, et al. (2011) strain differentiation and epidemiology of Mycobacterium tuberculosis. J Clin Reduced Transmissibility of East African Indian Strains of Mycobacterium Microbiol 31: 1987–1995. tuberculosis. PLoS ONE 6: e25075.

PLoS ONE | www.plosone.org 5 November 2011 | Volume 6 | Issue 11 | e27584 Virulence Characteristics and Genetic Affinities of Multiple Drug Resistant Uropathogenic Escherichia coli from a Semi Urban Locality in India

Savita Jadhav1., Arif Hussain2., Savita Devi2., Ashutosh Kumar2, Sana Parveen2, Nageshwari Gandham1, Lothar H. Wieler3, Christa Ewers3, Niyaz Ahmed2,4,5* 1 Department of Microbiology, Dr. D. Y. Patil Medical College and Hospital, Pimpri, Pune, India, 2 Pathogen Biology Laboratory, Department of Biotechnology, School of Life Sciences, University of Hyderabad, Hyderabad, India, 3 Veterinary Faculty, Institute for Microbiology and Epizootics, Freie Universitaet Berlin, Berlin, Germany, 4 Institute of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia, 5 Institute of Life Sciences, University of Hyderabad Campus, Hyderabad, India

Abstract Extraintestinal pathogenic Escherichia coli (ExPEC) are of significant health concern. The emergence of drug resistant E. coli with high virulence potential is alarming. Lack of sufficient data on transmission dynamics, virulence spectrum and antimicrobial resistance of certain pathogens such as the uropathogenic E. coli (UPEC) from countries with high infection burden, such as India, hinders the infection control and management efforts. In this study, we extensively genotyped and phenotyped a collection of 150 UPEC obtained from patients belonging to a semi-urban, industrialized setting near Pune, India. The isolates representing different clinical categories were analyzed in comparison with 50 commensal E. coli isolates fromIndiaaswellas50ExPECstrainsfromGermany.Virulent strains were identified based on hemolysis, haemagglutination, cell surface hydrophobicity, serum bactericidal activity as well as with the help of O serotyping. We generated antimicrobial resistance profiles for all the clinical isolates and carried out phylogenetic analysis based on repetitive extragenic palindromic (rep)-PCR. E. coli from urinary tract infection cases expressed higher percentages of type I (45%) and P fimbriae (40%) when compared to fecal isolates (25% and 8% respectively). Hemolytic group comprised of 60% of UPEC and only 2% of E. coli from feces. Additionally, we found that serum resistance and cell surface hydrophobicity were not significantly (p = 0.16/p = 0.51) associated with UPEC from clinical cases. Moreover, clinical isolates exhibited highest resistance against amoxicillin (67.3%) and least against nitrofurantoin (57.3%). We also observed that 31.3% of UPEC were extended-spectrum beta-lactamase (ESBL) producers belonging to serotype O25, of which four were also positive for O25b subgroup that is linked to B2-O25b-ST131-CTX-M-15 virulent/multiresistant type. Furthermore, isolates from India and Germany (as well as global sources) were found to be genetically distinct with no evidence to espouse expansion of E. coli from India to the west or vice-versa.

Citation: Jadhav S, Hussain A, Devi S, Kumar A, Parveen S, et al. (2011) Virulence Characteristics and Genetic Affinities of Multiple Drug Resistant Uropathogenic Escherichia coli from a Semi Urban Locality in India. PLoS ONE 6(3): e18063. doi:10.1371/journal.pone.0018063 Editor: Ramy Aziz, Cairo University, Egypt Received January 6, 2011; Accepted February 18, 2011; Published March 25, 2011 Copyright: ß 2011 Jadhav et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This study was funded by the University of Hyderabad through intramural, startup grants to Niyaz Ahmed and was carried out as part of the Indo- German Research Training Group - Internationales Graduiertenkolleg (GRK1673) - Functional Molecular Infection Epidemiology, an initiative of the German Research Foundation (DFG) and the University of Hyderabad (India), of which Niyaz Ahmed and Lothar Wieler are Speakers. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: Niyaz Ahmed is a Section Editor of PLoS ONE and a Member of the PLoS International Advisory Group. * E-mail: [email protected] . These authors contributed equally to this work.

Introduction and invade renal tissues [6]. Specific adhesion is mediated by certain adhesins which can be differentiated based on their Urinary tract infections (UTI) are the second most common receptor binding specificity. human infections and are mainly caused by uropathogenic E. coli. P fimbriae that are encoded by the E. coli pap (pyelonephritis- The severity of UTI depends both on the virulence of the bacteria associated pilus) operon are the most important mannose-resistant and the susceptibility of the host [1]. UPEC harbor numerous adhesins, although they are expressed by only a limited number of virulence factors including alpha-hemolysin, cytotoxic necrotizing E. coli serotypes. The most rampant P-fimbriated serotypes of the factor, adhesins and iron acquisition systems. These factors UPEC strains revealed one of the six O groups O1, O2, O4, O6, support their ability to adhere to uroepethelial cells, help resist O7, and O18 [6,7,8]. the bactericidal effect of serum and augment cell surface Hemolysin production is another important virulence property hydrophobicity thereby leading ultimately to tissue damage of UPEC. Hemolysins inflict direct cytotoxic effects on renal [2,3,4,5]. epithelium resulting in scaring. Alpha-hemolysin is described to be Adherence to the urinary tract mucosa might protect bacteria a lethal factor with dermonecrotic effects and is antigenic in from urinary lavage and in turn augment their ability to survive nature. Also, hemolysins are toxic to a series of host tissues and

PLoS ONE | www.plosone.org 1 March 2011 | Volume 6 | Issue 3 | e18063 Profiling Multi-Resistant UPEC from Western India cells including RBCs, leucocytes, epithelial and endothelial cells. Study population and strains The frequency of isolation of hemolytic E. coli significantly One hundred and fifty E. coli isolates were obtained from urine associates with the severity of the infection [7,8]. samples of human patients in cell counts of 105/ml. Fifty fecal E. The importance of cell surface hydrophobicity as a virulence coli isolates (that were used as controls), were isolated from the attribute (that facilitates bacterial adherence to mammalian cells) is feces of healthy individuals who had reported for routine health known for nearly a century now, thanks to the pioneering studies checkup between January 2009–March 2010 at Dr. D. Y. Patil of Mudd and Mudd [4]. It is an important factor helping E. coli to Medical College and Hospital, Pimpri, Pune, India. Patient adhere to various surfaces for colonization. Bacteria are lysed by background and provisional diagnosis of the infection were normal serum due to the activity of the complement system. The obtained from hospital records (Table S1). Identification of isolates alternate pathway of complement activation is potentially was done using standard microbiological techniques [13,14,15]. important than the classical pathway. Bacterial resistance to All strains were stored on 15% glycerol-supplemented Luria- killing by serum results from individual or combined effects of Bertani medium at 280uC. capsular polysaccharides, lipopolysaccharides and surface proteins [4]. Given this generic virulence ‘arsenal’ of UPEC, strains from DNA samples different geographical regions pose different disease severity and DNA samples of fifty ExPEC isolates from Germany and from should be genetically different. global sources [16] were obtained; these represented various To screen for the above mentioned virulence attributes and to animal and human specimens, including 17 UPEC and archetyp- find the most predominant serotypes among UPEC in the western ical newborn meningitis strain RS218 from humans, two UPEC Indian region, we analyzed 150 human clinical isolates. We from dogs and 31 avian pathogenic E. coli (APEC) from different intended to classify them epidemiologically into different serotypes birds. These strains were used for repetitive extragenic palin- and to obtain virulence marker profiles of UPEC (which have dromic (rep) PCR together with the aforementioned human different frequencies in different disease conditions ranging from clinical isolates. asymptomatic bacteriuria to chronic pyelonephritis). In addition, increasing antimicrobial resistance in bacterial pathogens is of Antimicrobial susceptibility test major concern as it can vary according to geographical and regional situations [9,10,11,12]. It is very relevant to ensure the The antimicrobial susceptibility testing was carried out on appropriate therapy based on full knowledge of the organisms that Mueller Hinton agar by disc diffusion method using the following cause UTI and their antibiotic susceptibility profiles. Therefore, it antimicrobial substances [17]: amoxicillin (10 mg), ceftazidime is necessary to do bacteriological testing also with reference to (30 mg), ciprofloxacin (10 mg), co-trimoxazole (25 mg), gentamicin extended-spectrum beta-lactamase (ESBL) producers with resis- (10 mg), nitrofurantoin (300 mg), nalidixic acid (30 mg), and tance to beta-lactam antibiotics, including third generation tetracycline (30 mg). Multi-drug resistant (MDR) strains were cephalosporins such as cefotaxime, ceftriaxone and ceftazidime. defined as those which showed resistance to three or more Not much information on ESBL producing organisms causing antimicrobial substances. Extended-spectrum beta-lactamase UTI is available from India. (ESBL) production was detected by the double-disk synergy The present work was essentially carried out as an important (DDS) test as recommended by the Clinical and Laboratory precursor of a larger study aimed at understanding the Standards Institute (CLSI, 2006). Its presence was assayed using transmission dynamics, population genetic structure and virulence the antibiotic disks comprising of ceftazidime and ceftazidime/ mechanisms of UPEC from India. Herein, we present a much clavulanic acid (30/10 mg). An isolate was graded ESBL producer needed snapshot comprising of the virulence characteristics and when its zone of inhibition varied by $5 mm among at least one antimicrobial resistance patterns of the available UPEC strains of the combination disks and its coordinate comprising of standard from western India representing the clinical conditions such as antibiotic disk. E. coli ATCC 25922 was used as a negative control symptomatic UTI, bacteriuria, pyelonephritis, cystitis, prostatitis, while K. pneumoniae ssp. pneumoniae ATCC 700603 served as positive septicemia and pyrexia of unknown origin (PUO). We sought to control [18,19]. define the clinical correlates of different virulence factors and their association with various biological and life-style related factors of Phenotypic assays to determine virulence factors the host. In addition, we used rep-PCR based DNA profiling to a) Alpha-hemolysin production. The detection of alpha- know the genetic affinities of our isolates and to explore if this hemolysin was performed by analysing the hemolytic zone observed simple PCR based genetic analysis helps in understanding their after overnight growth at 37uC on sheep blood (5%) agar [20]. spread patterns indicative of their diverse or clonal relationships. b) Haemagglutination. A suspension of human A +ve blood Also, we attempted to use this technique to know whether the and PBS was mixed on a VDRL (venereal diseases research bacterial isolates have any geographic inclination and how similar laboratory - test) cavity slide with a single colony of E. coli. After or different they are when compared to some of the well incubation on rotor at room temperature for some minutes, characterized isolates from western countries such as Germany. agglutination was seen. Similarly, haemagglutination was carried out in the presence of D-mannose. An ATCC E. coli 25922 strain Materials and Methods was used as a negative control for mannose sensitive haemagglutination assay and a known strain of E. coli repeatedly Ethics statement giving positive was taken as control for the assay [21]. Written informed consents were obtained from all patients and c) Cell Surface Hydrophobicity. Bacteria were tested for healthy controls for the use of their strains which were cultured as their hydrophobic property by using different molar part of compulsory diagnostic screening. Study protocols were concentrations of ammonium sulphate in VDRL tile; 40 mlof approved by the institutional ethics committee of Dr. D. Y. Patil bacterial suspension in PBS was added in each of the wells Medical College, Pune and by the Institutional Biosafety containing 1 M, 1.4 M and 2 M ammonium sulphate. Clumps Committee of the School of Life Sciences of the University of were seen by naked eyes. Strains were considered hydrophobic, if Hyderabad, India. they aggregated in the PBS concentration of #1.4 M [21].

PLoS ONE | www.plosone.org 2 March 2011 | Volume 6 | Issue 3 | e18063 Profiling Multi-Resistant UPEC from Western India

d) Serum Bactericidal Assay. Bacteria were diluted in periurban Pune. As such there is not much information available Hank’s balanced salt solution to get appropriate dilutions and from India on the genetic and phenotypic diversity of UPEC. This mixed with human serum in a sterile tube and incubated in a information should therefore be construed as the first systematic water bath at 37uC and inoculated on nutrient agar plates at 0 hr, analysis performed on a diverse type of patients/samples. 1 hr, 2 hr of incubation in water bath. Growth at 0 hr was taken as control. E. coli was considered sensitive if count dropped to 1% Virulence characteristics of UPEC from India and was considered resistant if .90% of the organisms survived a) Type 1 (MSHA) and P fimbriae (MRHA). In the present after 2 hrs of incubation. An E. coli isolate that was consistently study, 45% of the UTI isolates and 25% of fecal isolates showed serum resistant and a one which was consistently sensitive were the presence of type 1 (Mannose sensitive hemagglutination - used as positive and negative controls respectively [22]. MSHA) fimbriae. Thus the difference among UTI and fecal isolates was not statistically significant. It was found that 40% of Serotyping UTI isolates and 8% of E. coli isolates from the controls exhibited Typing of somatic antigens was performed at the National P fimbriae (Mannose resistant hemagglutination -MRHA). The Salmonella and Escherichia Centre, Central Research Institute, difference was statistically significant (P = 0.0002). Expression of Kasauli, India, using antisera against O antigens - O1 to O173. type 1 fimbriae was more discerned in isolates from the cases of O25 positive E. coli strains were further subjected to genotyping by simple UTI and cystitis. In our observation, maximum P fimbriae a recent method based on allele-specific PCR targeting the positive isolates were highly associated with simple UTI, while rfbO25b subgroup gene locus [23]. only 10% of such isolates caused pyelonephritis. As such there is no relation among the occurrence or severity of symptoms/the site Repetive extragenic palindromic PCR of infection and fimbrial expression (of P or type1) in case of DNA sequences of primers used for Rep - fingerprinting were as bacterial isolates from urine [28]. follows: Forward Rep2a2: 59-ACGGCTTATCGGGCCTA-39, Our observation was in agreement with another study [29], Reverse Rep1Ra1:59-GCGACGGCATCAGGC-39. PCR ampli- wherein the presence of type 1 fimbriae was seen in 71% of isolates fication was carried out in 20 ml reaction mixture consisting of of E. coli from UTI and in 60% of isolates of E. coli from the control fecal flora. Taken together, UTI associated isolates did not 106 Taq buffer, 25 mM MgCl2, 10 mM dNTP, 100% DMSO, 100 pM of each primer and 2 U of Taq DNA polymerase and show any significant expression of MSHA compared to controls; 5 ml of template. The PCR conditions were 95uC for 7 min, however, Najar et al. (2007) [29] showed an overall higher followed by 30 cycles of DNA amplification consisting of 45 s at expression of type 1 fimbriae in both the UTI isolates and controls. Many studies have demonstrated the presence of P fimbriae as 95uC, 1 min at 40uC and 8 min at 65uC followed by 16 min maximum in UTI isolates than in fecal isolates of healthy persons incubation at 65uC. The amplicons were run on 1.5% agarose gels and were analyzed by BionumericsH software [24,25]. [29,30]. Although, since we looked only at phenotypes of the isolates in vitro, it is difficult to dwell upon the real significance of the fimbrial Detection of O25b subgroup strain by PCR expression patterns vis a vis the niche tropism (upper or lower UTI) E. coli The newly described O25b O type were detected by using and its impact on different outcomes of infection – these are 9 the following primers gndbis.f (5 ATACCGACGACGCC- largely the phenomena due to microbial mechanisms operating in 9 9 GATCTG-3 ) and rfbO25b.r (5 TGCTATTCATTATGCG- vivo. Normally, the expression of P fimbriae is increased among 9 CAGC-3 ). Annealing temperature of 60uC was used to generate isolates associated with lower urinary tract infections, basically a PCR product of 300 bp with the conditions as previously resembling an ‘in vivo’ situation as there is an interplay between described [23]. type1-fimbriae and P-fimbriae expression (‘‘switch-off’’ of Type 1 fimbriae leads to increased motility of bacteria and finally to Statistical Analysis binding of P-fimbrial adhesin to receptors on kidney epithelial Chi square test was used to compare the occurrence of virulence cells). The higher or lower expression itself might therefore only markers in cases and controls. P values less than 0.05 was give an idea about the quantitative occurrence of P fimbriae, but considered significant. not about the different adhesion subunits which would be more relevant with respect to upper/lower UTI infections. Results and Discussion b) Hemolysin. Production of alpha hemolysin was observed in 60% of UPEC from the cases while only 2% of E. coli isolates Incidence of UTI in relationship with gender and from feces produced alpha hemolysin (P = 0.0001). Other studies socioeconomic status have also reported similar findings [31]. In our observation, In this study we observed a higher proportion of UTI in females hemolysin was highest in pyelonephritis, lower in cystitis and least (64%) than in males (36%). This is understandable due to the in asymptomatic bacteriuria. anatomy and is a consistent trend worldwide. Peak in the c) Serum resistance and cell surface hydrophobicity. incidence of UTI was observed in the age groups 11–21 and Serum resistance and cell surface hydrophobicity markers were 60–71 years. Among these, elderly patients are likely predisposed not significantly present in E. coli isolates from patients compared to conditions such as urinary tract obstruction, poor bladder to control fecal isolates: 55% and 76% of urinary E. coli isolates emptying, and diabetes mellitus, etc. These factors favor showed serum resistance and cell surface hydrophobicity colonization of bacteria and play an important role in UTI. respectively and the control isolates showed 40% and 70% for Other studies have also reported similar findings [26,27]. India has serum resistance and cell surface hydrophobicity respectively. a large infection burden and the genito-urinary infections are very The difference in both the cases was not statistically significant prominent. This may be due to a less affordable personal/ (p = 0.16/p = 0.51). This trend has been reported earlier and our community hygiene for some of the economically backword study confirmed the same for Indian isolates [32]. populations. We targeted these groups and analyzed strains Occurrence of virulence factors in UPEC strains confirms the obtained from such communities who reported for the UTI in association of UPEC with urinary pathogenicity. It was interesting

PLoS ONE | www.plosone.org 3 March 2011 | Volume 6 | Issue 3 | e18063 Profiling Multi-Resistant UPEC from Western India

Table 1. Occurrence of virulence phenotypes in E. coli isolates from cases and controls.

Number of isolates from cases (%) total Number of isolates from controls (%) total Virulence phenotypes cases = 150 controls = 50 P-value

MRHA 60 (40.0) 4 (8.0) 0.0002 MSHA 68 (45.3) 12 (24.0) 0.406 Hemolysin 90 (60.0) 1 (2.0) 0.00001 SR 83 (55.3) 20 (40.0) 0.16 CSH 120 (80.0) 35 (70.0) 0.51

doi:10.1371/journal.pone.0018063.t001 to note that UPEC with virulence factors were significantly more virulence potential of such pathogens. However, eight of the prevalent in patient groups than in controls (Table 1). Such a mucoid strains were susceptible to serum, perhaps due to the non significant difference was noticed earlier, indicating that several capsular factors and this might have a role in serum resistance. virulence factors act synergistically to cause infection of the urinary tract [33]. Nevertheless, it was interesting that 8% of the isolates Antimicrobial susceptibility profiles of UPEC (12/150) from infection cases did not reveal any virulence markers Antimicrobial susceptibility testing was carried out on all the under investigation. These were the isolates from complicated clinical isolates. The majority of isolates were sensitive to UTI. It appears that these isolates, though devoid of any virulence nitrofurantoin - 86 (57.3%) followed by ciprofloxacin - 78 (52%) factors as probed by us, were able to localize in the deeper tissues and nalidixic acid - 74 (49%). This study showed high resistance to of a compromised urinary tract where scarcely any drug reaches in amoxicillin - 101 (67.3%), tetracycline - 92 (61.3%), and therapeutic quantum. It may also be due to the reason that they cefotaxime - 68 (45.3%) (Figure 1). Upon testing for ESBL, we might possess virulence markers/mechanisms other than those found that 32 (21.3%) were ESBL producers. A high proportion tested by us. (31.3%) of these ESBL-producing isolates was belonging to the We screened 19 mucoid UPEC isolates by phenotypic and serotype O25. Four of these were positively tested for the subgroup biochemical characterization and found that all were from patient O25b (Figure 2), which has been linked to a clonally related group groups and were capsulated. Out of the 19 mucoid UPEC, 11 of highly virulent, multiresistant E. coli strains (B2-O25b-ST131- were capsulated and serum resistant. Capsule confers serum and CTX-M-15) that are emerging among humans and animals phagocyte resistance and this could be attributed to sialic acid worldwide [34,35]. Further studies, i.e. multi locus sequence residues that subvert the ability of bacterial surface to activate typing (MLST), macrorestriction analysis, and determination of complement by alternative pathway thereby augmenting the beta-lactamase enzymes are needed to unravel the phylogenetic

Figure 1. The antimicrobial susceptibility and resistance pattern of 150 UTI isolates from India. Nitrofurantoin was consistently the most active (59%) of the systemically active antimicrobials, with ceftazidime, giving similar results (54.7%). These were followed by ciprofloxacin (52%), nalidixic acid (49%) and gentamicin (48%), but sensitivity to amoxicillin was found to be low (32.7%). Sensivity profiles of Co-trimoxazole (36.7%) and tetracycline (38.7%) were quite similar. doi:10.1371/journal.pone.0018063.g001

PLoS ONE | www.plosone.org 4 March 2011 | Volume 6 | Issue 3 | e18063 Profiling Multi-Resistant UPEC from Western India

pathogens acquire resistance determinants and express a multire- sistant phenotype at the cost of their virulence properties, the frequent observation of hemolytic UPEC among our strain collection underlines the possible emergence of highly virulent multiresistant strains [36] in this area. It is particularly worrisome that more than half of the ESBL producers (53.1%) in our collection were resistant to ciprofloxacin suggesting that they may be resistant to all available fluoroquino- lones - the drug of choice for treatment of infections caused by ESBL-producing strains. Overall, nitrofurantoin and ciprofloxacin were found superior to amoxicillin and co-trimoxazole. None of the uropathogens from this study area were 100% susceptible to any of the antimicrobial substances used. Prior studies have shown a sensitivity rate of 95–100% to nitrofurantoin [37]. But in our study, nitrofurantoin showed only 57.3% sensitivity; the reason for this low Figure 2. PCR based detection of O25b subgroup that is linked to the B2-O25b-ST131-CTX-M-15 clonal group of strains. sensitivity and high resistance to almost all antimicrobials in the Among 32 ESBL producers in our collection (n = 150) four were positive study area may indicate difference in antimicrobial usage, infection for O25b subgroup (Lanes 1 to 4); lane 5, positive (strain showing control practices and other unrecognized factors including genetic consistently positive); lane 6, negative; M, 100 bp DNA ladder. propensity of these strains to accumulate mutations conferring doi:10.1371/journal.pone.0018063.g002 MDR phenotypes. Periodic review and formulation of antibiotic policy are needed to control acquisition of drug resistance. Further relatedness of these O25-ESBL-producing Indian isolates to the studies for better understanding of the interaction of different worldwide recognized clonal groups and to characterize the virulence factors at molecular level are necessary as most urovirulent remaining multiresistant isolates observed. ESBL isolates were strains express multiple virulence factors simultaneously. We believe frequently (65.6%) associated with a hemolytic phenotype, with a that the methods of detection of the above mentioned virulence higher rate than the non-ESBL-producing UPEC strains (58.5%). markers are reasonably easy and screening them in a clinical Hence, although it is a commonly accepted fact that bacterial microbiology laboratory is a worthwhile exercise.

Table 2. Distribution of various O-antigens in clinical isolates of E. coli.

O-Antigen Pyelonephritis Prostatitis Cystitis Simple UTI Septicaemia PUO Total

O1- 1-52-8 O2 - 1 - 1 - - 2 O9------0 O14 1 - - - - - 1 O20 1 - - - 1 1 3 O25 6 7 11 15 2 4 45 O44 1 - - 2 1 - 4 O45 - - 1 - - - 1 O60 1 1 4 1 - 7 O64 - - 1 1 - - 2 O76 - - 1 3 - - 4 O79 - 1 - - - - 1 O84 - - - 1 - - 1 O102 3 1 - 8 - 1 13 O116 - - - 1 - - 0 O120 2 2 3 6 - 1 14 O130 - 1 - 1 - - 2 Rough 1 1 2 2 - 6 O951 -----1 O100 - - - 1 - - 1 Non- viable - 1 - - 1 Non E. coli -112-15 Untypeable 5 - 3 13 3 4 28 Total 22 15 24 65 12 12 150

doi:10.1371/journal.pone.0018063.t002

PLoS ONE | www.plosone.org 5 March 2011 | Volume 6 | Issue 3 | e18063 Profiling Multi-Resistant UPEC from Western India

Serotyping and genotyping In our study, as many as 19 different serotypes (O1, O2, O9, O14, O20, O25, O44, O45, O60, O64, O76, O79, O84, O102, O116, O120, O130, O95, O100) were observed among the cultured UPEC. Out of these, 27 E. coli were untypeable; six expressed a rough O antigen and five were Escherichia species (non- E. coli; NEC). A majority of E. coli isolates expressed O antigens - O25 (31.9%), O102 (9.7%), O120 (8.3%), and O1 (5.6%). Serotype O25 was predominantly found to be isolated from patients with pyelonephritis, prostatitis, cystitis and simple UTI. Renal failure associated isolates mainly belonged to O25 and O102 serotypes (Table 2). To determine the relatedness of the strains, a dendrogram based on rep-PCR based fingerprint data was constructed in Bionumer- icsH. For this study, the clinical strains from India and 50 German isolates were used while the fecal isolates were excluded as the aim of this experiment was not to prove pathogenicity of the ExPEC but to see how diverse and clonal they are and to analyze Indian isolates in juxtaposition with German strains. Apart from the occurrence of singleton strains, the dendrogram obtained could be broadly classified into 5 major clusters: Indian-German cluster I (n = 15 strains), Indian cluster I (n = 63), mixed cluster I (n = 17) and mixed clusters IIa/IIb (n = 30), which include strains from all the three locales, as depicted in Figure 3. All the strains that clustered under Indian branch (exceptionally harboring one isolate from Germany) revealed very diverse pattern, except for a few ones - it appears that the bacteria we cultured were very diverse and the ones which were close in the cluster have no phenotypic resemblances with each other in terms of antimicrobial resistance profile or serotype or other virulence markers. This may be due to the reason that the UPEC in this study area could be highly diverse at sub-species and serotype levels. The mixed cluster I contained a group of 17 bacterial isolates from India, Germany and from global sources. This cluster predominantly comprises of APEC from cases of septicemia in chicken or goose, while the remaining seven strains were human UPEC, including strain CFT073, or they were septicemia- associated strains (SEPEC). The mixed cluster II was subdivided into two subclusters- subcluster IIa (n = 10) and subcluster IIb (n = 20). The subcluster IIa largely encompasses APEC. Eight out of ten members of this cluster were APEC and the rest were UTI pathogens from humans. In the subcluster IIb, human isolates as well as isolates from animals were grouped together. There was no discrete band pattern between strains isolated from different sources; out of twenty strains in subcluster IIb nine were APEC, seven were UPEC of humans and two from dog, one was a human SEPEC and finally the archetypical NMEC strain RS218 grouping within this subcluster. From this analysis, we may conclude that the technique of rep-PCR is not resolving enough to distinguishing strains that portray similar phenotypic markers; nevertheless the Indian ExPEC population was clearly distin- guishable from German ExPEC. We also found that the isolates from different animal sources were more or less grouping into distinct clusters, indicating that this technique can also be used to find the sources of infection in a reservoir such as a water body. Figure 3. Dendrogram based on rep-PCR, developed in In conclusion, our observations form an important baseline BionumericsH revealed genetic relationships of E. coli repre- senting Indian clinical isolates and ExPECs from global data-set towards understanding the virulence properties, antibiotic sources. Different geographic clusters are labeled. Upper scale denotes resistance profiles and genetic diversity of UPEC from India. We genetic distance. hope that such observations will be more meaningful towards doi:10.1371/journal.pone.0018063.g003 systematically unraveling the population genetic structure of UPEC and their propensity to spread, or emerge with multi drug resistance phenotypes in new epidemiological territories. In the whole genome sequencing of representative isolates from different backdrop of this work, future studies involving decipherment of MLST clades [2] shall be imminent. Further, the ‘omics’ inspired MDR mechanisms, lineage tracking based on MLST [16] and studies will ultimately help understanding of the functional

PLoS ONE | www.plosone.org 6 March 2011 | Volume 6 | Issue 3 | e18063 Profiling Multi-Resistant UPEC from Western India molecular epidemiology and infection biology of Indian UPEC at Acknowledgments greater details. On the clinical side, it will be possible to ascertain reservoirs of infection which could be important in understanding We would like to extend our thanks to Professors Jo¨rg Hacker and Seyed E. the mechanisms of chronic and recurrent UTI [38] in a Hasnain for their guidance and timely suggestions. We are also thankful to members of the pathogen evolution laboratory in University of Hyderabad community setting and how they are maintained within or in for their help and support. We acknowledge assistance provided by Mrs. G. the vicinity of different human populations of a multicultural and Navamallika at different stages of this study. densely populated country such as India. Author Contributions Supporting Information Conceived and designed the experiments: NA. Performed the experiments: Table S1 Clinical types, antibiogram, serotype and virulence SJ AH SD SP. Analyzed the data: AK CE NA. Contributed reagents/ characteristics of the UPEC isolates from Pune, India. materials/analysis tools: NG LHW. Wrote the manuscript: CE NA. (DOC)

References 1. Fowler JE, Stamey TA (1977) Studies of introital colonization in women with 20. Sakoulas G, Eliopoulos GM, Moellering RC, Jr., Wennersten C, Venkataraman L, recurrent urinary infections. The role of bacterial adherence. J Urol 117: et al. (2002) Accessory gene regulator (agr) locus in geographically diverse 472–476. Staphylococcus aureus isolates with reduced susceptibility to vancomycin. 2. Dobrindt U (2005) (Patho-)Genomics of Escherichia coli. Int J Med Microbiol Antimicrob Agents Chemother 46: 1492–1502. 295: 357–371. 21. Siegfried L, Kmetova M, Puzova H, Molokacova M, Filka J (1994) Virulence- 3. Johnson JR, Russo TA (2005) Molecular epidemiology of extraintestinal associated factors in Escherichia coli strains isolated from children with urinary pathogenic (uropathogenic) Escherichia coli. Int J Med Microbiol 295: 383–404. tract infections. J Med Microbiol 41: 127–132. 4. Leffler H, Svanborg-Eden C (1981) Glycolipid receptors for uropathogenic 22. Pezzlo M, York MK (2004) Urine Cultures. In: Isenberg HD, ed. Clinical Escherichia coli on human erythrocytes and uroepithelial cells. Infect Immun 34: Microbiology Procedures Handbook 2nd ed. Washington D. C.: American 920–929. Society of Microbiology. pp 31211–131231. 5. Hughes C, Hacker J, Roberts A, Goebel W (1983) Hemolysin production as a 23. Clermont O, Dhanji H, Upton M, Gibreel T, Fox A, et al. (2009) Rapid virulence marker in symptomatic and asymptomatic urinary tract infections detection of the O25b-ST131 clone of Escherichia coli encompassing the CTX- caused by Escherichia coli. Infect Immun 39: 546–551. M-15-producing strains. J Antimicrob Chemother 64: 274–277. 6. Hagan EC, Mobley HL (2007) Uropathogenic Escherichia coli outer membrane 24. Versalovic J, Koeuth T, Lupski JR (1991) Distribution of repetitive DNA antigens expressed during urinary tract infection. Infect Immun 75: 3941–3949. sequences in eubacteria and application to fingerprinting of bacterial genomes. 7. Blanco J, Alonso MP, Blanco M, Blanco JE, Gonzalez EA, et al. (1992) Nucleic Acids Res 19: 6823–6831. Establishment of three categories of P-fimbriated Escherichia coli strains that 25. Rademaker JLW, de Bruijn FJ (1997) Characterization and classification of show different toxic phenotypes and belong to particular O serogroups. FEMS Microbiol Lett 78: 131–136. microbes by rep-PCR genomic fingerprinting and computer-assisted pattern 8. Blanco M, Blanco J, Blanco JE, Alonso MP, Abalia I, et al. (1995) [Virulence analysis, 151–17. In Caetano-Anolle´s G, Gresshoff PM, eds. DNA markers: factors and O serogroups of Escherichia coli as a cause of community-acquired protocols, applications, and overviews, J. Wiley and Sons, New York, N.Y. urinary infections]. Enferm Infecc Microbiol Clin 13: 236–241. 26. Gales AC, Sader HS, Jones RN (2002) Urinary tract infection trends in Latin 9. Mathai D, Jones RN, Pfaller MA (2001) Epidemiology and frequency of American hospitals: report from the SENTRY antimicrobial surveillance resistance among pathogens causing urinary tract infections in 1,510 hospitalized program (1997–2000). Diagn Microbiol Infect Dis 44: 289–299. patients: a report from the SENTRY Antimicrobial Surveillance Program 27. Ulleryd P (2003) Febrile urinary tract infection in men. Int J Antimicrob Agents (North America). Diagn Microbiol Infect Dis 40: 129–136. 22 Suppl 2: 89–93. 10. Karlowsky JA, Jones ME, Thornsberry C, Critchley I, Kelly LJ, et al. (2001) 28. Johnson JR (1991) Virulence factors in Escherichia coli urinary tract infection. Prevalence of antimicrobial resistance among urinary tract pathogens isolated Clin Microbiol Rev 4: 80–128. from female outpatients across the US in 1999. Int J Antimicrob Agents 18: 29. Najar GA, Nejad MM, Mansouri S (2007) The comparison between virulence 121–127. factors of Escherichia coli isolated from urinary tract infections and fecal flora. Res 11. Sahm DF, Thornsberry C, Mayfield DC, Jones ME, Karlowsky JA (2001) Pharm Sci 2: 99–103. Multidrug-resistant urinary tract isolates of Escherichia coli: prevalence and 30. Jacobsen SM, Stickler DJ, Mobley HL, Shirtliff ME (2008) Complicated patient demographics in the United States in 2000. Antimicrob Agents catheter-associated urinary tract infections due to Escherichia coli and Proteus Chemother 45: 1402–1406. mirabilis. Clin Microbiol Rev 21: 26–59. 12. Dombek PE, Johnson LK, Zimmerley ST, Sadowsky MJ (2000) Use of repetitive 31. Minshew BH, Jorgensen J, Swanstrum M, Grootes-Reuvecamp GA, Falkow S DNA sequences and the PCR To differentiate Escherichia coli isolates from (1978) Some characteristics of Escherichia coli strains isolated from extraintes- human and animal sources. Appl Environ Microbiol 66: 2572–2577. tinal infections of humans. J Infect Dis 137: 648–654. 13. Quackenbush RL, Falkow S (1979) Relationship between colicin V activity and 32. Puzova H, Siegfried L, Kmetova M, Filka J, Takacova V, et al. (1994) virulence in Escherichia coli. Infect Immun 24: 562–564. Fimbriation, surface hydrophobicity and serum resistance in uropathogenic 14. Collee JG, Duguid JP, Fraser AG, Marmion BP, Simmons A (1996) Laboratory strains of Escherichia coli. FEMS Immunol Med Microbiol 9: 223–229. strategy in the diagnosis of infective syndromes. In: Collee JG, Fraser AG, 33. Johnson JR (2003) Microbial virulence determinants and the pathogenesis of Marmion BP, Simmons A, eds. Mackie and McCartney practical medical urinary tract infection. Infect Dis clin North Am 17(2): 261–71. microbiology. 14th ed, Churchill Livingstone Inc: London. pp 53–94. 34. Ewers C, Grobbel M, Stamm I, Kopp PA, Diehl I, et al. (2010) Emergence of 15. Barron EJ, Fingold MS (1990) Diagnostic Microbiology. In: Baily, Scott, eds. 8th human pandemic O25:H4-ST131 CTX-M-15 extended-spectrum-beta-lacta- ed. Mosby Public Philadelphia Toronto, Baltimore. pp 259–260. mase-producing Escherichia coli among companion animals. J Antimicrob 16. Ewers C, Li G, Wilking H, Kiessling S, Alt K, et al. (2007) Avian pathogenic, Chemother 65: 651–660. uropathogenic, and newborn meningitis-causing Escherichia coli: how closely 35. Nicolas-Chanoine MH, Blanco J, Leflon-Guibout V, Demarty R, Alonso MP, related are they? Int J Med Microbiol 297: 163–176. et al. (2008) Intercontinental emergence of Escherichia coli clone O25:H4- 17. Clinical and Laboratory Standards Institute (2006) ‘Performing standards for antimicrobial susceptibility testing’, Fifteenth information supplement, Approved ST131 producing CTX-M-15. J Antimicrob Chemother 61: 273–281. standard MS45-A. Wayne, PA. 36. Wiles TJ, Kulesus RR, Mulvey MA (2008) Origins and virulence mechanisms of 18. Babypadmini S, Appalaraju B (2004) Extended spectrum -lactamases in urinary uropathogenic Escherichia coli. Exp Mol Pathol 85: 11–19. isolates of Escherichia coli and Klebsiella pneumoniae - prevalence and 37. Tambekar DH, Dhanorkar DV, Gulhane SR, Khandelwal VK, Dudhane MN susceptibility pattern in a tertiary care hospital. Indian J Med Microbiol 22: (2006) Antibacterial susceptibility of some urinary tract pathogens to commonly 172–174. used antibiotics. Afr J Biotechnol 5: 1562–1565. 19. Tankhiwale SS, Jalgaonkar SV, Ahamad S, Hassani U (2004) Evaluation of 38. Hannan TJ, Mysorekar IU, Hung CS, Isaacson-Schmid ML, Hultgren SJ (2010) extended spectrum beta lactamase in urinary isolates. Indian J Med Res 120: Early severe inflammatory responses to uropathogenic E. coli predispose to 553–556. chronic and recurrent urinary tract infection. PLoS Pathog 6: e1001042.

PLoS ONE | www.plosone.org 7 March 2011 | Volume 6 | Issue 3 | e18063