Analysis of the Interaction of Francisella Tularensis Lvs

Clemson University TigerPrints

All Dissertations Dissertations

5-2013 ANALYSIS OF THE INTERACTION OF FRANCISELLA TULARENSIS LVS AND LEGIONELLA PNEUMOPHILA BIOFILMS WITH COMPONENTS OF THE AQUATIC ENVIRONMENT Uma Mahajan Clemson University, [email protected]

Follow this and additional works at: https://tigerprints.clemson.edu/all_dissertations Part of the Microbiology Commons

Recommended Citation Mahajan, Uma, "ANALYSIS OF THE INTERACTION OF FRANCISELLA TULARENSIS LVS AND LEGIONELLA PNEUMOPHILA BIOFILMS WITH COMPONENTS OF THE AQUATIC ENVIRONMENT" (2013). All Dissertations. 1117. https://tigerprints.clemson.edu/all_dissertations/1117

This Dissertation is brought to you for free and open access by the Dissertations at TigerPrints. It has been accepted for inclusion in All Dissertations by an authorized administrator of TigerPrints. For more information, please contact [email protected].

ANALYSIS OF THE INTERACTION OF FRANCISELLA TULARENSIS LVS AND LEGIONELLA PNEUMOPHILA BIOFILMS WITH COMPONENTS OF THE AQUATIC ENVIRONMENT

A Dissertation Presented to the Graduate School of Clemson University

In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy Microbiology

by Uma Vikas Mahajan May 2013

Accepted by: Dr. Tamara L. McNealy, Committee Chair Dr. Matthew Turnbull Dr. Peter H. Adler Dr. Susan Chapman

ABSTRACT

Biofilms are complex communities of organisms enclosed in a matrix (EPS) composed of polysaccharides, proteins and DNA. Biofilms are the base of the aquatic ecosystem and interact with the biotic as well as the abiotic components in the environment. Biofilms interact with biotic components of the environment such as protozoa and aquatic invertebrates. In this project, the biotic interactions of Francisella tularensis LVS biofilms are studied in relation to the mosquito larvae of Culex quinquefasciatus. F. tularensis is the causative agent of tularemia and is transmitted by ticks, deer flies, and mosquitoes. Mosquitoes are believed to play a role in transmission of tularemia in Europe. Hence, using C. quinquefasciatus as a model system, we wanted to test the effects of acquisition of F. tularensis LVS biofilms on the larvae. Our results demonstrate that F. tularensis LVS can form biofilms in media as well as persist in simulated natural water. The persistence of the bacteria is important as the bacteria can now serve as food source for the aquatic mosquito larvae. The results also show uptake and localization of F. tularensis LVS within the mosquito larvae suggesting a potential mechanism for bacterial persistence in the environment. The data further shows effects of

F. tularensis LVS ingestion on mosquito life history traits, including fecundity.

Along with the biotic components in the environment, the biofilms also interact with abiotic components that primarily include anthropogenic sources such as metal contaminants, and release of nanoparticles in the aquatic system. Naturally occurring nanoparticles are generated due to mineral weathering and occur ubiquitously. Thus, they

are likely to interact with bacterial biofilms in the aquatic environment. In this study, we investigated the abiotic interactions of L. pneumophila biofilms by studying their interaction with nanoparticles and biocides. L. pneumophila is the causative agent of

Legionnaires’ Disease and affects the elderly and immune-compromised individuals.

Legionella biofilms in industrial cooling towers pose a serious health risk due to generation of aerosols that can contain bacteria. Previously, our lab showed a change in the morphology of Legionella biofilms upon exposure to 18 nm gold particles. We hypothesized that combination of nanoparticle exposure followed by biocide treatment will significantly reduce the biomass of the biofilms. Upon exposure to nanoparticles followed by bleach, a significant reduction in the biomass of the biofilms was observed.

The results also suggest that, although not significantly different, the combination of nanoparticles and chlorine dioxide delays the regrowth of the biofilm. Studying the impact of both biotic and abiotic interactions with biofilms is vital in understanding persistence and dissemination strategies of infectious bacteria.

iii

DEDICATION

This work is dedicated to my parents who have struggled hard in life to give me this opportunity and family who have supported me in all my endeavors.

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Tamara McNealy, for giving me the opportunity to conduct research in her lab, and for her patience in mentoring me over the past 5 years. She has constantly motivated me and without her help and belief in my abilities, this dissertation would not have been possible. I would also like to thank my committee members, Dr. Matthew Turnbull, Dr. Peter Adler, and Dr. Susan Chapman for their knowledge and help. Special thanks to Dr. Matthew Turnbull for his guidance, constant support and encouragement. Thanks to Dr. Terri Bruce, the manager of the

Jordan Hall Imaging Facility, where all of my microscopy was performed, for her help and expertise in the field of microscopy. I want to specially thank the past lab members

Tara Raftery and Jonathan Gravgaard, for assistance in daily lab tasks, for their friendship and inspiration. I also want to thank my fellow graduate students Brennen

Jenkins and Andrew Harmon for their support and friendship, for helping me with lab work from time to time. Thank you to all of the undergraduates who helped with the daily maintenance of the lab.

I would like to acknowledge Clemson University, the Clemson University

Research Fund, Creative Inquiry, and the Calhoun Honors College for funding and support. Finally, I am very grateful for my kind and loving husband Neeraj, who is my source of inspiration and support, my sweet little daughter Anushka, my parents, my brother Rohit, my parents in-law and friends in India who have consistently supported me

through their immense love, reassurance and motivation, helping me to succeed as a graduate student.

TABLE OF CONTENTS

Page

TITLE PAGE ...... i ABSTRACT ...... ii DEDICATION ...... iv ACKNOWLEDGEMENTS ...... v LIST OF FIGURES ...... viii LIST OF TABLES ...... x INTRODUCTION ...... 1 FRANCISELLA TULARENSIS LVS INTERACTIONS WITH CULEX QUINQUEFASCIATUS LARVAE ...... 16

INTRODUCTION ...... 17 MATERIALS AND METHODS...... 21 RESULTS ...... 26 DISCUSSION ...... 35 INTERACTION OF LEGIONELLA PNEUMOPHILA BIOFILMS WITH GOLD NANOPARTICLES ...... 41

INTRODUCTION ...... 42 MATERIALS AND METHODS...... 52 RESULTS ...... 56 DISCUSSION ...... 62 CONCLUSION ...... 68 REFERENCES ...... 76

vii

LIST OF FIGURES

Figure Page

1. Illustration of the key steps in biofilm formation ...... 3

2. F. tularensis LVS establishes biofilms in media and MHW ...... 27

3. COMSTAT analysis F. tularensis LVS biofilms in media and MHW ...... 28

4. Culex quinquefasciatus feeds on both planktonic and biofilm F. tularensis LVS ...... 29

5. F. tularensis LVS localizes in the midgut cells and Malpighian tubule cells of the larvae ...... 31

6. Isolation of hemocytes from either control larvae or planktonic F. tularensis larvae...... 33

7. Standard curve using F. tularensis LVS genomic DNA...... 34

8. Treatment of L. pneumophila biofilms with 0.5mg/L chlorine ...... 45

9. Sequential treatment of biofilms with 18nm AuNPs and chlorine ...... 51

10. Illustration of treatments of L. pneumophila biofilms used in the biocide assay ...... 54

11. Sequential treatments of L.pneumophila biofilms with 18nm AuNPs and chlorine dioxide ...... 586

viii

List of Figures (Continued)

Figure Page

12. Combination treatment using bleach ...... 59

13. Combination treatment using chlorine dioxide ...... 61

LIST OF TABLES

Table Page

1. Treatments explaining the biocide assay set up ...... 50

CHAPTER ONE

INTRODUCTION

Overview of biofilms

Structure of biofilms

Composition of the matrix

Biofilms and disease

Overview of biofilms

Biofilms are communities of microorganisms embedded in a sticky substance called extrapolymeric matrix (EPS). These communities primarily include bacteria, fungi, and protozoa. Interest in studying biofilms is growing after the realization that the biofilms are ubiquitous and colonize every natural and man-made niche. Biofilms relevant to medical industry, such as those on medical devices and implants, have gained further importance as the biofilms are resistant to antibiotics. Depending on the type of bacteria and the environment, the bacteria within the biofilms can be up to 1000 times more resistant to an antibiotic (1). In aquatic environments, biofilms serve as reservoirs for persistence of bacteria and act as protective mechanisms against harsh environmental conditions. Some bacteria that exist within the biofilm niche and are related to disease include Pseudomonas aeruginosa that causes cystic fibrosis, Mycobacterium ulcerans, the causative agent of buruli, Staphylococcus aureus that forms biofilms on medical implants and devices, Francisella tularensis responsible for tularemia, and Legionella pneumophila that causes Legionnaires’ Disease.

Biofilm formation is a dynamic process and begins when planktonic cells attach to a surface and the initial reversible attachment becomes irreversible (Figure 1). The bacteria within the biofilm start replicating and the biofilm grows, attracting numerous community members including other bacteria, fungi and protozoa. As the biofilm matures, EPS is secreted which provides channels for nutrient exchange and quorum sensing. Biofilm formation protects the bacteria from disinfectants and antibiotics.

Dispersal of the biofilm allows the bacteria to colonize a new surface and start this process again.

Biofilms are the natural niche of the majority of microbes. They are found in all natural and human environments including glaciers, old rocks, and vents deep within the sea. Along with their association with abiotic environments, they also form on biotic surfaces of microorganisms, plants, and animals. Biofilms serve as the base of the food chain as food eaten by higher organisms such as protozoa and insect larvae, which are in turn eaten by fish. Many biofilm bacteria have defense mechanisms to protect them from predation by invertebrate hosts and some of these mechanisms depend on the morphology of the biofilm. Understanding the morphology of biofilms is a key step in studying the biofilm ecology and interaction with other environmental components. The morphology

of biofilms depends greatly on the environment in which it is present. Some of the key factors that influence the morphology of biofilms are presented in the next section.

Structure of biofilms

Information about different biofilm morphologies is usually obtained by studying single or dual species biofilms grown in lab. Pseudomonas has been used as a model organism for biofilm studies due to the robustness of biofilm formation and availability of genetic tools. These studies, and those conducted with other species such as

Escherichia coli, Vibrio cholerae, and Staphylococcus aureus, have revealed that biofilm morphology has certain typical phenotypes. Biofilm morphology can be described as mushroom and pillar like (irregular peaks), flat (low peaks, high surface area) or have a filamentous. The morphology depends on the environment in which the biofilms are growing. Filamentous structures are generally found in flowing waters such as drainage pipes or rivers. Biofilms with flat topology covering a large surface area are often observed in single species lab conditions. The mushroom body structures are found predominantly in static biofilms (2)

A variety of intrinsic factors such as various types of motility (flagellar, swarming, twitching) and presence of quorum sensing molecules affect the morphology of the biofilm (3).Type IV pili of Pseudomonas aeruginosa play a role in twitching motility and biofilm formation. Mutants for type IV pili (Δ pilA) establish biofilms with different morphology than the wild type strains. Further analysis demonstrates that flagellar mutants of P. aeruginosaPAO1 strain form biofilms that are similar to those

formed by type IV pili mutants suggesting that flagella are involved in the topology of biofilms (4). Pili or fimbriae also play a role in biofilm formation in E. coli and V. cholerae biofilms (5), (6).

Biofilm morphology in bacteria is also influenced by nutritional cues and quorum sensing. In P. aeruginosa, the biofilm morphology is influenced by changes in carbon sources. P. aeruginosa grown in a medium with succinate as a carbon source developed biofilms with a homogeneous flat topology, whereas supplementation with glucose led to biofilms containing large aggregates of cells (4). Carbon source requirement for the biofilm formation can be regulated by quorum sensing cues. Biofilms of quorum sensing mutants were irregular in the presence of either succinate or glucose (7). Another study showed that quorum sensing mutants of P. aeruginosa showed a significant difference in the biofilm morphology compared to the wild type when grown in presence of glucose

(8). V. cholerae quorum sensing mutants also displayed an altered biofilm phenotype (9) compared to wild type biofilms. Quorum sensing mutants in V. cholerae produce thicker biofilms than the wild type bacteria due to overexpression of Vibrio polysaccharide (vps) genes (10). Serratia liquefaciens, an opportunistic pathogen, is capable of infecting plants and animals and survives in soil and water. Quorum sensing plays a critical role in the biofilm morphology in S. liquefaciens, as wild type bacteria (strain MG1) form a mature biofilm with thick aggregates of cells and quorum sensing mutants exhibit a thin, single layered, immature biofilm (11).

Extrinsic factors such as presence of surfactants and temperature also affect biofilm morphology (3). Rhamnolipids (surfactants produced by P. aeruginosa) play a role in maintaining the biofilm 3D architecture in P. aeruginosa PAO1. Mutants do not form robust and well-structured biofilms like the wild type P. aeruginosa PAO1 and lack the channels that are characteristic of the wild type biofilms (12). Temperature also plays a role in the morphology of biofilms. In Legionella pneumophila, biofilms at 37°C form a multinucleate, mycelial mat like structure with filamentous bacteria, but at 25°C, the bacteria were shorter and rod shaped (13). Based on the above examples, it is evident that multiple factors within the external environment affect biofilm morphology.

Intrinsic and extrinsic factors also affect the dispersal of biofilms. Certain environmental conditions that no longer favor residing within the biofilms can induce dispersal events. This dispersal allows release of bacteria to colonize new, more favorable environments. A well-known determinant of biofilm dispersal is nutrient availability.

Early studies showed that when Acinetobacter biofilm that was grown under low nutrient conditions was switched to a rich growth medium, it led to a widely dispersed cell arrangement (14). On the other hand, in P. putida biofilms, removal of citrate leading to carbon starvation induced dissolution of biofilms (15). In a minimal medium, an increase in the concentration of various carbon and/or nitrogen sources such as glutamate, succinate, citrate, glucose, and ammonium chloride led to P. aeruginosa biofilm dissolution (16).

Some bacteria can trigger dispersal events by producing different molecules. The oral pathogen Aggregatibacter actinomycetemcomitans produces Dispersin B, a protein that hydrolyzes the glycosidic linkages of poly-β-1, 6-N-acetylglucosamine (PNAG) and triggers biofilm dispersal (17). Secreted bacterial nuclease, NucA induces biofilm dispersal in marine biofilms (18).

The above examples highlight that quorum sensing, motility, and nutritional cues play a significant role in the biofilm morphology and dispersal. Understanding biofilm morphology will aid in understanding bacterial ecology. The outermost layer, the EPS, is the first to interact with any environmental components that come in contact with the biofilm. It is crucial to understand the organization of the biofilm matrix, which is described in the next section.

Composition of the matrix

The biofilm matrix acts as a barrier to antimicrobials, disinfectants and any other harsh environmental conditions, thereby protecting the bacteria within the biofilm. The extrapolymeric matrix (EPS) is composed mainly of water, proteins, and polysaccharides that provide the biofilm with a stable network that promotes nutrient exchange and cell- cell signaling. Polysaccharides, proteins and DNA are the primary components of the matrix and are described below.

i. Exopolysaccharides

Exopolysaccharides are the primary component of a majority of biofilms. One of the most commonly studied matrix exopolysaccharides is Β-1-6-N-acetyl-D-glucosamine

(PGA or PNAG) and many bacteria produce this polysaccharide (3). Escherichia coli require PNAG for both surface attachment and formation of biofilms. Attachment to surfaces is blocked if mutations are in this locus. Disruption of this locus with metaperiodate resulted in dispersion of biofilm associated cells implying a role for this polysaccharide in cell-cell as well as cell-surface adhesion (19).Staphylococcus aureus and S. epidermidis produce PNAG or the closely related polysaccharide intracellular adhesin (PIA) and both are synthesized by the action of four proteins encoded by genes organized on a single operon, icaADBC(20),(21). PIA and PNAG serve as adhesins and play a critical role in biofilm formation for Staphylococcus. A PIA-like polymer is produced by Yersinia pestis and plays a role in the transmission of the bacterium via the flea. The bacteria form biofilms in the flea digestive tract and are transmitted to humans upon biting. The block produced in the flea digestive system by Yersinia requires genes on hms locus which shows a sequence similarity to ica locus. Thus, biofilm formation by

Yersinia aids in its transmission to humans.

Polysaccharide production plays an important role in the pathogenesis of P. aeruginosa. This pathogen colonizes the lungs of patients with cystic fibrosis (CF). Initial lung colonization occurs by the non-mucoid forms of Pseudomonas, which then convert to a mucoid form (22). The mucoid phenotype is due to overproduction of alginate, a polymer of β-1-4 linked mannuronic acid and glucuronic acid (23). When biofilms made by the non-mucoid PAO1 strain are compared to an isogenic strain that produces a

mucoid phenotype as a result of deregulation of the alginate synthesis genes, the biofilm is 1000 times more resistant to the antibiotic tobramycin (24).

The genes associated with Vibrio cholerae exopolysaccharides are found in two operons of vpsA (A to –K) and vpsL (L to –Q). Both of these are required for synthesis and export of the VPS polysaccharide (25), (26). Activation of vps genes is correlated with biofilm formation as well as rugose colony morphology. Comparisons of the polysaccharide isolated from rugose colonies and from the biofilm matrix show a difference in composition. Carbohydrate analysis performed on a non-rugose strain of V. cholerae O139 shows presence of N-acetylglucosamine, glucose, galactose and mannose

(27) while that isolated from the rugose V. cholerae El Tor shows presence of mainly glucose and galactose(25).

Cellulose is another exopolysaccharide that is commonly present in the plant kingdom and found to play a role in biofilm formation especially in Enterobacteriaceae family (Escherichia coli, Salmonella species, Klebsiella species). Cellulose is an important component of the extracellular matrix in E. coli and S. typhimurium (28), (29).

Studies in Agrobacterium tumefaciens reveal association between virulence, biofilm formation and presence of cellulose. Mutants deficient in cellulose production showed a reduction in biofilm formation and root colonization. Overproduction of cellulose in the mutant restored biofilm formation and partial root colonization (30). In addition to cellulose and PNAG, E.coli can also produce colanic acid. Colanic acid is a complex,

branched polymer. It is important to build the characteristic, complex three dimensional structure of the biofilm that the mutants are unable to produce (31).

Lectins are sugar binding proteins and can facilitate cell-cell interactions by binding to polysaccharides of the matrix or sugar moieties on surfaces of other cells (3). In P. aeruginosa, two lectins that are thought to play a role in biofilm formation are LecA and

LecB (32). LecA is specific for D-galactose and LecB is specific for L-fucose and its derivatives. LecB is important for biofilm formation by promoting cell-cell interactions by binding to the surface of biofilm cells. Binding occurs as a consequence of its interaction with fucose containing ligands on cell surface (33). The above results are consistent with an expression profiling study that demonstrated that both lecA and lecB were induced in biofilms of P. aeruginosa (34). V. cholerae El Tor utilizes a secreted protein, RbmA, to build robust biofilms. RbmA is also necessary for the rugose phenotype of V. cholerae El Tor. Mutant biofilms of V. cholerae El Tor A1552 are more sensitive to sodium dodecyl sulfate than the wild type. RbmA plays a role in maintaining the integrity of Vibrio biofilms (35).

ii. Proteins

Proteins within the biofilm matrix are adhesive molecules that help maintain a stable biofilm structure. Curli fimbriae are proteinaceous appendages that contribute to the adhesive properties in E. coli and Salmonella species (36). In E. coli, cell-cell contacts and cell substratum interactions are mediated by curli (37). In S. typhimurium, a curli

mutant was severely compromised in biofilm formation (38). P. aeruginosa is also thought to use fimbriae as a part of its biofilm matrix. Mutants that cannot synthesize

CupA (chaperon-usher pathway, genes involved in novel assembly of fimbriae) fimbriae make weak pellicles (liquid-air interface biofilms) that are easily disrupted (39). Several transcriptional profiling studies have shown that fimbriae/ pilus gene expression is upregulated in biofilms compared to planktonic structures. This supports the idea that fimbriae and pili may be thought of as proteinaceous structural components of the matrix

(40), (41). Legionella pneumophila, Francisella tularensis, and Proteus species also use fimbriae to maintain biofilm integrity

The family of proteins known as bap proteins (named after Staphylococcus aureus bap, biofilm associated proteins) is a multi-domain protein family that shares similarity in a number of bacterial species and are involved in promoting biofilm formation (42). In S. aureus V329 strain, disruption of bap causes inhibition of surface accumulation and intracellular adhesion. Moreover, human clinical isolates that have the bap gene form thicker biofilms (43). In some situations, presence of Bap and other proteins eliminate the need for exopolysaccharides in the biofilm matrix (3). In presence of Bap, inactivation of icaADBC operon (involved in synthesis and export of PNAG) does not cause a reduction in the biofilm formation (44). Many S. aureus strains do not possess the icaADBC operon but can still form robust biofilms (45). Enterococcus faecalis protein, Esp (entorococcal surface protein) shows 33% sequence similarity to

Bap and is involved in biofilm formation in vitro (46).When Esp is expressed on the surface of Esp deficient strains, biofilm formation was significantly increased (47). The

involvement of Esp in biofilm formation was further confirmed when heterologous expression of Bap in Esp deficient and biofilm deficient strains of E. faecalis restored the capacity to form biofilms (48). In P. fluorescens and P. putida, LapA is a secreted protein similar to Bap. Mutations in lapA impairs adhesion as well as formation of biofilm in both P. fluorescens and P. putida on biotic and abiotic surfaces (49), (50). Therefore, proteins play an important role in biofilm adhesion and intercellular cementing. iii. DNA

Extracellular DNA (eDNA) is the third major component of the biofilm matrix (3).

Extracellular DNA has been found in significant amounts in the biofilms of soil bacterium Bacillus cereus, respiratory pathogen Bordetella bronchiseptica, and

Streptococcus mutans among others (51),(52),(53).The eDNA in the matrix of P. aeruginosa biofilm is required for maintaining the integrity of biofilms (54), (55).

Addition of DNase to the medium inhibited biofilm formation by P. aeruginosa and preformed biofilms were disintegrated upon addition of DNase (54). In Enterococcus faecalis, chromosomal DNA is released as a consequence of autolysis of cells from within the biofilm (56) and incorporated into the EPS. The extracellular protease GelE is responsible for this autolysis and can be inhibited by another extracellular protease, SprE

(56). GelE mutants show significant reduction in biofilm formation, while absence of

SprE increases the rate of biofilm formation. This increased rate was attributed to increase in autolysis of cells leading to increased amounts of eDNA (56). eDNA is an important part of the matrix for S. aureus as well, and quantitative real time PCR shows

that the source of eDNA here is also chromosomal DNA (57). A mutant in cidA (involved in regulating cell lysis and release of eDNA) exhibits decreased cell lysis resulting in defective biofilms due to decreased amounts of eDNA (57). Extracellular DNA is required for initial attachment and biofilm formation in 41 strains of Listeria monocytogenes. DNase treatment of Listeria biofilms results in biofilm dispersal, confirming the essential role of extracellular DNA in the matrix (58).

In many cases, it is difficult to isolate, purify and identify the exact composition of the EPS from a bacterial biofilm. Additionally, the EPS composition of biofilms of different organisms will be affected by the growth conditions and strains of the bacteria involved. This analysis is difficult to conduct with environmental biofilms as these usually involve mixed species. Therefore, it is important to remember that most information on the matrix has been obtained from studying model organisms in single species biofilms.

The properties of the EPS confer protection to bacteria against harsh conditions and allow persistence of bacteria in the environment. Prolonged persistence of pathogens within biofilms can allow and increase the potential for human exposure. The next section describes pathogenic species in environmental biofilms that lead to human disease.

Biofilms and disease

Biofilms play a key role in the ability of pathogens to colonize environmental niches. Human pathogens within these biofilms are protected from antimicrobials,

disinfection strategies and other environmental stresses. Campylobacter jejuni exists in the environment in biofilms. This bacterium, among others, is responsible for human bacterial gastroenteritis. Consumption of contaminated water and poultry are the most common routes of infection. In microcosm experiments, C. jejuni survived until the experiment was terminated (up to 42 days) at 30°C and 4°C. When rRNA fluorescent probes were used to detect the bacteria, longer persistence times in biofilms were observed than traditional culturing. Thus, biofilms play an essential role in the survival of this organism outside the human host and allow prolonged persistence of this pathogen in the environment (59).

Vibrio cholerae is also an environmental pathogen. The bacteria form biofilms on copepods, crustaceans with chitinous exoskeletons, a variety of zooplankton and phytoplankton. In particular, the bacteria form biofilms on copepods, part of the marine zooplankton community (60). Zooplankton-associated V. cholerae biofilms survive longer in sea water than planktonic cells alone. Zooplanktons are thought to be the principal reservoir of V. cholerae in the marine ecosystem (61). V. cholera resides in the gut and on the surface of zooplankton where it is not only protected from the external environment but can also use the improved nutrient conditions and thrive (60), (62). A colonized copepod can contain up to 104 bacteria at a time; thus a bloom in the copepod population can result in a high number of V. cholerae per ml of water that constitutes an infectious dose (63).

Numerous other bacteria require biofilm forming ability for survival, persistence or transmission. Biofilm formation of Yersinia pestis in the flea vector is essential in the transmission of this pathogen to new hosts (64). Biofilm formation in Salmonella typhimurium is enhanced on abiotic surfaces outside hosts (1). Legionella pneumophila also exist in aquatic environments as biofilms that allow prolonged persistence of the bacteria.

Bacterial and specifically, pathogen persistence within biofilms can contribute to disease ecology, transmission dynamics and community structure. Biofilms interact and respond to biotic and abiotic components in the aquatic ecosystem. Studying detailed interactions provides understanding of biofilm survival and persistence mechanisms which then allow for development of optimal prevention and treatment options for biofilms. In aquatic environments, biotic connections of biofilms involve interactions with protozoa and insects. In the next chapter, the study of the interaction of the biofilms of a pathogenic bacterium Francisella tularensis LVS with the mosquito larvae of Culex quinquefasciatus are described.

CHAPTER TWO

FRANCISELLA TULARENSIS LVS INTERACTIONS WITH

CULEX QUINQUEFASCIATUS LARVAE

Introduction

Materials and Methods

Results

Discussion

This chapter was published as:

Mahajan, U. V., Gravgaard, J., Turnbull, M., Jacobs, D. B., & McNealy, T. L. (2011).

Larval exposure to Francisella tularensis LVS affects fitness of the mosquito Culex quinquefasciatus. FEMS microbiology ecology, 78(3), 520-530.

Introduction

Francisella tularensis is an intracellular pathogenic bacterium responsible for causing tularemia, commonly known as rabbit fever. Human cases are rare in the US, but occur sporadically. About 200 cases are reported each year in the USA from 1990-2000

(65). Cases have occurred recently in Martha’s Vineyard (66), Japan (67), and northern

China (68) but currently the majority of them occur in Finland and Sweden (69). Human infections arise by inhaling aerosols, drinking water contaminated with the bacteria, hunting and skinning of animals infected with Francisella, or through arthropod bites.

The organism has been isolated from over 250 species of birds, reptiles and mammals

(70). Rodents, rabbits and beavers act as reservoirs for Francisella in the environment but most likely the primary niche of the bacterium is a biofilm (71), (72). Protozoa play a role in the survival of this pathogen probably through interaction with the biofilm. Francisella tularensis is also able to persist within amoebal cysts. Biofilms and amoebae then offer the possibility for long term survival of F. tularensis in the environment during inter- epidemic periods (73).

There are two main subspecies of F. tularensis: F. tularensis tularensis (type A) and F. tularensis holarctica (type B). F. tularensis tularensis is primarily terrestrial in association while F. tularensis holarctica is more water associated (74). Type A causes infections in North America and is highly associated with tick bites; contrastingly, type B is responsible for more widespread infections across the entire Northern hemisphere and is transmitted by ticks, deer flies and mosquitoes (74). A third type, F. novicida is

generally not considered to be a human pathogen but causes a similar disease in mice and hence is a good model for studying tularemia. The bacteria most likely survive in the environment by forming biofilms. In 2010, Durham-Colleran and colleagues demonstrated the ability of F. novicida to establish biofilms (75). F. novicida also uses chitinases to breakdown chitin and can grow on surfaces containing chitin. Mutants lacking chiA and chiB cannot form biofilms (76).

Arthropods such as mosquitoes, deer flies and ticks are predominant vectors for

Francisella (74). Circumstantial evidence suggests that mosquitoes play a role in transmission of tularemia in Europe (69), (77). In Scandinavia, the majority of tularemia cases are believed to be due to mosquito bites (69). Ten species of mosquitoes have been found to naturally carry F. tularensis, possibly through acquisition of blood meals from infected animals (78). Mosquito larvae however, are also a possible source as they are occupants of aquatic ecosystems and feed on bacteria. Culex quinquefasciatus larvae are found in water with high bacterial loads and organic matter. This genus has a wide range across the northern hemisphere. It is a known vector of several pathogenic agents including West Nile Virus, Western Equine Encephalitis Virus, St. Louis Encephalitis

Virus and Wuchereria bancrofti (causative agent of lymphatic filariasis). The range of this mosquito species, as well as of the mosquito genera Aedes, overlaps with the distribution range of F. tularensis (79). Therefore, mosquito larvae feeding on F. tularensis biofilm could result in persistence of the bacteria in adults and for possible transmission.

Feeding habitats of mosquito larvae are well studied. Their diet mainly consists of bacteria and other detritus in the aquatic environment (80). Historically, mosquitoes have been shown to be involved in mechanical transmission of F. tularensis (81). Recent work with mosquito cell lines, however, demonstrates that F. novicida can invade and replicate within arthropod cell lines (82), (83). F. novicida was found to replicate in Sual B hemocyte-like cells and proliferate in Drosophila S2 cells (82), (83). Hemocytes are phagocytic cells of the insect immune system similar to mammalian macrophages and during human infection F. tularensis invades and replicates within macrophages (73).

Other studies used the intra-hemocoelic injection of live bacteria within arthropods such as Galleria mellonella and D. melanogaster (84)(85). In the G. mellonella caterpillar model, F. tularensis LVS killed the caterpillars and the rate of killing was dose dependent. Administration of antibiotics to the caterpillars, before injecting the bacteria, delayed death of the caterpillars (84). The bacteria were able to replicate in vitro and in vivo in D. melanogaster hemocytes (85). Although the above studies are informative for analysis of virulence traits, they tell us little about the interaction after a normal environmental exposure of larvae to the bacteria. The most likely interaction of the larvae and the bacteria in the aquatic environment will occur via ingestion of the bacteria by the larvae.

In addition to what happens from the pathogen side after interaction between larvae and biofilms, the consequences of the effect of feeding on microbes on the larvae are also of interest. The interaction of mosquitoes with Francisella can not only contribute not only to pathogen persistence in the environment but could also have effects

on the mosquito ecology. Bacteria in the environment influence mosquito life history traits such as time to pupation, wing length, fecundity and mortality. Their effects may be apparent immediately in the larval stage or may not be seen until later in adult life.

Stresses incurred in juvenile life stages can alter and impact the adult organism. Bauer fiend et al (86) demonstrated that larval food stress in Bicyclus anynana butterflies led to reduction in pupal mass, adult size, an increase in larval developmental rate, and reduction in adult wing lengths (86). A similar effect was observed in C. quinquefasciatus wing length when larvae were raised in presence of a large number of larvae and insufficient space for development (87). Ingestion of Pseudomonas entomophila kills larvae and adults of Drosophila (88) and adult survival, locomotory behavior and fecundity are impacted in the Drosophila parasitoid Leptopilana heterotoma infected by Wolbachia (89).

The analysis of the interaction of F. tularensis and C. quinquefasciatus larvae is therefore twofold. First, we hypothesize that oral acquisition of pathogens by larvae will affect the persistence and survival of the pathogen. Second, we predict effects on fitness traits of the mosquito. The development of an oral acquisition model for biofilms and insect larvae will allow for the study of pathogen uptake and persistence in larvae and provide the opportunity to examine the effects of larval exposure to pathogens on adult fitness in a natural environment.

Materials and Methods

Organisms and plasmids: Francisella tularensis subspecies holarctica LVS (live vaccine strain; provided by Dr. Karen Elkins, FDA) was cultured on chocolate agar (CA) supplemented with Isovitalex and grown at 37°C with 5% CO2. Bacteria were used from a 3-day-old agar plate. The GFP expressing plasmid, pKK214 (provided by Dr. Anders

Sjostedt, University of Umea, Sweden), was used to transform F. tularensis LVS and maintained in experiments using 10µg/ml tetracycline. Mueller Hinton broth (MHB) supplemented with Isovitalex, 0.33mM ferric pyrophosphate, 5.55mM dextrose, 0.71mM calcium chloride and 0.95mM magnesium chloride was used to establish bacterial biofilms. Sterile moderately hard water (90) was used in exposure experiments and for rearing of mosquitoes. The mosquito, Culex quinquefasciatus, was reared in the arthropod containment facility at Clemson University at 27°C using previously published guidelines (91). Sterile dog biscuit slurry was used to rear larvae; adults were blood fed using cow blood purchased from the Godley Snell Animal Research Facility at Clemson

University.

Biofilm establishment: F. tularensis LVS biofilms were established in 4-well Labtec

^8 chambers (Nunc) by adding 1ml of a bacteria suspension adjusted to OD600=0.18 (~10 bacteria/ml) in MHB to each well. Biofilms were incubated at 37°C for 5 or 15 days.

Media was changed every 5 days to provide fresh nutrients. For imaging, biofilms were washed twice with ultrapure water to remove non-adherent bacteria and stained with the nucleic acid binding dye Syto 11 at 3µM (Invitrogen) for 30 minutes in dark. Syto 11 is a

permeable dye, resulting in the staining of all cells in the biofilm. After staining, biofilms were washed again with ultrapure water and imaged using a Nikon TiE Eclipse confocal microscope. To analyze the persistence of biofilm in natural water, biofilms were set up as described, but the initial suspension was made in moderately hard water (MHW). 1ml of this suspension was added directly to the chamber slide and imaged at the same time points. Biofilms were then quantitatively analyzed using COMSTAT (92), which calculated the biomass, maximum thickness, and roughness coefficient. The experiment was performed three times and statistical analysis performed on COMSTAT data using

ANOVA.

Mosquito biofilm interaction: F. tularensis LVS-GFP was used in all assays. Bacteria were provided to mosquito larvae as either planktonic or biofilm stage. For biofilms, large chambers containing two sterile glass slides were used. Bacteria were suspended in

MHB at OD600=0.3 and 5ml of this suspension was added to 20ml MHB supplemented with 10µg/ml tetracycline. The chambers were incubated at 37°C with 5% CO2 for 5 days. The biofilms were then washed twice with 10ml MHW to remove any unattached cells. Next, 25 ml of sterile filtered MHW was added to the chambers and the biofilms then incubated at 27°C for 24 hours to allow equilibration of the biofilm prior to their use in mosquito interaction experiments. Biofilms were then washed a final time with 10ml

MHW to remove traces of tetracycline and fresh sterile 25ml MHW was added. Third instar C. quinquefasciatus larvae were collected from the lab colony and starved for 24 hours in sterile MHW at 27°C to remove all food traces from the gut tract. Ten larvae were then placed in a container containing established biofilms. For planktonic assays,

ten larvae were placed in a beaker containing 47ml sterile MHW and 3ml of a planktonic suspension of F. tularensis LVS (OD600=0.5 in MHW). For controls, ten larvae were placed in a sterile beaker containing 50 ml sterile MHW and 150 µl sterile dog biscuit slurry. Larvae were exposed to either a single, 24 hour exposure of the bacteria or to continuous exposure of the bacteria. After the single exposure, larvae were transferred to fresh beakers containing 50 ml sterile MHW and 150 µl dog biscuit slurry until pupation.

In continuous exposure experiments, larvae were provided control, planktonic or biofilm conditions as described above with the food source being replaced every two days until pupation.

Microscopy-Uptake Assay: The experimental cohorts (control, planktonic, biofilm) were collected after a single exposure and placed in separate wells of a 24 well plate on day 1 and 3 post exposure. A 1:1 mixture of 50% glycerol and 0.1% agarose was added to each well to immobilize the larvae. The solution was allowed to solidify prior to imaging. Whole larvae where then imaged using the Nikon AZ100 with Nikon Ri-1 camera. Images were captured using bright field and FITC epi-fluorescence modes.

Images were analyzed using the Nikon Elements software and control and experimental samples were processed equivalently. The experiment was repeated three times with 10 larvae in each container.

Microscopy-Bacterial Localization Assay: Cohorts of larvae were collected after single or continuous exposure and washed with ultrapure water. Larvae were then fixed with

500µl, 4% EMS grade paraformaldehyde (EMS) for 2 days. Each larva was then

dissected under a stereoscope and rinsed three times with 1X phosphate buffered saline+

0.1% Triton X 100 (PBS-T). For blocking, 2X PBS-TB (2X PBS+ 0.1% Triton X 100 +

3% bovine serum albumin) were added and larvae incubated at 4°C for 1 hour. Larvae were washed three times and the primary antibody (F. tularensis LVS anti-LPS, kindly provided by Dr. Jim Drake, Albany Medical College) was prepared in 2X PBS-TB (1:50) and added to the slide for incubation overnight at 4°C. Larvae were washed and the secondary antibody (anti-mouse FITC, Southern Biotech) added for 4 hours. DAPI was used to stain eukaryotic cell nuclei (1:100). After final washing, samples were mounted with Vectashield and imaged on the Nikon TiE microscope. Images were analyzed using

Nikon Elements software.

Hemocyte isolation: Cohorts of ten larvae from each treatment were used for isolation of hemocytes. The larvae were washed in PBS and then using the dissection microscope hemocytes were isolated and collected in Schneider’s medium and 10mM of phenylthiourea (PTU) to prevent melanization. The collecting tubes were kept on ice to delay melanization. Hemocytes were then gently pipetted on a glass slide and cover slipped using Vectashield. Images were captured using the Nikon TE2000 using the 20X objective.

Real time PCR: To quantify the number of bacteria within the larvae real time PCR assay using a hydrolysis probe (Roche: Universal probe library number 155) for the fopA gene (Francisella specific) was designed. Hydrolysis probes hybridize within the amplicon and are degraded by Taq polymerase. Upon degradation of the probe,

fluorescence proportional to amount of template is recorded. Qiagen blood and tissue kit was used to isolate genomic DNA from 102 to 107 bacteria for generating the standard curve. All PCR reactions were run in 10 µl volumes. Each reaction contained 5 µl Probes master mix (2X), 0.2µl probe, 0.4µl each forward and reverse primer, 1µl template DNA and 3µl water. Roche LC480 was used to perform the real time assay. PCR cycling was performed at 95°C for 10 seconds and the annealing temperature for probe was 60°C for

40 cycles.

Statistical Analyses: Statistical analyses were performed using SAS (9.2). ANOVA was performed on the biofilm and fitness assays. A p value less than 0.05 considered significant.

Results

Francisella tularensis LVS establishes biofilms and persists in simulated natural water environments

Although biofilm formation in vitro is well documented for Francisella novicida

(75), this has not yet been demonstrated for F. tularensis LVS. We show that F. tularensis LVS establishes biofilm in media (MHB) and grows at 37°C for at least 15 days (Fig 2a and b). We also wanted to know if F. tularensis LVS could establish and maintain a biofilm in MHW without the addition of nutrients or host cells. To test this, we incubated the bacteria with MHW and performed confocal analysis on these biofilms after 5 and 15 days. We found that the bacteria could form and persist in the biofilms for at least 15 days (Figure 2c and d).

COMSTAT analysis (92) for biofilms was performed and indicated that 15-day biofilms in MHB have more biomass than the 5 day old biofilms but due to variation across the biofilms the average of the biomass of these biofilms are not significant (Figure 3a). The maximum thickness measurements of the 15 day old biofilms however, are significantly greater than 5 day old biofilms in media (Figure 3b).

COMSTAT analysis for biofilms established in MHW (92) showed no significant difference in the total biomass or maximum thickness of the biofilms (Figure 3a and b).

The above results support the earlier results that F. tularensis LVS can establish robust biofilms in presence of media and MHW supports persistence of bacteria in absence of a host cell.

Culex quinquefasciatus can feed on F. tularensis LVS

Mosquito larvae can feed from water columns as well as surfaces. To assess if the larvae could acquire bacteria orally in both planktonic (water column) and biofilm (surface) form, we exposed the larvae to GFP F. tularensis LVS to both forms. The uptake of the bacteria was qualitatively analyzed by fluorescence microscopy. Larvae were imaged 24 and 72 hours post exposure. At both time points, fluorescence in the gut of the larvae is easily detectable in the planktonic (Figure 4b and e) as well as the biofilm (Figure 4c and f) exposed larvae and is absent in the control larvae (Figure 4a and d). As larvae were

only exposed to bacteria for 24 hours and switched back to clean MHW with dog biscuit slurry, presence of green fluorescence in the gut region at 72 hours suggests the persistence of the bacteria in the larvae.

F. tularensis LVS localizes in the Malpighian tubule cells and midgut cells of the mosquito larvae

Since the bacteria were present in the gut and knowing that F. tularensis is an intracellular pathogen, we hypothesized the localization of bacteria to the midgut cells and performed localization studies. These studies were performed to determine if F. tularensis is capable of crossing the gut epithelium making them capable of cellular localization outside of the gut lumen. Larvae exposed to a single 24h exposure of the bacteria show intracellular presence of F. tularensis LVS in both the midgut (MG) and

Malpighian tubule (MT) cells (Figure 5a and c). 72 hours after exposure the bacteria remained localized in the MT cells (Figure 5d). The presence of bacteria in these cells

(especially apical MT) suggests possible translocation of the bacteria from the gut to the hemocoel of the larvae. When exposed continuously to F. tularensis LVS, the bacteria localized prominently to the cells of the Malpighian tubules (Figure 5e).

Isolation of hemocytes from larvae fed with planktonic GFP F. tularensis LVS

To assess whether the bacteria cross the gut epithelium, enter the hemocoel of the larvae and interact with the hemocytes, we attempted to isolate the hemocytes from the control larvae and larvae exposed to planktonic GFP F. tularensis LVS. However, we could not proceed with the analysis as melanization of the hemocytes occurred under five minutes of isolation in spite of addition of phenylthiourea (PTU) which should delay melanization. Some success in isolation of hemocytes was obtained after repeated trials.

However, as seen in Figure 6, the control (b) as well as the treated (d) hemocytes shows the presence of green fluorescence. Hence, no conclusions could be drawn from this experiment.

Real time PCR analysis

To determine the number of the bacteria persisting or replicating within the larvae following single and continuous exposures, we performed real time PCR analysis. An increase in the number of bacteria from day1 through pupation would indicate replication of Francisella while reduction or maintenance of the same number of bacteria would suggest non- replicative association. A standard curve was generated using genomic

DNA isolated from 102 bacteria to 107 bacteria. A no template control (water) and mosquito control (DNA from larvae with no Francisella) as well as a positive control

(Francisella genomic DNA) were included in the run. The efficiency of the reaction was

1.956 which implies very low variation among each sample Crossing point (Cp) is a point in the PCR reaction when the sample fluorescence crosses the threshold. When this curve was used along with experimental samples we could not detect difference in Cp values of control larvae and larvae exposed to F. tularensis LVS. The likely explanation for inconclusive results is that the significantly low ratio of bacterial DNA to mosquito DNA, making it difficult to detect F. tularensis in samples.

Discussion

Biofilm formation is a key survival strategy for many bacteria. Bacteria such as

Vibrio cholerae, Legionella pneumophila and Francisella tularensis persist in the environment within the biofilms (93). F. tularensis LVS is a capable biofilm former in both media and moderately hard water (MHW). In our experimental set up, biofilms of F. tularensis LVS persisted for at least 15 days. Although these bacteria grow in laboratory media, F. tularensis is an intracellular pathogen and requires a host cell to replicate in the environment. Persistence of the biofilms in moderately hard water (MHW) is important because it demonstrates that the bacteria can survive in the environment in absence of a host cell and the persistence within the biofilms may allow the bacteria to interact with protozoa and other aquatic insects that can serve as host cells.

Many bacteria have protective mechanisms to protect against protozoan grazing.

Pseudomonas aeruginosa biofilms formed grazing-resistant micro-colonies in presence of the surface feeding flagellate Rhynchomonas nasuta. Absence of the predator caused no change in the morphology of the biofilm (94). In another study, grazing effects of on marine bacterial biofilms were tested using nano-flagellates. Results demonstrated secretion of an anti-protozoan chemical, violacein produced predominantly within biofilms that protects the biofilms from grazing by nano-flagellates (95). Grazing by flagellates on planktonic Vibrio cholerae eliminated planktonic bacteria but the biofilm population was protected (96). The above examples highlight the different outcomes on the population dynamics of biofilm communities.

Other organisms besides protozoa graze on bacterial biofilms. Our results show that the mosquito C. quinquefasciatus can feed on F. tularensis LVS in both planktonic and the biofilm form. The green fluorescence from the bacteria carrying the GFP plasmid is present even 72 hours post-exposure suggesting that the bacteria persist within the larvae for at least that long. The feeding on the bacteria by the larvae is not surprising as mosquito larvae are known to feed on microbes and organic matter (80). However; the interaction between the larvae and bacteria could be a mechanism for persistence of the bacteria during the inter-epidemic periods. Invertebrates that feed on mosquito larvae carrying F. tularensis could allow for trophic transfer of the pathogen. As mosquitoes are possible vectors for F. tularensis, the above interaction can provide useful information about disease ecology and persistence of the bacteria in the environment.

When Tribenbach and colleagues fed Aedes aegypti and Anopheles gambiae planktonic F. novicida, the bacteria did not persist after molting (97).However; evidence suggests that mosquitoes are predominant vectors for tularemia in Europe, especially

Sweden (98). In fact, larvae reared in water endemic for F. tularensis contained bacteria that persisted through molts to the adults. In the above study F. tularensis type B was present in A. punctor, A. vexans, A. stiticus and Culex pipiens/ torrentium(99).

Aquatic insect larvae are also a food source for higher invertebrates. Presence of viable bacteria within the larvae implies a potential for the transmission of the bacteria to higher organisms. This is observed when Mycobacterium ulcerans is ingested by aquatic insects. The bacterium was then transmitted to mice through insect bites. The bacteria

were later isolated from the salivary glands of water bugs (100). Similarly, trophic transfer of F. tularensis persisting within C. quinquefasciatus larvae can occur and have similar transmission outcomes as seen in M. ulcerans. The above results suggest that mosquito larvae can play a vital role in maintenance and possibly transmission of pathogens.

Since the mosquito larvae ingested F. tularensis LVS, we wanted to examine the sites of localization of the bacteria within the larvae after the exposure. Our results show localization of bacteria within the midgut cells and Malpighian tubule cells of the mosquito larvae. F. tularensis is an intracellular bacterium and readily invades and multiplies within the alveolar epithelial cells; thus localization of the bacteria within the midgut cells is not unexpected (101). However, colonization of the Malpighian tubule cells is more surprising as bacteria must move against the current from hindgut into the

Malpighian tubules. Malpighian tubule colonization can also occur via histolysis of tissues during molt and metamorphosis. The most likely route however, is for the bacteria to cross the midgut epithelium into the hemocoel and interact with the hemocytes that could then potentially transfer them to the Malpighian tubules. The presence of the bacteria within the midgut cells suggests that this last hypothesis may be the case.

Hemocytes are the mammalian equivalent of macrophages in insects. F. tularensis is known to invade and survive within macrophages (102). F. novicida has been shown to invade and replicate within a mosquito hemocyte-like cell line (82). In aquatic mosquito larvae, chitin is a major component of the exoskeleton (cuticle) as well as the peritrophic matrix which lines the midgut epithelium and protects the mosquito from bacterial

invasion into the hemocoel. Therefore, organisms that can produce chitinases and breakdown chitin can potentially circumvent host barriers. Bacteria could also use chitinases in escaping the midgut after ingestion. Francisella possesses chitinases genes, and since we observed localization of F. tularensis LVS in the midgut cells, we can hypothesize that bacteria escape into the hemocoel and interact with hemocytes using chitinases to degrade the midgut epithelium (76), (103). We attempted to isolate the hemocytes to determine if the bacteria were present within the larval hemocytes but were unsuccessful in our isolation due to melanization.

Bacterial communities and the higher organisms feeding on them can affect community ecology. For instance, barnacles settle on a biofilm depending upon the community structure of that biofilm (104). The settlement response of Balanus amphitrite and Balanus trigonus depends on composition of the microbial community and not on the temperature or salinity of the water (104). Thus, environmental conditions can dictate the mode of choice for settlement for the two barnacles. Biofilm communities can also affect the metamorphosis of some aquatic larvae as seen in larvae of the mussel, Mytilusgallo provincialis. Pediveliger larvae (final stage of larval metamorphosis of mussels) settled as well as morphed in response to biofilms (105). Settlement and metamorphosis to the post- larval stage significantly increased in the presence of mature biofilms compared to young biofilms (105). If the biofilm was treated with UV, heat or ethanol and the conditioned water from the treated biofilms used, there was no settlement and metamorphosis of the mussels, suggesting that the mussels required cues from the live bacteria within the biofilm (105).

The relationship between the mosquito larvae and the bacteria is not mutually exclusive and uptake of F. tularensis by the larvae can have significant effects on the fitness of the mosquito. Certain strains of Pseudomonas fluorescens when fed to Drosophila melanogaster cause a dose-dependent response and delay in the development from larvae to the adults. In the same study, another strain of P. fluorescens causes lethality in the

Drosophila larvae (106). Feeding Escherichia coli and Micrococcus luteus to

Trichoplusia ni (red flour beetle) larvae negatively affected time to pupation and pupation mass (107). These stresses although occurring in the larval stages of the insect had a significant impact on adult survival. In collaborative research in our lab, we observed similar results with F. tularensis LVS-C. quinquefasciatus interaction. Larvae continuously exposed to F. tularensis LVS alone do not develop to the pupal stage (data not shown). Thus, supplementation of the continuous exposure treatments with dog biscuit slurry was necessary. Even with supplementation, in both continuously exposed biofilm and planktonic F. tularensis LVS larvae fitness decreased in the mosquitoes

(108).

The observed effect on larval fitness upon exposure to F. tularensis LVS led us to test effect of larval exposure to F. tularensis LVS on mosquito fecundity. In collaboration with a fellow graduate student, the fecundity costs associated with the bacterial uptake in control or biofilm exposed larvae were investigated. Mosquitoes were raised in either control or biofilm environments, allowed to emerge and placed in mating chambers in combinations of the two conditions. All pairings in which one of the pair had been raised in the biofilm environment resulted in significantly fewer eggs than the control. The

above results suggest that exposure to F. tularensis LVS as juveniles has significant negative fecundity costs in adult mosquito life (108). Similarly, when mosquito larvae of

A. aegypti, A. gambiae and Culex fatigans were exposed to Aspergillus parasiticus there was a significant reduction in the percentage of egg hatch in the exposed larvae (109).

These data suggest that microbial communities can have a substantial impact on the ecology of other organisms involved in the aquatic system.

The interaction between organisms of diverse trophic levels and microbial communities is not well understood but is an indispensable component of infectious disease ecology. Transient aquatic organisms such as mosquito larvae may play a significant role in the persistence and dissemination of pathogens in the aquatic environment. This work, for the first time, demonstrates that larvae can acquire pathogens from the biofilm via a natural feeding mode (oral) and the bacteria may be able to use this host as a mechanism for maintenance in the environment. Our work suggests that the pathogen may also be playing a role in larval population dynamics. Although further work is needed to corroborate the mechanisms, uptake of pathogens by mosquito larvae may provide protection and nutrients for pathogens within cells, transport of pathogens through movement of larvae and serve as a source for disease transmission.

This work emphasizes the need to study the interaction between transient aquatic organisms and environmental pathogens, not only in the laboratory but also in relation to environmental aquatic systems. Understanding the causes which influence persistence and dispersal of biofilms and the dynamics of biofilm interactions with vector species can lead to significant insight on outbreaks, transmission and prevalence of disease.

CHAPTER THREE

INTERACTION OF LEGIONELLA PNEUMOPHILA BIOFILMS WITH

GOLD NANOPARTICLES

Introduction

Materials and Methods

Results

Discussion

Introduction

Legionella pneumophila is the causative agent of Legionnaires’ Disease (LD), a potentially fatal pneumonia. L. pneumophila is ubiquitous in natural and human-made aquatic environments. The bacteria can survive in a broad range of pH and temperatures, suggesting that Legionella is capable of enduring a variety of environmental growth conditions. They can be found in a variety of human-made environments such as cooling towers, hot water systems, plumbing and air conditioning systems. The bacteria exist within biofilms in these aquatic systems. When the biofilms are disturbed, the bacteria can integrate into aerosols produced by such systems resulting in an exposure risk to immune-compromised and elderly individuals. The first identified outbreak of LD occurred in 1976 in Philadelphia at the American Legion conference (110). Since then, this pathogen has been responsible for several community acquired outbreaks of LD

(111). Diagnosed cases of LD have increased from 1110 cases in 2000 to 3522 cases in

2009(MMWR, 2011) but the number of cases is thought to be under-reported. Infection by L. pneumophila causes pneumonia and many cases of pneumonia are treated and do not warrant culturing of the causative agent.

Naturally occurring biofilms contain multiple species. L. pneumophila persists within biofilms formed by Flavobacterium species, Pseudomonas fluorescens and

Klebsiella pneumoniae and in a mixed biofilm formed by Flavobacterium and K. pneumonia (112). Along with survival within mixed biofilms the bacteria grow within amoebae. L. pneumophila actually require amoebae or other host cells for replication

(73). Growth of the bacteria within amoebae protects them from external environmental stresses when amoebae encyst. L. pneumophila can also survive within amoebal cysts for long periods until favorable conditions including, nutrient availability and appropriate temperature to grow (113). The ability of L. pneumophila to survive and replicate within the amoebae suggests that amoebae and other protozoa are reservoirs for environmental amplification of Legionella (114), (73).

Most cases of LD are attributed to inhalation of aerosols containing bacteria. For instance, an outbreak in 2006 in Pampilona, Spain was a result of contaminated aerosols generated from cooling towers (115). A 2012 outbreak in Quebec City was attributed to aerosols generated from a nearby cooling tower. Over 180 confirmed cases were reported and 18 people died (116). Prevalence of L. pneumophila in cooling towers and other human-made aquatic systems is a common phenomenon and strategies to prevent or reduce colonization of L. pneumophila biofilms are regularly employed.

Cooling towers are used for industrial cooling processes and air conditioning facilities. They are a cheap, efficient way to transfer heat to the outside environment by evaporating hot water with air steam. Cooling tower construction and function provide an ideal environment for growth of biofilms. The towers have large surface areas and a constant supply of flowing or stagnant water. Recirculation of water through these towers creates a risk of aerosol formation, which can contain microbes (117). Often times outbreaks of LD are directly linked to presence of Legionella biofilms within the cooling tower (118), (119).

Appropriate disinfection strategies for removal of Legionella biofilms from cooling towers and plumbing systems are needed on a routine basis. Several disinfectant techniques are employed for removal of biofilms such as ultraviolet light, ozone, and oxidizing and non-oxidizing biocides. Oxidizing biocides used primarily used include bromine, chlorine, iodine and chlorine dioxide. Chlorine in water forms hydrochlorous acid which is responsible for killing the organisms. The amount of chlorine added depends upon the demand (organic content), pH and temperature at which it is used

(120). Chlorine dioxide is an active biocide and is used extensively due to its high efficiency. It is a gas and must be produced on-site as it degrades quickly. Oxidizing biocides are effective and cheap; however, prolonged use of these agents causes corrosion issues within systems (120). Previous results from our lab showed no significant difference in the biofilm biomass upon exposure to 0.5mg/L chlorine alone

(Figure 8). The EPA recommended concentration for chlorine is 0.5-1 mg/L but, concentration of chlorine required to effectively kill L. pneumophila is high as 512 mg/L

(121).

Some commonly used non-oxidizing agents are quaternary ammonium salts, amines, copper salts, and organo-sulfur compounds. Quaternary ammonium compounds change the permeability of bacterial cell walls and cause internal osmotic change leading to cell death of the bacteria. These compounds are most effective in an alkaline pH range of 8 or higher but the pH in the cooling towers is often between 6 and 8. The main problem with non-oxidizing biocides is requirement of long exposure times and need for high concentration (120). Often a combination of both oxidizing and non-oxidizing biocide is essential for complete eradication of Legionella biofilms.

The activity of disinfectants is higher in planktonic cells than it is for biofilm cells. When chlorine was used as a disinfectant, 1 mg/L was sufficient to kill planktonic

bacteria but up to 4 times higher concentration was required to kill the bacteria within biofilms. Similarly, a chlorine concentration of 0.4 mg /L for 15 minutes was sufficient to kill planktonic Legionella but more than 3 mg/L was needed for killing biofilm- associated bacteria (120). When the efficacy of chlorine was tested against L. pneumophila residing within cysts of Acanthamoebae polyphaga, the bacteria could tolerate at least up to 50 mg/L chlorine (122). To keep L. pneumophila below the detectable limit, routine residual chlorine levels must be between 2 to 6mg/L (123),

(124). Continuous use of biocides poses several issues. First, prolonged use of biocides leads to corrosion in pipes and cooling towers. Also, when water treated with the biocides is purged from the systems and released it can pose risks such as soil and water pollution.

Finally, use of chlorine gas and chlorine dioxide involves an occupational risk of exposure of biocides to the people administering them.

Therefore, most industries do not continually treat with water systems with biocides. Thus bacteria that survive develop tolerance for the biocide and it results in evolution of bacteria with a higher tolerance for the biocide. Therefore, biofilms regrow upon removal of the agent. Because the EPS of the biofilm prevents thorough penetration of the biocide, some bacteria always survive (125). When a hot water system in a hospital was treated to remove Legionella colonization using hyper chlorination, the bacteria recolonized the system within two months of treatment (126). In a hospital, chlorine dioxide eradication of L. pneumophila biofilms was unsuccessful in complete removal of the biofilm and cases of LD were reported until a sterile water policy was adopted (127).

Current treatments with biocides are only useful when used on a routine basis and

therefore novel ways for removal of L. pneumophila biofilms that will also prevent or delay the regrowth of the bacteria are essential. Novel treatment options will lead to a reduction in the concentration of biocides required for removal of L. pneumophila biofilms, reduce the cost of treatment and increase overall efficiency.

Nanotechnology is an exciting and rapidly developing field. The global market for nanotechnology is rapidly increasing. Nanotechnology is widely applied in everyday consumer products such as creams, lotions, washing machines, cycles, and cars. New research areas using nanotechnology are under development for biological applications from disease diagnosis to novel therapies. Nano-materials exist in a variety of forms and shapes such as nanowires, nano-rods, nano-crystals, dendrimers, fullerenes, quantum dots, and nanoparticles (NPs) with a variety of core composition and properties like super paramagnetic NPs.

Silver NPs have been widely used for a variety of biological applications to include those involving bacteria due to its antimicrobial activity. Silver NPs exhibit dose- dependent bactericidal effects on multidrug resistant Escherichia coli O157, P. aeruginosa and Staphylococcus aureus (128). The study further suggests that the mechanism of action of silver NPs is similar to that of silver ions and the particles exert the bactericidal effect due to inhibition of cell wall synthesis (128). Exposure of copper oxide (CuO) NPs to L. pneumophila caused concentration and time dependent effects on growth of the bacteria leading to cell death. Transcriptome analysis revealed up- regulation of virulence and DNA synthesis genes (129). When E. coli and B. subtilis were

exposed to cerium oxide (CeO2) nanoparticles, a size and concentration dependent inhibition effect was observed (130). Increasing concentration of titanium dioxide (TiO2), silicon dioxide(SiO2) and zinc oxide (ZnO)NPs to E. coli and Bacillus subtilis led to increased antibacterial activity and SiO2NPs were found to be most toxic (131). Chitosan

NPs and copper adsorbed chitosan NPs were antibacterial against E. coli, S. aureus,

Salmonella typhimurium and Salmonella choleraesuis at a minimum inhibitory concentration of 0.25 µg/ml and minimum bactericidal concentration of 1 µg/ml (132).

NPs could potentially be used as to deliver antimicrobial silver to bacteria within biofilms. Cell viability of biofilms from wastewater biofilms was not affected even after application of 200 mg/L silver NPs but the viability was reduced when the cells loosely associated with EPS were removed. However, when bacteria were isolated from the wastewater biofilms and treated as a planktonic culture the bacteria were killed by application of 1 mg/L silver NPs in one hour. Silver ions at 200 mg/L were more effective in reducing cell viability than the silver NPs suggesting that silver ions may play a significant role in the bactericidal action and the biofilm EPS may be preventing the interaction of silver ions from the NPs (133).

Nanoparticles are not a new phenomenon and exist in the natural environment.

They are present in soil, water, volcanic ash, forest-fire smoke, clouds and clay. Mineral weathering and nano-biomineralization can generate naturally occurring NPs. Glover and colleagues explain that in natural environments, metal objects can oxidize and release cations (134). Silver and copper cations can then be spontaneously reduced by photo catalytic action to form metallic NPs and can interact with biofilms in the aquatic

environment (134). NPs characterized from the fungus Arthrobotrys oligospora, (fungal kingdom) are about 360 nm in size and have an immune-stimulatory effect in mouse macrophages as well as antitumor properties (135). Naturally occurring NPs from English

Ivy are efficient in blocking ultraviolet light and show decreased cell cytotoxicity when compared to TiO2 NPs (136).Due to the small size and high surface area, there can be dramatic effects on interactions of NPs with biofilms and higher organisms feeding on biofilms. Since biofilms as well as NPs occur ubiquitously in the aquatic environment, anthropogenic activities (increase in the use of NPs) and modifications of biofilms (by use of biocides and nanomaterial) that serve as the base of the food web can have significant ecological impacts. NPs may undergo a dramatic change in presence of ultraviolet light, in presence of ligands and other organic and inorganic material.

Spontaneous generation of NPs from human-made objects and natural processes suggests that NPs have been in direct contact with aquatic ecosystems and studying the trophic interactions of NPs and organisms will elucidate previously unknown organism response mechanisms.

Our previous data demonstrated that exposure to18 nm citrate-capped gold NPs altered the morphology of L. pneumophila biofilms and led to a significant reduction in surface area coverage of the NP- treated biofilms as compared to the controls. A reduction in surface area was not observed with 50 nm NPs or a higher concentration of gold NPs implying a size and concentration dependent effect on morphology of L. pneumophila biofilms (137). Similar size dependent results were observed when the antibacterial effect of ZnO NPs was tested on E. coli and S. aureus and reduction in

particle size led to increased antibacterial activity (138). Further, when L. pneumophila biofilms with the gold NP treatment induced altered morphology were presented to the amoebae, Acanthamoebae polyphaga, there was a reduction in the capacity of amoebae to graze on L. pneumophila biofilms (139).

Building on this previous research we can now ask how the understanding of biofilm mechanisms can be used in developing new ways for effective removal of L. pneumophila biofilm. The dispersal events triggered by NP exposure suggest a destabilization event. A combination treatment using nanoparticles to first destabilize the biofilm, followed by exposure to the biocide may be more effective due to increased penetration of the biocide. This type of treatment could aid in removing the biofilm and at inactivating the remaining biofilm to delay or prevent regrowth. Additionally this combination of nanoparticles with biocides could reduce the effective concentration of the biocide required for eradication of the biofilm. Initial analysis testing the efficacy of chlorine alone on L. pneumophila biofilms revealed no significant differences in the biomass of biofilms. However, exposure to 18 nm gold NPs followed by chlorine treatment revealed significant differences in the biomass of biofilms from controls as well as chlorine alone and NP alone (figure 9, unpublished data, Tara Raftery)

These data suggest that combination treatment using NPs and bleach could be useful for biofilm eradication. We therefore further investigated the effect of a combination of 18 nm gold NPs and the more efficient biocide chlorine dioxide and hypothesized that the combination treatment would be effective in removal of L. pneumophila biofilm and delay regrowth of the bacteria.

Materials and Methods

Organisms and growth conditions: Legionella pneumophila Philadelphia 1 strain

(ATCC 33152) was used in all experiments. Bacteria were grown on buffered charcoal yeast extract (BCYE) agar for three days at 26°C and a suspension of the bacteria was made in ACES-buffered yeast extract (AYE) to set up biofilms as described in detail below.

Nanoparticles: 18nm citrate capped gold nanoparticles were used for all experiments.

Nanoparticles (NPs) were synthesized as described in Gole and Murphy (140).NPs were synthesized in deionized water and characterized in ultrapure water (18 milliohm). NPs were characterized using dynamic light scattering (DLS) and Zeta potential. DLS measures the hydrodynamic diameter of the particles in solution by evaluating the variations in the intensity of scattered light from particles as they undergo Brownian motion (Malvern Instruments, 2012). Zeta potential measures the stability and aggregation of nanoparticles. Values less than-30mV and more than +30mV are considered stable for nanoparticles. Zeta potential is determined by calculating the velocity of the particles as they move across the laser beam in a liquid suspension.

Concentration of 1µg/liter of the NPs was used for all experiments.

Biofilm establishment: Biofilms were set up as described in (137). Briefly, L. pneumophila biofilms were established in sterile chambers containing two sterile glass slides. A suspension of the bacteria was made in AYE (OD600 = 0.6) and 5ml of this suspension was added to each chamber with 22.5ml of ultrapure water and 2.5ml of AYE

for initial bacterial attachment to the slides. The chambers were incubated at 26°C for 24 hours and then media replaced with 30ml 100% AYE and further incubated for 4 days to promote biofilm growth and maturation.

Biocide assay: L. pneumophila biofilms were set up in sterile chambers containing two sterile glass slides as described above. On day 5, the media was removed and the biofilm was washed with ultrapure water to remove unattached cells. It was then replaced with 30 ml moderately hard water (MHW, (90)) to simulate a natural aquatic environment. At this point, six treatments using six different chambers were set up in MHW as shown in the table below:

Table 1: Treatments set up and time line for biocide assay

Treatment Treatment Treatment change at day 7 change at day 5

CONTROL MHW MHW

BIOCIDE (BC) MHW Bleach (3 hours, 0.5mg/L) or chlorine dioxide (2 hours, 0.1 ppm)

NANOPARTICLE (NP) 1µg/L MHW AuNP

COMBINATION 1µg/L Bleach (3 hours, 0.5mg/L) or chlorine dioxide (2 hours, 0.1 (NPBC) AuNP ppm)

ERYTHROMYCIN MHW EM (2 hours, 0.25µg/ml ) (EM)

CIPROFLOXACIN MHW CF (2 hours, 0.004µg/ml (CF)

All treatments except the NP and NPBC received 30 ml of MHW until day 7, while NP and NPBC treatments were exposed to 1µg/L of Au NPs for 48h. On day 7, all biofilms were washed with ultrapure water. BC and NPBC were then exposed to chlorine-dioxide for two hours or chlorine for three hours and the antibiotics were added to the respective chambers at the concentrations mentioned above for two hours in MHW. Control and NP treatments received MHW for the two hour period. An illustration of all treatments is shown below. All biofilms were then washed with ultrapure water and slides were removed and air dried completely.

Microscopy analysis: All slides were fixed in 100% methanol for 15 minutes and air dried again. They were then stained with Syto 11 nucleic acid dye (3µm in ultrapure water, Invitrogen) for 30 minutes. Slides were washed with ultrapure water to remove excess stain and air dried completely. Slides were mounted using a mixture of PBS- glycerol and covered with coverslip. Confocal microscopy (Nikon, TiE Eclipse) was performed on each slide and the images analyzed using COMSTAT (92) for biomass, surface area and roughness of the biofilms. The experiment was performed independently at least three times and data were analyzed using ANOVA.

Regrowth assay (Microscopy): Biofilms for regrowth assay were set up as above and divided into four treatments: Control, biocide (bleach or chlorine dioxide), nanoparticle, and combination treatment. Biofilms were exposed as before on day 5 with nanoparticles and on day 7 to the biocide. After treatment, one slide from each treatment was removed and stained using the Syto 11 dye for confocal microscopy. The remaining slide in each treatment chamber was washed and provided fresh media and incubated at 26° C for five additional days. On day 12, the media was removed and slides were washed and stained with the Syto 11 dye and confocal microscopy was performed. The images from all slides were analyzed using COMSTAT. The regrowth experiment was independently performed at least three times and data were analyzed using ANOVA.

Statistical analyses: All experiments were performed independently and the data were analyzed using ANOVA with 95% confidence limits.

Results

Effect of chlorine dioxide and antibiotics against Legionella pneumophila biofilms

The efficacy of chlorine dioxide biocide for removal of L. pneumophila biofilms was analyzed. Our results show that chlorine dioxide is effective against L. pneumophila biofilms as seen by a reduction in the biomass compared to the controls (Figure 11).

Antibiotic treatments with erythromycin (EM) and ciprofloxacin (CF) used as positive controls also demonstrate a loss of biomass after exposure (Figure 11).

Combination treatment (NPBC) does not significantly reduce the biomass of the L. pneumophila biofilm from either treatment alone

Previous data from our lab shows a reduction in the biofilm coverage upon exposure to combination 18nm gold nanoparticles followed by bleach (unpublished data, Tara

Raftery). We wanted to determine if combination of nanoparticles and chlorine dioxide could further reduce the biomass of Legionella biofilms. Legionella biofilms were exposed to treatments as described in the methods. Microscopy and COMSTAT analysis demonstrate that BC alone, NP alone and the combination treatment all significantly reduce biofilm biomass in comparison to the control. However, the combination treatment of NPBC was not significantly different from either the NP or BC treatment alone (Figure 12).

Combination treatment affects the regrowth of the biofilm

The bigger problem with eradication efforts of Legionella biofilms using biocides is that the biofilms regrow after removal of the biocides. Most systems test for the presence of

Legionella through water column analysis and not by examining the towers for the presence of biofilms. Therefore, when water column levels drop below detectable levels, treatment is removed. We investigated if the combination treatment (NPBC) was more efficient at preventing regrowth of the biofilms, potentially increasing the time between needed treatments. When bleach was used as the biocide, regrowth of combination did not differ significantly from controls or individual treatments (Figure 13).

Next chlorine dioxide was used as the biocide. All treatments in this experiment lead to less regrowth than the controls. Control biofilms grew up to 1.4 fold in 5 days while chlorine dioxide alone grew 0.06 times after removal of treatment. NP and NPBC treatments continued to reduce biomass of the biofilms post treatment. Due to variation of biomass in COMSTAT analysis, significant differences between BC, NP and NPBC were not obtained (p>0.05). Although a statistical difference between the treatments was not found, a trend suggesting that the combination treatment was more efficient was seen.

(Figure14). Thus, although the NPBC (using chlorine dioxide) treatment is unable to remove the biofilm more efficiently than that of NP or BC treatment alone (Figure 14) it is possible that the combination will prevent regrowth better than either treatment alone.

Discussion

Biofilms respond not only to biotic but also abiotic factors of the environment. In this study, we investigated the interaction of Legionella pneumophila with metallic nanoparticles (NPs). Previous work suggested that exposure to NPs induces dispersal of

L. pneumophila biofilms (137). Here, the specific aim is to use our basic understanding of the interaction of NPs and biofilms in the application of NPs in an effort to eradicate biofilms. Biofilms are an environmental niche for Legionella and readily persists within biofilms in natural and human-made systems. Of particular interest, is the ability of the bacteria to form biofilms in man- made systems such as cooling towers, hot water systems and industrial plumbing systems. Growth of bacterial biofilms in these human- made systems poses a tremendous health risk for humans as well as a high cost of maintenance for the systems. Cooling towers allow for amplification of Legionella along with its dissemination in the environment by generation of aerosols causing many outbreaks of LD (120). Many treatment approaches to remove or minimize the growth of biofilms are employed by industries and hospitals but there are numerous problems associated with these treatments. Heat and biocides (oxidizing and non-oxidizing) are frequently used as the primary treatment options but are not completely efficient as the biofilms regrow quickly after removal of the agent. Additionally, concentration of biocides required for removal of biofilms is often higher than the safety levels recommended by the EPA. When chlorination is used to treat the biofilm, it is required that the free chlorine be maintained at 0.5 mg/L to 1.0 mg/L but the minimum inhibitory concentration required to kill Legionella is reported to be as high as 512 mg/L (119).

Upon completion of the process, if the water from the treatment plants needs to be emptied into natural waters, dechlorination procedure needs to be performed adding to the cost of maintenance. Nanotechnologies have been employed to develop surfaces that are resistant to microbial growth. However, for existing structures it is expensive to replace all surfaces with new nanomaterial. Our investigation uses the unique properties of NPs in an applied format to remove existing biofilms.

Many studies have focused on bacterial-NP interaction. Zinc oxide (ZnO) NPs are internalized by Streptococcus agalactiae and Staphylococcus aureus and are lethal to these bacteria in a concentration dependent manner (141). Escherichia coli and S. aureus were completely inhibited upon exposure to ZnO NPs at 3.4 mM and 1 mM respectively.

Inhibition of growth was accompanied by corresponding loss of cell viability (142).

Brayner and colleagues show a similar effect with E.coli and ZnO NPs and suggest that upon contact with bacterial cells the NPs, increase in membrane permeability leading to cellular uptake of the particles and their accumulation in the bacterial membrane (143).

Silver NPs were four times more inhibitory to planktonic E.coli than to biofilms but did penetrate up to 40 µm in a thick biofilm after a one hour exposure (144).

Sloughing of Pseudomonas putida biofilms occurred on exposure to silver NPs in presence and absence of organic material (145). In marine biofilms, a concentration dependent reduction in volume to surface area after exposure to silver NPs is observed. In addition, although the relative abundance of all taxonomic groups of bacteria remained unchanged, the community structure was altered and normal biofilm development was

affected (146). Above studies highlight the differences in responses of various bacteria to nanoparticles. Biofilms respond to NPs in a variety of ways, likely based on the characteristics of the NPs and composition of the biofilm.

Previously published data in our lab indicated a size and concentration dependent effect of gold NPs on L. pneumophila biofilm. Specifically, smaller sizes of 4 nm and 18 nm and lower concentration of gold NPs were effective in destabilization of L. pneumophila biofilms (137). We hypothesized that a combination treatment using NP exposure followed by the biocide treatment would yield greater success in removing L. pneumophila biofilms.

When used alone, bleach did not reduce the biomass of L. pneumophila biofilms but prior exposure to 18 nm gold NPs followed by bleach significantly reduced the biomass of the biofilms as compared to the controls. The reduction was significantly greater than that of bleach or NP treatments alone (unpublished data, Tara Raftery). We extended this approach to test the effectiveness of chlorine dioxide in combination with

NPs. Our results demonstrated that biomass of biofilms treated with chlorine dioxide is significantly lower than the controls but that the combination treatment of NP and chlorine dioxide is not statistically different from chlorine dioxide or NP treatment alone.

Chlorine dioxide is one of the primary treatments for L. pneumophila biofilms in cooling towers because of its effectiveness Chlorine dioxide disrupts the protein synthesis in bacteria leading to cell death (147). In a hospital hot water system, 3-5 ppm of chlorine dioxide was maintained in the water after the shock treatment of the system with 50-80

ppm chlorine dioxide and ensured removal of Legionella below detectable limits (148).

The concentration used in our experiments (0.1 ppm) is significantly lower than the typical concentration used in cooling towers (ranges from 1-10 ppm) and is useful for treating single species L. pneumophila biofilms in the lab. However, the complex nature of environmental biofilms along with presence of protozoa in cooling towers requires the use of a higher concentration of chlorine dioxide.

The bigger issue with Legionella biofilms in cooling towers is regrowth.

Treatment of cooling towers in London using chlorinated phenol at 100 ppm inhibited bacterial growth for at least three days, but Legionella were detectable after 14 days

(149). We therefore examined if a combination treatment of NPs-chlorine dioxide was more effective at preventing regrowth. We analyzed biofilm regrowth using both NP- bleach and NP-chlorine dioxide five days after treatment removal. Significant differences between control biofilm regrowth and individual treatments of NP, BC and combination treatment were observed. However, no significant difference between the combination treatment, biocide alone or NP treatment alone was observed. A trend suggesting that the bacteria grow more slowly in the combination treatment than after chlorine dioxide alone was seen. Extension of the post-treatment analysis may provide a more accurate representation of the regrowth within the biofilm.

Another interesting possibility from this investigation revolves around the need of

Legionella in the environment to use protozoan cells as a means of survival and replication. The protozoan cells in addition to serving as a host cell also protect

Legionella from environmental stresses. Protozoa are the primary grazers of the biofilms in aquatic environments. Thomas et al showed that presence of amoebae in cooling towers confers protection to the bacteria from the disinfectants and helps Legionella recover their initial levels more quickly after biocide treatment is stopped (150). Data from our lab also demonstrated reduced grazing by Acanthamoebae polyphaga after exposure to biofilms treated with 18 nm gold NPs (139). This suggests that the change in the morphology of the biofilms induced due to the NP exposure reduces the capacity of amoebae to graze on L. pneumophila biofilms. It can therefore be hypothesized that a change in the biofilm morphology can significantly alter the interaction between trophic levels (139). Therefore, alteration of the biofilms by NP exposure can possibly reduce regrowth potential not only by improving the effectiveness of the biocide, but also by reducing interaction with host cells and thus reducing bacterial replication. It is also likely that the protozoa themselves will interact with nanoparticles.

Several studies have investigated the fate of nanoparticles within protozoa.

Although Raftery et al (139)saw no effect on amoebae viability or replication after exposure to gold NPs, other studies have shown negative impacts on protozoa by NPs.

Effect of silver NPs in presence and absence of light was tested on the ciliate

Tetrahymena pyriformis. The authors found that toxicity of NPs is higher than silver ions in absence of light but the activity of silver NPs reduced greatly in presence of light

(151). Cadmium selenide quantum dots that were accumulated in P. aeruginosa were not only transferred to, but also biomagnified in Tetrahymena thermophila. Cadmium concentrations in the protozoa were up to five times higher than in the bacteria and the

bacteria prevented the protozoan lysis (152). These results are significant because they imply that trophic transfer and bio-magnification of nanoparticles is possible and intact quantum dots remain available for higher trophic levels (152).

Cooling tower environment hosts Legionella biofilms and amoebae, that can be used as host cells by the bacteria to escape disinfection strategies. Future studies can examine response of amoebae to the biofilms exposed to biocides and a combination of nanoparticles and biocides. Novel treatment options using NPs to destabilize the biofilm matrix first, followed by treatment with biocides will reduce the requirement for high concentration of biocides. It will increase time between the treatments and this will further reduce corrosion issues in the towers and lead to a reduction in the economic costs associated with the treatments.

CONCLUSION

Bacteria are among the most successful forms of life. One of the most important survival mechanisms for bacteria is their ability to form biofilms. Biofilms are present in a variety of environments: from, growing environmentally on rocks and in aquatic systems and in man-made environments such as industries, plumbing systems, and medical implants and in the human body. The ability of biofilms to resist disinfection and other removal strategies can significantly affect the ecology of the environment in which they are present. Biofilms are a food source for many invertebrates and play important roles in nutrient cycling into higher trophic levels. But, biofilms also play a role in numerous diseases such as tuberculosis, Legionnaires’ Disease, cholera, plague, and orthodontic diseases. Biofilms can also cause fouling issues in water supplies. Good and bad, biofilms play an important role in shaping ecosystems.

As more and more has been learned from studies on biofilms, this information has been applied to answer some of the bigger questions in ecology and disease. Questions regarding transmission dynamics, evolution of virulence traits, trophic level interactions, population ecology and anthropogenic effects on ecosystems have all been addressed through the study of biofilms. Cholera epidemics occur during the monsoon season or after a natural disaster such as floods (152). In countries such as Bangladesh and India sanitation and clean sources of water are frequently unavailable. Cholera epidemics usually occur in the monsoon season or after a natural disaster such as a flood (153) and these outbreaks can cause millions of cases. The disease occurs upon ingestion of high

numbers of Vibrio cholerae via contaminated water or food. The bacteria produce cholera toxin in the small intestine causing diarrhea. Disease transmission is facilitated due to shedding of large numbers of bacteria via patients. However, for years the connection to the environmental reservoir was missing. Basic research on Vibrio biofilm persistence led to the discovery of biofilms of the bacteria on copepods in the environment.

Phytoplankton and zooplankton blooms occur every year in Bangladesh during summer.

Each year, the blooms are followed by monsoons and outbreaks of cholera (154), (60).

Huq et al (60)identified an association between V. cholerae biofilms and copepods.

Biofilm formation on the copepods improves the survival of V. cholerae during inter- epidemic periods. By realizing that a single copepod can carry up to 104 cells of V. cholerae(60)and that the zooplankton bloom can result in increased numbers of this organism (carrying the bacteria) in the water, researchers were able to modify peoples’ behaviors during this time to reduce the incidence of disease. Adopting simple measures like using a sari cloth to filter water reduced the number of cholera cases (63).

Additionally, the recognition of the correlation with the blooms and monsoons led to development of predicative tools. El Nino Southern Oscillation (ENSO) is a band of warm ocean waters that can cause variations in surface temperatures in tropical Eastern

Pacific. ENSO was hypothesized to play a role in cholera emergence in Peru in 1991-

1992 after a century of absence of disease when ENSO coincided with cholera epidemic

(154), (61). To test if ENSO played a role in cholera disease dynamics, the above hypothesis was tested in cholera endemic regions of Bangladesh. There were increased number of reported cases and transmission rates upon warming of the Pacific regions due

to ENSO (155). Other studies have tested ocean temperatures and climate variation patterns as useful predictors of cholera outbreaks in Bangladesh (156) (157). Thus, understanding of V. cholerae biofilms has led to our understanding of cholera disease dynamics.

Currently we do not understand where Francisella bacteria exist in the environment and how they survive outside of mammalian reservoirs. Our work begins to elucidate this issue by answering some basic questions such as, how do the bacteria persist outside the host in the environment, can benthic grazers feed on the bacteria, how does this interaction affect both, the pathogen and the grazer? Our data demonstrates that

F. tularensis LVS can successfully form biofilms in media and persist in moderately hard water in absence of a host cell. Additionally, mosquito larvae can graze on F. tularensis

LVS biofilms and the bacteria persist in the larvae for at least 72 hours. The larvae are transient hosts for F. tularensis but they can serve to disseminate the bacteria in the environment. Biomagnification of the bacteria can occur when higher invertebrates such as fish feed on mosquito larvae with F. tularensis. Thus, studying the interaction of

Francisella tularensis and mosquito larvae can lead to similar understanding of persistence and transmission of the bacteria in the environment as observed in Vibrio.

Another question regarding environmental pathogens is how selection pressures play a role in maintenance of virulence traits. Grazing arthropods have significant impacts on environmental pathogens with a biofilm niche and affect the population dynamics of bacterial biofilms. The ability of a pathogen to maintain and adapt to novel

transmission mechanisms depends on selection pressures exerted in the environmental hosts. Each time a pathogen encounters a new environment, antimicrobial, or a new host; the pathogen must adapt or will face extinction. In aquatic ecosystems, selection is exerted by the grazing arthropods on the bacteria. Since humans are a dead end host for both F. tularensis and Legionella pneumophila selection of virulence traits depends on pressures exerted by protozoan host cells. Protozoa and amoebae frequently serve as the reservoirs for survival and replication of pathogens in the environment. For a long time, we could not understand why a commensal bacterium of cattle, Escherichia coli O157:H7 coded for virulence factors that killed its human hosts. A possible explanation is that these virulence determinants are required for the survival of E.coliO157:H7 in other environments and virulence against humans is accidental. This was found to be true; survival of the bacteria was enhanced in protozoa Tetrahymena pyriformis (158), 2007).

Human macrophages perform similar functions compared with amoebae in that they engulf and sequester foreign particles. Genetic analysis of F. tularensis has shown that the virulence factors expressed in macrophages and amoebae are the same (159). (159).

Therefore, understanding how biofilms develop and interact with higher trophic levels such as those examined in this dissertation will allow us to identify selection pressures that play a role in pathogen virulence.

Studying trophic interactions is vital to understand how communities respond to environmental change. However, these studies when performed in laboratory settings pose a challenge. Often times, mimicking the natural route of interaction of organisms is difficult. For instance, several F. tularensis interactions with insects are studied using cell

lines or intra-hemocoelic injections of live bacteria as well as use of model organisms such as Drosophila melanogaster (84),(82),(83). Our model for acquisition of the bacteria from the planktonic and the biofilm form mimics natural grazing on bacteria that occurs in the environment. This model system can be extended to other mosquito species and bacterial pathogens. Our model allows us to examine the effect of pathogen on the mosquito larvae as well as elucidate mechanisms of persistence of F. tularensis.

Environmental bacteria can alter mosquito life history traits and stresses incurred as juveniles can have a negative impact on the mosquito life history. Collaborative work from our lab shows that exposure to F. tularensis LVS as juveniles affects the fecundity of mating pair includes mosquitoes exposed to F. tularensis LVS biofilms. Thus, exposure to bacteria and stresses encountered in the larval stages can have significant impacts on mosquito life history traits. Mosquitoes are predominant vectors of F. tularensis in Europe, and change in ecology of the mosquito vector can affect transmission of tularemia (74). Our study sheds light on the environmental persistence of

Francisella and suggests that mosquito larvae can be potential reservoirs for maintenance of the bacteria in the aquatic environment.

Along with biotic interactions, biofilms also respond to abiotic metal contaminants, disinfectants and nanoparticles (NPs) that form a part of the aquatic ecosystems due to natural and anthropogenic activities. One fundamental question regarding use of nanotechnology is how do NPs integrate with the base of the food web- biofilms and how do the biofilms then respond to them? Can trophic transfer of NPs occur and how does it affect the next level of interaction? Although nanotechnology

offers potentially new and exciting ways for treating disease and delivery of drugs, understanding of the impact of use and release of NPs in the environment is essential.

The study of aquatic biofilms and NPs along with the impacts on higher trophic levels is important as it will shed light on the fate of nanoparticles in the aquatic environment.

Multiple studies have characterized the fate of NPs in aquatic ecosystems and most information about effect of nanoparticle exposure to organism comes from toxicity studies. A multi-trophic level study evaluating effects of various metallic NPs on algae, daphnids and zebra fish demonstrated that out of silver, copper, titanium dioxide, cobalt, aluminum and nickel NPs, nano-silver and nano-copper cause toxicity in daphnids, algae and zebra fish. Larger organisms like zebra fish were less susceptible to nano-metal induced toxicity than filter-feeding invertebrates such as daphnids (160). A study by

Heinlaan et al showed that zinc oxide and copper oxide NPs were extremely toxic to the bacterium Vibrio fischeri and crustaceans Daphnia magnia and Thamnocephalus platyrus(161). Titanium dioxide NPs caused high toxicity in D. magnia when exposure time was increased from 48 hours to 72 hours (162). Chronic exposure to the titanium dioxide NPs for 21 days resulted in growth retardation, mortality and reproductive defects. The NPs accumulated within the daphnids and ultimately led to interference in food intake and reproductive collapse (162).

Previous work from our lab shows an effect on morphology of L. pneumophila biofilms upon exposure to gold NPs (137). Pseudomonas aeruginosa and Staphylococcus epidermidis biofilms were inhibited upon exposure to 100nM silver nanoparticles after

treatment for 2 hours (163). Data from our lab also demonstrates an effect of grazing of amoebae upon exposure to biofilms treated with NPs (139).Due to their small size and high surface area available for interactions, careful evaluation of the effects of the NPs on the environment as well as potential for bioaccumulation is required. It is certain that bacterial biofilms in the environment will interact with NPs; therefore effects of this interaction must be evaluated in order to understand their environmental impact as biofilms serve as the base of the food chain. Grazing resistance of Serratia marcescens biofilms by two protozoans with different feeding styles was noted in a study by Queck et al (164). The biofilms in batch conditions formed micro-colonies and were protected by the suspension feeder, flagellate Bodo saltans, but not from the surface feeder A. polyphaga. Contrastingly, the filamentous biofilms under flow conditions were protected from A. polyphaga (164). This suggests that changes in morphology of biofilms affect next level trophic interactions. Therefore, the published results from our lab showing that a morphology change induced by NPs can alter the grazing ability of amoebae (137) are significant in answering the fundamental questions about use and effects of NPs.

Biofilms serve as the base of the food chain in the aquatic ecosystem and any effect on the biofilms due to the biocides or NPs can affect the grazer population and be magnified up the food chain. If the biomass of the biofilms is reduced due to the biocide, nanoparticles or the combination treatment, and the biofilms are more easily dispersed; less biomass is available for the primary aquatic grazers, which may lead to significant change in the ecology of the grazing population. Although my dissertation examines an applied area of nanotechnology, it builds from the initial basic research about biofilm-NP

interactions conducted in our lab, reemphasizing importance of studying interactions of bacterial biofilms with environmental components.

In summary, basic biofilm research studying biofilm interactions with biotic and abiotic components will enhance our understanding about pathogen persistence, mechanisms of transmission, disease dynamics and effect of anthropogenic activities on the ecology of ecosystems. In the evolutionary arms race, pathogens are constantly adapting and evolving; thus understanding details about their environmental existence will open new avenues for prevention and treatment of disease.

REFERENCES

1. Parsek, MR and Singh, PK.Bacterial biofilms: an emerging link to disease pathogenesis. s.l. : Annual Reviews in Microbiology., 2003. 57:677-701.

2. Hall-Stoodley, L, Costerton, JW and Stoodley, P.Bacterial biofilms: from the natural environment to infectious diseases. s.l. : Nature Reviews Microbiology., 2004. 2:95-108.

3. Karatan, E and Watnick, P.Signals, regulatory networks, and materials that build and break bacterial biofilms. s.l. : Microbiology and Molecular Biology Reviews., 2009. 73:310-347.

4. Klausen, M, et al., et al.Biofilm formation by Pseudomonas aeruginosa wild type, flagella and type IV pili mutants. s.l. : Mol. Microbiol. 48:1511-1524, 2003. 48:1511- 1524.

5. Pratt, LA and Kolter, R.Genetic analysis of Escherichia coli biofilm formation: roles of flagella, motility, chemotaxis and type I pili. s.l. : Mol. Microbiol., 1998. 30:285-293.

6. Watnick, PI, et al., et al.The absence of a flagellum leads to altered colony morphology, biofilm development and virulence in Vibrio cholerae O139. s.l. : Mol. Microbiol., 2001. 39:223-235.

7. Shrout, JD, et al., et al.The impact of quorum sensing and swarming motility on Pseudomonas aeruginosa biofilm formation is nutritionally conditional. s.l. : Mol. Microbiol., 2006. 62:1264-1277.

8. De Kievit, TR, et al., et al.Quorum-sensing genes in Pseudomonas aeruginosa biofilms: their role and expression patterns. s.l. : Appl. Environ. Microbiol., 2001. 67:1865-1873.

9. Hammer, BK and Bassler, BL.Quorum sensing controls biofilm formation in Vibrio cholerae. s.l. : Mol. Microbiol., 2003. 50:101-104. doi: 10.1046/j.1365- 2958.2003.03688.x..

10. Zhu, J and Mekalanos, JJ.Quorum Sensing-Dependent Biofilms Enhance Colonization in Vibrio cholerae. s.l. : Developmental Cell., 2003. 5:647-656. doi: 10.1016/S1534-5807(03)00295-8.

11. Labbate, M, et al., et al.Quorum sensing-controlled biofilm development in Serratia liquefaciens MG1. s.l. : J. Bacteriol., 2004. 186:692-698. 76

12. Davey, ME, Caiazza, NC and O'Toole, GA.Rhamnolipid surfactant production affects biofilm architecture in Pseudomonas aeruginosa PAO1. s.l. : J. Bacteriol., 2003. 185:1027-1036.

13. Piao, Z, et al., et al.Temperature-regulated formation of mycelial mat-like biofilms by Legionella pneumophila. s.l. : Appl. Environ. Microbiol., 2006. 72:1613-1622.

14. James, G, et al., et al.Digital image analysis of growth and starvation responses of a surface-colonizing Acinetobacter sp. s.l. : J. Bacteriol., 1995. 177:907-915.

15. Gjermansen, M, et al., et al.Characterization of starvation‐induced dispersion in Pseudomonas putida biofilms. s.l. : Environ. Microbiol., 2005. 7:894-904.

16. Sauer, K, et al., et al.Characterization of nutrient-induced dispersion in Pseudomonas aeruginosa PAO1 biofilm. s.l. : J. Bacteriol., 2004. 186:7312-7326..

17. Kaplan, JB, et al., et al.Detachment of Actinobacillus actinomycetemcomitans biofilm cells by an endogenous β-hexosaminidase activity. s.l. : J. Bacteriol., 2003. 185:4693-4698.

18. N, Reindert, Hall, MJ and Burgess, JG.Dispersal of biofilms by secreted, matrix degrading, bacterial DNase. s.l. : PLoS One, 2010. e15668..

19. Wang, X, Preston, JF and Romeo, T.The pgaABCD locus of Escherichia coli promotes the synthesis of a polysaccharide adhesin required for biofilm formation. s.l. : J. Bacteriol., 2004. 186:2724-2734.

20. Cramton, SE, et al., et al.The intercellular adhesion (ica) locus is present in Staphylococcus aureus and is required for biofilm formation. s.l. : Infect. Immun. , 1999. 67:5427-5433.

21. Heilmann, C, et al., et al.Molecular basis of intercellular adhesion in the biofilm‐forming Staphylococcus epidermidis. s.l. : Mol. Microbiol., 2006. 20:1083-1091.

22. Govan, JR and Deretic.V.Microbial pathogenesis in cystic fibrosis: mucoid Pseudomonas aeruginosa and Burkholderia cepacia. s.l. : Microbiological reviews, 1996. 539-574..

23. Evans, LR and Linker, A.Production and characterization of the slime polysaccharide of Pseudomonas aeruginosa. s.l. : J. Bacteriol., 1973. 116:915-924.

24. Hentzer, Morten, et al., et al.Alginate overproduction affects Pseudomonas aeruginosa biofilm structure and function. s.l. : Journal of Bacteriology, 2001. 5395- 5401.

25. Yildiz, FH and Schoolnik, GK.Vibrio cholerae O1 El Tor: identification of a gene cluster required for the rugose colony type, exopolysaccharide production, chlorine resistance, and biofilm formation. s.l. : Proceedings of the National Academy of Sciences., 1999. 96:4028-4033.

26. Moorthy, S and Watnick, PI.Identification of novel stage‐specific genetic requirements through whole genome transcription profiling of Vibrio cholerae biofilm development. s.l. : Mol. Microbiol., 2005. 57:1623-1635.

27. Kierek, K and Watnick, PI.Environmental determinants of Vibrio cholerae biofilm development. s.l. : Appl. Environ. Microbiol., 2003. 69:5079-5088.

28. Zogaj, X, et al., et al.The multicellular morphotypes of Salmonella typhimurium and Escherichia coli produce cellulose as the second component of the extracellular matrix. s.l. : Mol. Microbiol., 2008. 39:1452-1463.

29. Solano, C, et al., et al.Genetic analysis of Salmonella enteritidis biofilm formation: critical role of cellulose. s.l. : Mol. Microbiol., 2002. 43:793-808.

30. Matthysse, AG, et al., et al.The effect of cellulose overproduction on binding and biofilm formation on roots by Agrobacterium tumefaciens. s.l. : Mol. Plant-Microbe Interact., 2005. 18:1002-1010.

31. Danese, PN, Pratt, LA and Kolter, R.Exopolysaccharide production is required for development of Escherichia coli K-12 biofilm architecture. s.l. : J. Bacteriol., 2000. 182:3593-3596.

32. Diggle, SP, et al., et al.The galactophilic lectin, LecA, contributes to biofilm development in Pseudomonas aeruginosa. s.l. : Environ. Microbiol., 2006. 8:1095-1104.

33. Tielker, D, et al., et al.Pseudomonas aeruginosa lectin LecB is located in the outer membrane and is involved in biofilm formation. s.l. : Microbiology, 2005. 151:1313- 1323.

34. Waite, RD, et al., et al.Clustering of Pseudomonas aeruginosa transcriptomes from planktonic cultures, developing and mature biofilms reveals distinct expression profiles. s.l. : BMC Genomics., 2006. 7:162.

35. Fong, JC, et al., et al.Identification and characterization of RbmA, a novel protein required for the development of rugose colony morphology and biofilm structure in Vibrio cholerae. s.l. : J. Bacteriol., 2006. 188:1049-1059.

36. Gerstel, U, Park, C and Römling, U.Complex regulation of csgD promoter activity by global regulatory proteins. s.l. : Mol. Microbiol., 2004. 49:639-654.

37. Prigent-Combaret, C, et al., et al.Developmental pathway for biofilm formation in curli-producing Escherichia coli strains: role of flagella, curli and colanic acid. s.l. : Environ. Microbiol., 2000. 2:450-464. doi: 10.1046/j.1462-2920.2000.00128.x..

38. Jonas, K, et al., et al.Roles of curli, cellulose and BapA in Salmonella biofilm morphology studied by atomic force microscopy. s.l. : BMC Microbiology, 2007. 7:70.

39. Friedman, L and Kolter, R.Two genetic loci produce distinct carbohydrate-rich structural components of the Pseudomonas aeruginosa biofilm matrix. s.l. : J. Bacteriol., 2004. 186:4457-4465.

40. Beloin, C, et al., et al.Global impact of mature biofilm lifestyle on Escherichia coli K‐12 gene expression. s.l. : Mol. Microbiol. , 2004. 51:659-674.

41. Domka, J, et al., et al.Temporal gene‐expression in Escherichia coli K‐12 biofilms. s.l. : Environ. Microbiol., 2006. 9:332-346.

42. Lasa, I and Penadés, JR.Bap: a family of surface proteins involved in biofilm formation. s.l. : Res. Microbiol., 2006. 157:99-107.

43. Cucarella, C, et al., et al.Bap, a Staphylococcus aureus surface protein involved in biofilm formation. s.l. : J. Bacteriol., 2001. 183:2888-2896.

44. Cucarella, C, et al., et al.Role of biofilm-associated protein bap in the pathogenesis of bovine Staphylococcus aureus. s.l. : Infect. Immun., 2004. 72:2177-2185.

45. Tormo, MÁ, et al., et al.Bap-dependent biofilm formation by pathogenic species of Staphylococcus: evidence of horizontal gene transfer? s.l. : Microbiology., 2005. 151:2465-2475.

46. Toledo-Arana, A, et al., et al.The enterococcal surface protein, Esp, is involved in Enterococcus faecalis biofilm formation. s.l. : Appl. Environ. Microbiol., 2001. 67:4538- 4545.

47. Tendolkar, PM, et al., et al.Enterococcal surface protein, Esp, enhances biofilm formation by Enterococcus faecalis. s.l. : Infect. Immun., 2004. 72:6032-6039. 79

48. Lasa, I.Towards the identification of the common features of bacterial biofilm development. International Microbiology. s.l. : International Microbiology, 2006. 9:21- 28.

49. Espinosa-Urgel, M, Salido, A and Ramos, JL.Genetic analysis of functions involved in adhesion of Pseudomonas putida to seeds. s.l. : J. Bacteriol., 2000. 182:2363- 2369.

50. Hinsa, SM, et al., et al.Transition from reversible to irreversible attachment during biofilm formation by Pseudomonas fluorescens WCS365 requires an ABC transporter and a large secreted protein. s.l. : Mol. Microbiol., 2003. 49:905-918.

51. Vilain, S, et al., et al.DNA as an adhesin: Bacillus cereus requires extracellular DNA to form biofilms. s.l. : Appl. Environ. Microbiol., 2009. 75:2861-2868.

52. Conover, Matt S., Mishra, Meenu and Deora, Rajendar.Extracellular DNA is essential for maintaining Bordetella biofilm integrity on abiotic surfaces and in the upper respiratory tract of mice. s.l. : PloS one, 2011. 6, no. 2 (2011): e16861.

53. Montanaro, Lucio, et al., et al.Extracellular DNA in biofilms. s.l. : The International journal of artificial organs, 2011. 34, no. 9 (2011): 824.

54. Whitchurch, CB, et al., et al.Extracellular DNA required for bacterial biofilm formation. s.l. : Science., 2002. 295:1487-1487.

55. Matsukawa, M and Greenberg, E.Putative exopolysaccharide synthesis genes influence Pseudomonas aeruginosa biofilm development. s.l. : J. Bacteriol., 2004. 186:4449-4456.

56. Thomas, VC, et al., et al.Regulation of autolysis-dependent extracellular DNA release by Enterococcus faecalis extracellular proteases influences biofilm development. s.l. : J. Bacteriol., 2008. 190:5690-5698.

57. Rice, KC, et al., et al.The cidA murein hydrolase regulator contributes to DNA release and biofilm development in Staphylococcus aureus. s.l. : Proceedings of the National Academy of Sciences., 2007. 104:8113-8118.

58. Harmsen, M, et al., et al.Role of extracellular DNA during biofilm formation by Listeria monocytogenes. s.l. : Appl. Environ. Microbiol., 2010. 76:2271-2279.

59. Buswell, CM, et al., et al.Extended Survival and Persistence of Campylobacter spp. in Water and Aquatic Biofilms and Their Detection by Immunofluorescent-Antibody and - rRNA Staining. s.l. : Appl. Environ. Microbiol., 1998. 64:733-741.

60. Huq, A, et al., et al.Ecological relationships between Vibrio cholerae and planktonic crustacean copepods. s.l. : Appl. Environ. Microbiol., 1983. 45:275-283.

61. Colwell, RR.Global climate and infectious disease: the cholera paradigm. s.l. : Science, 1996. 274:2025.

62. Huq, A, et al., et al.Influence of water temperature, salinity, and pH on survival and growth of toxigenic Vibrio cholerae serovar 01 associated with live copepods in laboratory microcosms. s.l. : Appl. Environ. Microbiol., 1984. 48:420-424.

63. Colwell, RR, et al., et al.Reduction of cholera in Bangladeshi villages by simple filtration. s.l. : Proceedings of the National Academy of Sciences., 2003. 100:1051-1055.

64. Hinnebusch, B. J. and Erickson, D. L.Yersinia pestis biofilm in the flea vector and its role in the transmission of plague. s.l. : Bacterial biofilms, 2008. pp. 229-248. Springer Berlin Heidelberg, 2008..

65. Feldman, Katherine A., et al., et al.An outbreak of primary pneumonic tularemia on martha's vineyard. s.l. : New England Journal of Medicine, 2001. 345 (22): .

66. Matayas, Bela T., Nieder, Henry S. and Telford III, Sam R.Pneumonic tularemia on martha's vineyard. s.l. : Annals of the New York Academy of Sciences, 2007. 1105 (1): 351-77.

67. Ohara, Yoshiro, Tadashi, Sato and Morio, Homma.Arthropod-borne tularemia in Japan: Clinical analysis of 1,374 cases observed between 1924 and 1996. s.l. : Journal of Medical Entomology, 1998. 35 (4): 471-3.

68. Pang, Z. C.The investigation of the first outbreak of tularemia in shandong peninsula. s.l. : Zhonghua Liu Xing Bing Xue Za Zhi = Zhonghua Liuxingbingxue Zazhi, 1987. 8 (5) (Oct): 261-3.

69. Eliasson, H., J., et al., et al.The 2000 tularemia outbreak: A case-control study of risk factors in disease-endemic and emergent areas, sweden. s.l. : Emerging Infectious Diseases, 2002. 8 (9) (Sep): 956-60.

70. Hopla, C. E.The ecology of tularemia. s.l. : Advances in Veterinary Science and Comparative Medicine, 1974. 18 (0): 25-53.

71. Oyston, Petra CF, Sjöstedt, Anders and Titball, Richard W.Tularaemia: Bioterrorism defence renews interest in francisella tularensis. s.l. : Nature Reviews Microbiology, 2004. 2 (12): 967-78.

72. Santic, Marina, et al., et al.Francisella tularensis travels a novel, twisted road within macrophages. s.l. : Trends in Microbiology, 2006. 14 (1): 37-44.

73. Abd, H., et al., et al.Survival and growth of Francisella tularensis in acanthamoeba castellanii. s.l. : Applied and Environmental Microbiology, 2003. 69 (1) (Jan): 600-6.

74. KEIM, PAUL, JOHANSSON, ANDERS and WAGNER, DAVID M.Molecular epidemiology, evolution, and ecology of Francisella. s.l. : Annals of the New York Academy of Sciences, 2007. 30-66, http://dx.doi.org/10.1196/annals.1409.011.

75. Durham-Colleran, M. W., Verhoeven, A. B. and van Hoek, M. L.Francisella novicida forms in vitro biofilms mediated by an orphan response regulator. s.l. : Microbial Ecology, 2010. 59 (3) (Apr): 457-65.

76. Margolis, Jeffrey J., et al., et al.Contributions of Francisella tularensis subsp. novicida chitinases and sec secretion system to biofilm formation on chitin. s.l. : Applied and Environmental Microbiology, 2010. 76 (2): 596-608.

77. Hanke, C. A., et al., et al.Ulceroglandular tularemia in a toddler in germany after a mosquito bite. s.l. : European Journal of Pediatrics, 2009. 168 (8) (Aug): 937-40.

78. Ryden, P., Sjostedt, A. and Johansson, A.Effects of climate change on tularaemia disease activity in sweden. s.l. : Global Health Action, 2009. 10.3402/gha.v2i0.2063.

79. Eliasson, H. and Back, E.Tularaemia in an emergent area in sweden: An analysis of 234 cases in five years. s.l. : Scandinavian Journal of Infectious Diseases, 2007. 39 (10): 880-9.

80. Merritt, R. W. and Dadd, R. H., Walker,E. D.Feeding behavior, natural food, and nutritional relationships of larval mosquitoes. s.l. : Annual Review of Entomology, 1992. 37 : 349-76.

81. C.B., Philip and R.R., Parker.Public health weekly reports for OCTOBER 21, 1932. s.l. : Public Health Reports, 1932. 47 (43) (Oct 21): 2077-103.

82. Read, A., et al., et al.Francisella genes required for replication in mosquito cells. s.l. : Journal of Medical Entomology, 2008. 45 (6) (Nov): 1108-16.

83. Santic, Marina, et al., et al.Intracellular fate of Francisella tularensis within arthropod‐derived cells. s.l. : Environmental microbiology, 2009. 11, no. 6 (2009): 1473- 1481.

84. Aperis, G., et al., et al.Galleria mellonella as a model host to study infection by the francisella tularensis live vaccine strain. s.l. : Microbes and Infection / Institut Pasteur, 2007. 9 (6) (May): 729-34.

85. Vonkavaara, M., et al., et al.Drosophila melanogaster as a model for elucidating the pathogenicity of francisella tularensis. s.l. : Cellular Microbiology, 2008. 10 (6) (Jun): 1327-38.

86. Bauerfeind, Stephanie S. and Fischer, Klaus.Effects of food stress and density in different life stages on reproduction in a butterfly. s.l. : Oikos, 2005. 111 (3): 514-24.

87. Mpho, M., Holloway, G. J. and Callaghan, A.Fluctuating wing asymmetry and larval density stress in culex quinquefasciatus (diptera: Culicidae). s.l. : Bulletin of Entomological Research, 2000. 90 (3) (Jun): 279-83.

88. Vodovar, Nicolas, et al., et al.Drosophila host defense after oral infection by an entomopathogenic pseudomonas species. s.l. : Proceedings of the National Academy of Sciences of the United States of America, 2005. 102 (32): 11414-9.

89. Fleury, F., et al., et al.Physiological cost induced by the maternally-transmitted endosymbiont Wolbachia in the Drosophila parasitoid Leptopilina heterotoma. s.l. : Parasitology, 2000. 121, no. 05 (2000): 493-500.

90. EPA, US.Methods for measuring the acute toxicity of effluents and receiving waters to freshwater and marine organisms. 2002.

91. Geavgaard, Jonathan C., Turnbull, Matthew W. and McNealy, Tamara L.A novel technique for feeding and confirming uptake of bacteria in larvae of the southern house mosquito, culex quinquefasciatus (diptera: Culicidae). s.l. : Florida Entomologist, 2010. 93 (4): 577-83.

92. Heydorn, A., et al., et al.Quantification of biofilm structures by the novel computer program COMSTAT. s.l. : Microbiology, 2000. 146 ( Pt 10) (Pt 10) (Oct): 2395-407.

93. Vezzulli, Luigi, et al., et al.Environmental reservoirs of Vibrio cholerae and their role in cholera. s.l. : Environmental Microbiology Reports, 2010. 2, no. 1 (2010): 27-33.

94. Matz, C, et al., et al.Microcolonies, quorum sensing and cytotoxicity determine the survival of Pseudomonas aeruginosa biofilms exposed to protozoan grazing. s.l. : Environ. Microbiol., 2004. 6:218-226. doi: 10.1111/j.1462-2920.2004.00556.x..

95. Matz, C, et al., et al.Marine Biofilm Bacteria Evade Eukaryotic Predation by Targeted Chemical Defense. s.l. : PLoS ONE, 2008. 3(7).

96. Matz, C, et al., et al.Biofilm formation and phenotypic variation enhance predation- driven persistence of Vibrio cholerae. s.l. : Proc. Natl. Acad. Sci. U. S. A., 2005. 102:16819-16824.

97. Triebenbach, Alison N., et al., et al.Detection of Francisella tularensis in Alaskan mosquitoes (Diptera: Culicidae) and assessment of a laboratory model for transmission. s.l. : Journal of medical entomology, 2010. 47, no. 4 (2010): 639.

98. Guryčová, Darina.First isolation of Francisella tularensis subsp. tularensis in Europe. s.l. : European journal of epidemiology, 1998. 14, no. 8 (1998): 797-802.

99. Lundstrom, JO, et al., et al.Transstadial Transmission of Francisella tularensis holarctica in Mosquitoes, Sweden. s.l. : Emerg. Infect. Dis., 2011. 17:794-799.

100. Marsollier, L, et al., et al.Aquatic insects as a vector for Mycobacterium ulcerans. s.l. : Appl. Environ. Microbiol., 2002. 68:4623-4628.

101. Hall, JD, et al., et al.Francisella tularensis replicates within alveolar type II epithelial cells in vitro and in vivo following inhalation. s.l. : Infect. Immun., 2007. 75:1034-1039.

102. Fortier, AH, et al., et al.Live vaccine strain of Francisella tularensis: infection and immunity in mice. s.l. : Infect. Immun., 1991. 59:2922-2928.

103. Chandler, JC, et al., et al.Differential chitinase activity and production within Francisella species, subspecies, and subpopulations. s.l. : J. Bacteriol., 2011. 193:3265- 3275.

104. Lau, SC, et al., et al.Roles of bacterial community composition in biofilms as a mediator for larval settlement of three marine invertebrates. s.l. : Aquat. Microb. Ecol., 2005. 38:41-51.

105. Bao, W, et al., et al.Larval settlement and metamorphosis of the mussel Mytilus galloprovincialis in response to biofilms. s.l. : Mar. Biol., 2007. 150:565-574.

106. Olcott, MH, et al., et al.Lethality and developmental delay in Drosophila melanogaster larvae after ingestion of selected Pseudomonas fluorescens strains. s.l. : PloS One., 2010. 5:e12504.

107. Freitak, D, et al., et al.Immune system responses and fitness costs associated with consumption of bacteria in larvae of Trichoplusia ni. s.l. : BMC Biology., 2007. 5:56.

108. Mahajan, UV, et al., et al.Larval exposure to Francisella tularensisLVS affects fitness of the mosquito Culex quinquefasciatus. s.l. : FEMS Microbiol. Ecol., 2011. 78:520-530.

109. Nnakumusana, ES.Laboratory infection of mosquito larvae by entomopathogenic fungi with particular reference to Aspergillus parasiticus and its effects on fecundity and longevity of mosquitoes exposed to sporal infections in larval stages. s.l. : Current Science, 1985. 54:1221-1228.

110. Fraser, David W., et al., et al.Legionnaires' disease. s.l. : New England Journal of Medicine, 1977. 297, no. 22 (1977): 1189-1197.

111. Broome, C.Current issues in epidemiology of LegionellosisIn Anonymous Legionella: Proceedings of the 2nd International Symposium. s.l. : American Society of Microbiology. Washington, DC., 1983.

112. Stewart, CR, Muthye, V and Cianciotto, NP.Legionella pneumophila Persists within Biofilms Formed by Klebsiella pneumoniae, Flavobacterium sp., and Pseudomonas fluorescens under Dynamic Flow Conditions. s.l. : PloS One., 2012. 7:e50560.

113. Smith-Somerville, HE, et al., et al.Survival of Legionella pneumophila in the cold- water ciliate Tetrahymena vorax. s.l. : Appl. Environ. Microbiol., 1991. 57:2742-2749..

114. Nahapetian, K, et al., et al.The intracellular multiplication of< i> Legionella pneumophila in protozoa from hospital plumbing systems. s.l. : Res. Microbiol., 1991. 142:677-685.

115. Castilla, J., et al., et al.A large Legionnaires' disease outbreak in Pamplona, Spain: early detection, rapid control and no case fatality. s.l. : Infect. , 2008. 136:823-832.

116. mail, ProMed.Legionella pneumophilahumans-Canada (Quebec) 20120920.1303215. 2012.

117. Pagnier, I, Merchat, M and La Scola, B.Potentially pathogenic amoeba- associated microorganisms in cooling towers and their control. s.l. : Future Microbiology., 2009. 4:615-629.

118. Fiore, AE, et al., et al.Epidemic Legionnaires' disease two decades later: old sources, new diagnostic methods. s.l. : Clinical Infectious Diseases., 1998. 26:426-433.

119. Leoni, E, et al., et al.Legionella waterline colonization: detection of Legionella species in domestic, hotel and hospital hot water systems. s.l. : J. Appl. Microbiol., 2004. 98:373-379.

120. Kim, B, et al., et al.Literature review—efficacy of various disinfectants against Legionella in water systems. s.l. : Water Res., 2002. 36:4433-4444.

121. Garcia, M. T., et al., et al.Persistence of chlorine sensitive Legionella pneumophila in hyperchlorinated installations. s.l. : Journal of applied microbiology, 2008. 105, no. 3 (2008): 837-847.

122. Kilvington, S and Price, J.Survival of Legionella pneumophila within cysts of Acanthamoeba polyphaga following chlorine exposure. s.l. : J. Appl. Microbiol., 1990. 68:519-525. doi: 10.1111/j.1365-2672.1990.tb02904.x..

123. Lin, YS, et al., et al.Disinfection of water distribution systems for Legionella. s.l. : Semin. Respir. Infect., 1998. 13:147-159.

124. Wright, J, Ruseska, I and Costerton, J.Decreased biocide susceptibility of adherent Legionella pneumophila. s.l. : J. Appl. Microbiol., 1991. 71:531-538.

125. Gong, AS, et al., et al.Extraction and analysis of extracellular polymeric substances: Comparison of methods and extracellular polymeric substance levels in Salmonella pullorum SA 1685. s.l. : Environ. Eng. Sci., 2009. 26:1523-1532.

126. Borella, P., et al., et al.Prevention and control of Legionella infection in the hospital environment. s.l. : Annali di igiene: medicina preventiva e di comunità, 2000. 12, no. 4 (2000): 287.

127. Hosein, I. K., et al., et al.Point-of-care controls for nosocomial legionellosis combined with chlorine dioxide potable water decontamination: a two-year survey at a Welsh teaching hospital. s.l. : Journal of Hospital Infection, 2005. 61, no. 2 (2005): 100- 106.

128. Lara, HH, et al., et al.Bactericidal effect of silver nanoparticles against multidrug- resistant bacteria. s.l. : World Journal of Microbiology and Biotechnology., 2010. 26:615-621.

129. Lu, J, et al., et al.Legionella pneumophila Transcriptional Response Following Exposure to CuO Nanoparticles. s.l. : Appl. Environ. Microbiol., 2013.

130. Pelletier, DA, et al., et al.Effects of engineered cerium oxide nanoparticles on bacterial growth and viability. s.l. : Appl. Environ. Microbiol., 2010. 76:7981-7989.

131. Adams, LK, Lyon, DY and Alvarez, PJ.Comparative eco-toxicity of nanoscale TiO2, SiO2, and ZnO water suspensions. s.l. : Water Res., 2006. 40:3527-3532..

132. Qi, L, et al., et al.Preparation and antibacterial activity of chitosan nanoparticles. s.l. : Carbohydr. Res., 2004. 339:2693-2700. doi: 10.1016/j.carres.2004.09.007..

133. Sheng, Z and Liu, Y.Effects of silver nanoparticles on wastewater biofilms. s.l. : Water Res., 2011. 45:6039-6050. doi: 10.1016/j.watres.2011.08.065..

134. Glover, RD, Miller, JM and Hutchison, JE.Generation of Metal Nanoparticles from Silver and Copper Objects: Nanoparticle Dynamics on Surfaces and Potential Sources of Nanoparticles in the Environment. s.l. : ACS Nano., 2011. 5:8950-8957.

135. Wang, Y, et al., et al.Naturally Occurring Nanoparticles from Arthrobotrys oligospora as a Potential Immunostimulatory and Antitumor Agent. s.l. : Advanced Functional Materials., 2012.

136. Xia, L, et al., et al.Naturally occurring nanoparticles from English ivy: an alternative to metal-based nanoparticles for UV protection. s.l. : Journal of Nanobiotechnology., 2010. 8:12..

137. Stojak, AR, et al., et al.Morphological responses of Legionella pneumophila biofilm to nanoparticle exposure. s.l. : Nanotoxicology., 2011. doi: 10.3109/17435390.2010.550696.

138. Nair, S, et al., et al.Role of size scale of ZnO nanoparticles and microparticles on toxicity toward bacteria and osteoblast cancer cells. s.l. : J. Mater. Sci. Mater., 2009. Med. 20:235-241.

139. Raftery, TD, Lindler, H and McNealy, TL.Altered Host Cell–Bacteria Interaction due to Nanoparticle Interaction with a Bacterial Biofilm. s.l. : Microb. Ecol., 2013. 65:496-503..

140. Gole, A and Murphy, CJ.Polyelectrolyte-coated gold nanorods: synthesis, characterization and immobilization. s.l. : Chemistry of Materials., 2005. 17:1325-1330.

141. Huang, Z, et al., et al.Toxicological effect of ZnO nanoparticles based on bacteria. s.l. : Langmuir, 2008. 24:4140-4144.

142. Reddy, KM, et al., et al.Selective toxicity of zinc oxide nanoparticles to prokaryotic and eukaryotic systems. s.l. : Appl. Phys. Lett., 2007. 90:213902-213902-3..

143. Brayner, R, et al., et al.Toxicological impact studies based on Escherichia coli bacteria in ultrafine ZnO nanoparticles colloidal medium. s.l. : Nano Letters, 2006. 6:866-870.

144. Choi, O, et al., et al.Interactions of nanosilver with Escherichia coli cells in planktonic and biofilm cultures. s.l. : Water Res., 2010. 44:6095-6103. doi: 10.1016.

145. Fabrega, J, Renshaw, JC and Lead, JR.Interactions of silver nanoparticles with Pseudomonas putida biofilms. s.l. : Environ. Sci. Technol., 2009. 43:9004-9009.

146. Fabrega, J, et al., et al.Impact of silver nanoparticles on natural marine biofilm bacteria. s.l. : Chemosphere, 2011. 85:961-966.

147. Benarde, M. A., Snow, W. B. and Olivieri, V. P.Chlorine dioxide disinfection temperature . s.l. : Journal of Applied Microbiology, 1967. 30, no. 1 (1967): 159-167.

148. Walker, J, et al., et al.Control ofLegionella pneumophila in a hospital water system by chlorine dioxide. s.l. : J. Ind. Microbiol. Biotechnol., 1995. 15:384-390.

149. Kurtz, J. B., et al., et al.Legionella pneumophila in cooling water systems. s.l. : J. Hyg, 1982. (1982): 369-381.

150. Thomas, V, et al., et al.Amoebae in domestic water systems: resistance to disinfection treatments and implication in Legionella persistence. s.l. : J. Appl. Microbiol., 2004. 97:950-963.

151. Shi, J, et al., et al.Effect of light on toxicity of nanosilver to Tetrahymena pyriformis. s.l. : Environmental Toxicology and Chemistry., 2012.

152. Werlin, R, et al., et al.Biomagnification of cadmium selenide quantum dots in a simple experimental microbial food chain. s.l. : Nature Nanotechnology., 2010. 6:65-71.

153. Faruque, Shah M., et al., et al.Induction of the lysogenic phage encoding cholera toxin in naturally occurring strains of toxigenic Vibrio cholerae O1 and O139. s.l. : Infection and immunity, 1998. 66, no. 8 (1998): 3752-3757.

154. Epstein, Paul R.Algal blooms in the spread and persistence of cholera. s.l. : Biosystems, 1993. 31, no. 2 (1993): 209-221.

155. Pascual, M.Predicting endemic cholera: the role of climate variability and disease dynamics. s.l. : Climate research, 2008. 36.2 (2008): 131.

156. Magny, E., et al., et al.Environmental signatures associated with cholera epidemics. s.l. : Proceedings of the National Academy of Sciences, 2008. 105, no. 46 (2008): 17676-17681.

157. Matsuda, F., S. Ishimura, Y. Wagatsuma, T. Higashi, T. Hayashi, A. S. G. Faruque, D. A. Sack, and M. Nishibuchi.Prediction of epidemic cholera due to Vibrio cholerae O1 in children younger than 10 years using climate data in Bangladesh. s.l. : Epidemiology and infection, 2008. 136, no. 01 (2008): 73-79.

158. Steinberg, Karyn Meltz, and Bruce R. Levin.Grazing protozoa and the evolution of the Escherichia coli O157: H7 Shiga toxin-encoding prophage. s.l. : Proceedings of the Royal Society, 2007. B: Biological Sciences 274, no. 1621 (2007): 1921-1929.

159. Lauriano, CM, et al., et al.MglA regulates transcription of virulence factors necessary for Francisella tularensis intraamoebae and intramacrophage survival. s.l. : Proceedings of the National Academy of Sciences., 2004. 101:4246-4249.

160. Griffitt, RJ, et al., et al.Effects of particle composition and species on toxicity of metallic nanomaterials in aquatic organisms. s.l. : Environmental Toxicology and Chemistry, 2008. 27:1972-1978. doi: 10.1897/08-002.1.

161. Heinlaan, M, et al., et al.Toxicity of nanosized and bulk ZnO, CuO and TiO2 to bacteria Vibrio fischeri and crustaceans Daphnia magna and Thamnocephalus platyurus. s.l. : Chemosphere, 2008. 71:1308-1316.

162. Zhu, X, et al., et al.Comparative toxicity of several metal oxide nanoparticle aqueous suspensions to Zebrafish (Danio rerio) early developmental stage. s.l. : Journal of Environmental Science and Health Part A., 2008. 43:278-284.

163. Kalishwaralal, K, et al., et al.Silver nanoparticles impede the biofilm formation by Pseudomonas aeruginosa and Staphylococcus epidermidis. Colloids and Surfaces. s.l. : B: Biointerfaces., 2010. 79:340-344.

164. Queck, S, et al., et al.The role of quorum sensing mediated developmental traits in the resistance of Serratia marcescens biofilms against protozoan grazing. s.l. : Environ. Microbiol., 2006. 8:1017-1025. doi: 10.1111/j.1462-2920.2006.00993.x..