Biosynthesis of Mannose-Containing Cell Wall Components Important in Mycobacterium

Home , Middlebrook 7H9 Broth, Mycobacterium microti, Pott disease

Biosynthesis of mannose-containing cell wall components important in Mycobacterium

tuberculosis virulence

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

in the Graduate School of the The Ohio State University

Tracy Lynn Keiser

Graduate Program in Microbiolgy

The Ohio State University

2014

Dissertation Committee:

Dr. Larry S. Schlesinger, MD, Advisor

Dr. Samantha King, PhD

Dr. Robert Munson, PhD

Dr. Stephanie Seveau, PhD

Tracy Lynn Keiser

2014

Abstract

Tuberculosis (TB) remains a worldwide scourge. Mycobacterium tuberculosis

(M.tb) is one of the oldest and most successful pathogens in human history. It has developed a plethora of ways to subvert the host immune response and make a home out of what should be its undoing, its host cell niche, the macrophage phagosome.

Mycobacterium species are compositionally unique organisms belonging to the phyla,

Actinomycetes, which also includes Corynebacterium, Rhodococcus, Nocardia and

Streptomyces. Their cell envelope includes structural components of both Gram negative and Gram positive bacteria in addition to molecules exclusive to mycobacterium, like mycolic acids. This unique cell envelope configuration provides M.tb with a physical barrier to environmental insults and is responsible for the variable retention of the Gram stain. M.tb coats itself with mannosylated molecules such as the abundant mannosylated lipoglycans which have mannosyl motifs that resemble those of mammalian glycoproteins. This molecular mimicry enables M.tb to take advantage of alveolar macrophages that have up-regulated surface receptors like the C-type lectin mannose receptor to gain entry into the cell and enhance its survival.

Synthesis of the mannosylated lipoglycans involves several enzymes and pathways, and thus is difficult to study as a whole. The emphasis of this thesis will be on synthesis of the mannose donor molecules for these lipoglycans. The terminal mannose caps as well as the mannan structures in the core of these molecules are synthesized through a variety of specific mannosyltransferases that use the donors GDP-mannose and polyprenyl phosphate mannose (PPM) that are products of the mannose donor biosynthesis pathway. The putative genes of this pathway in M.tb are orthologs with

100% sequence identity to those in the attenuated vaccine strain M. bovis BCG and include manA (an isomerase), manB (a phosphomannomutase), manC (a GDP-mannose pyrophosphorylase), and ppm1 (polyprenyl-phosphate mannose synthase). Additionally, there are several other neighboring genes, like whiB2 (Fe-S clustering molecule and transcriptional regulator), Rv3256c, Rv3258c (hypothetical proteins), and Rv3253c (a postulated membrane flipase) whose functions are unknown, but are potentially contributing members of the mannose donor biosynthesis pathway.

In this thesis we examined expression and some functional characterization of genes involved in the putative mannose donor biosynthetic pathway. Our results for the transcriptional profile in broth show that there are differences in expression of certain genes between M.tb and BCG under the same growth conditions despite their identical sequences. The expression profile of M.tb and BCG mannose donor biosynthesis genes in human macrophages is of particular importance not only because macrophages are the natural host cell niche for M.tb but also because the expression profiles are highly reproducible among different donors. It is of particular interest that the genes Rv3256c,

iii

Rv3258c and ppm1 were highly expressed in M.tb 2 hours post-infection of macrophages and then gradually decreased.

Due to this up regulation at early time points post infection, characterization of the genes Rv3256c, Rv3258c and ppm1was of great interest and a focus of this work.

Ppm1over-expressiom in M. smegmatis or M. tuberculosis provided no phenotype while over-expression of Rv3258c was unstable in the surrogate organism, M. smegmatis, and also showed no phenotype in M.tb. On the other hand, over-expression of Rv3256c reduced rather than increased cell wall mannosylated lipoglycans compared to the vector control. Despite the decrease in these molecules, over-expression of Rv3256c showed an increase in association with and survival in human macrophages. Thus, although the function still remains a mystery, evidence is provided for the role of Rv3256c in virulence.

TB has become romanticized in our culture claiming the lives of some of our most beloved characters. Although we understand that the disease is caused solely by this unicellular organism, the interaction between host and microbe is incredibly complex. A more complete understanding of the entry and acclimation of M.tb into host cell macrophages could possibly allow us to develop new ways to prevent infection, limit dissemination following primary infection and possibly help enhance immune protection.

Dedication

Without the love and unwavering support of my family, none of this would have been possible. First and foremost to my father for being my number 1 fan and sparking my infatuation with science. My sisters Angie and Kimberly Keiser for showing me that when you want something enough in life, you will find a way to do it. And to the rest of my huge, fun and kind family whose names would go for pages but are certainly not forgotten.

Science opened many new doors for me both professionally and socially. Their everyday support in a career fraught with disappointment was made bearable. Too many to name, but a few worth a special mention start with Anice, Ben and Oliver Daigle for always being my Columbus home. To Mels Lodder and Neinke Willigenburg for their genuine kindness. Last but not least, my best friends Ran Furman and Geoffrey

Gonzalez. The three of us started as classmates, quickly became friends and are now considered my family. Cheers!

Acknowledgements

I would like to thank my advisor Larry Schlesinger for providing me with the opportunity pursue my degree in his lab, for his patience when I wasn’t and for the smile on his face…….. most of time anyway! Thanks to the entire Schlesinger lab with special mention to Abul Azad for the introduction to genetics. To Evelina Guirado for her support and input. To Murugesan Rajaram for always giving me a fresh perspective and to all the ladies I spent the majority of the time with, huddled in a lab bay: Tracy Carlson,

Michelle Brooks, Bin Ni and Eusondia Arnett. I would also like to acknowledge our collaborator, Jordi Torrelles and Jesus Arcos for guiding me through the biochemistry.

Vita

May, 1994……………………………...Versailles High School

2005……………………………………B.S. Microbiology, The Ohio State University

2007 to present………………………....Graduate Teaching Associate, Department of

Microbiology, The Ohio State University

Publications

1. Yang L, Sinha T, Carlson TC, Keiser TL, Torrelles JB. and Schlesinger LS. Changes in the major cell envelope components of Mycobacterium tuberculosis during in vitro growth. Glycobiology, April 2013.

2. Cummings HE, Barbi J, Reville P., Oghumu S., Zorko N, Sarkar A, Keiser TL, Lu B, Rückle T, Varikuti S., Lezama-Davila C., Wewers MD, Whitacre C, Radzioch D., Rommel C, Seveau S, and Satoskar AR. Critical role for phosphoinositide 3-kinase gamma in parasite invasion and disease progression of cutaneous leishmaniasis. PNAS, Jauary 2012.

3. Keiser TL, Azad AK, Guirado E and Schlesinger LS. A comparative transcriptional study of the putative mannose donor biosynthesis genes in the virulent Mycobacterium tuberculosis and attenuated Mycobacterium bovis BCG strains. Infection and Immunity, November, 2011.

4. Martin J, Duncan FJ, Keiser TL, Shin S, Kusewitt DF, Oberyszyn T, Satoskar AR, Vanbuskirk AM. Macrophage migration inhibitory factor (MIF) plays a critical role in pathogenesis of ultraviolet-B (UVB) -induced nonmelanoma skin cancer (NMSC). FASEB. 2008 Oct 24.

vii

5. Rosas LE, Barbi J, Snider H, , Satoskar AA, Keiser TL, Papenfuss T, Durbin J, Radzioch D, Glimcher LH, Satoskar AR. Cutting Edge: STAT1 and T-bet play opposite roles in determining outcome of visceral leishmaniasis caused by L. donovani. Journal of Immunology, 2006; Jul 1:177(1):22-5.

6. Rosas LE, Satoskar AA, Roth K, Keiser TL, Barbi J, Hunter CA, de Sauvage F, Satoskar AR. Interleukin-27R (WSX-1/T-cell receptor) gene deficient mice display enhanced resistance to Leishmania donovani infection but develop severe liver immunopathology. American Journal of Pathology. 2006; Jan:168(1):158-69.

7. Rosas LE, Keiser TL, Barbi J, Satoskar AA, Septer A, Kaczmarek J, Lezama-Davilla CM, Satoskar AR. Genetic background influences immune responses and disease outcome of cutaneous L. mexicana infection in mice. International Journal of Immunology. 2005; 17:1347.

8. Rosas LE, Keiser T, Pyles R, Durbin J, Satoskar AR. Development of protective immunity against cutaneous leishmaniasis is dependent on STAT1-mediated IFN signaling pathway. European Journal of Immunology. 2003: 33:701.

Fields of Study

Major Field: Microbiology

viii

Table of Contents

Abstract ...... ii Dedication ...... v Acknowledgements ...... vi Vita...... vii Publications ...... vii Table of Contents ...... ix List of Figures ...... xii List of Tables ...... xiv Chapter 1: Introduction ...... 1 1.1 History of Tuberculosis ...... 1 1.2 M.tb pathogenesis and the host response ...... 3 1.2.1 Drug resistance ...... 6 1.3 The mycobacterial cell wall ...... 7 1.3.1 Mycobacterium pseudo capsule ...... 8 1.3.3: The Mycolyl-Arabinogalactan-Peptidoglycan Complex ...... 9 1.3.2 Cell wall-associated proteins ...... 11 1.3.2.1 Membrane-bound proteins ...... 11 1.3.2.2 Secreted or unbound proteins ...... 12 1.3.3 Peripheral lipids and carbohydrate-linked lipids...... 13 1.3.4 Phosphatidyl-myo-inositol-based lipoglycans: Structure, biosynthesis and Effector function ...... 15 1.3.4.1 Structure and biosynthesis ...... 16 1.3.4.2 Host cell effector functions ...... 17 1.3.5 Putative Mannose donor biosynthetic pathway...... 18 1.3.5.1: GDP-mannose pyrohoshorylase, manC ...... 19 1.3.5.2: Phosphomannomutases, manB and pmmB ...... 20

1.3.5.3: Phosphomannose isomerase, manA ...... 21 1.3.5.4: Neighboring genes and their postulated functions Rv3253c and whiB2 ...... 21 1.3.5.5: Rv3258c ...... 22 1.3.5.6: PPM synthase, ppm1 ...... 22 1.3.5.7: Bioinformatics and putative function of Rv3256c ...... 25 1.3.5.8: Rv3256c as a potential glucosamine-6-phosphate synthase, glms ...... 26 1.3.5.9: Sugar donor molecules, GDP-mannose and polyprenolphosphate mannose ...... 30 1.4: Model organisms...... 31 1.5: Genetic systems in mycobacterium ...... 34 1.5.1: Plasmids used in mycobacterial research ...... 35 1.5.2: Gene deletion systems in Mycobacterium ...... 37 1.6: Animal models for TB ...... 39 Chapter 2: Transcriptional Profiling of the Putative Mannose Donor Biosynthetic Pathway in Mycobacterium ...... 42 2.1: Introduction of the Putative Mannose Donor Biosynthetic Pathway ...... 42 2.4: Materials and Methods ...... 46 2.4.1: Mycobacterial strains and growth media ...... 46 2.4.2: Isolation of human monocyte-derived macrophages (MDMs) ...... 47 2.4.3: Bacterial lysis, RNA isolation and Real-time PCR...... 47 2.5: Results ...... 48 2.5.1: Transcriptional profiling of mannose donor biosynthesis genes of M.tb and BCG grown in broth culture ...... 49 2.5.2: Transcriptional profiling of mannose donor biosynthesis genes of M.tb and BCG in macrophages ...... 53 2.6: Discussion ...... 57 Chapter 3: Elucidation of Rv3256c, a Putative Enzyme Involved in the Mannose Donor Biosynthetic Pathway in Mycobacterium ...... 60 3.1: Introduction ...... 60 3.4: Materials and Methods ...... 62 3.4.1: Mycobacterial strains, plasmids and growth media ...... 62 3.4.2: Isolation of human monocyte-derived macrophages (MDMs) ...... 63 3.4.3: Enumeration of bacterial CFUs and cell association with macrophages ...... 64

3.4.4: Western blotting and silver staining ...... 65 3.4.5: 2 Dimensional Thin Layer Chromatography ...... 65 3.4.5: Bioinformatics ...... 66 3.5: Results ...... 66 3.5.1: Analysis of mannolipids ...... 68 3.5.2: Effect of Rv3256c over-expression on bacterial survival and association with macrophages ...... 69 3.5.3: Cloning of Rv3256c and protein production...... 72 3.6: Discussion ...... 73 Chapter 4: Partial Characterization of ppm1 and Rv3258c in the Putative Mannose Donor Biosynthetic Pathway in Mycobacterium ...... 77 4.1: Introduction ...... 77 4.2: Materials and Methods ...... 78 4.2.1: Mycobacterial strains, plasmids and growth media ...... 78 4.2.2: Bioinformatics ...... 78 4.2.3: Isolation of human monocyte-derived macrophages (MDMs) ...... 79 4.2.4: Coomassie staining ...... 80 4.2.5: Bacterial lysis, RNA isolation and Real-time PCR ...... 80 4.2.6: Total carbohydrate analysis ...... 81 4.3: Results ...... 82 4.3.1: Cloning of Rv3258c ...... 82 4.3.2: Polyprenol monophosphomannose synthase, ppm1 ...... 86 4.4: Discussion ...... 90 4.4.1: Rv3258c ...... 90 4.4.2: PPM synthase ...... 91 Chapter 5: Synthesis and Future Directions ...... 93 References ...... 98

List of Figures

Figure 1.1. Cartoon depiction of the fate of inhaled M.tb in context of the human host

(114)…………………………………………………………………………...4

Figure 1.2. Mycobacterium tuberculosis cell wall depiction……………………………...7

Figure 1.3. The biosynthesis of PIMs, LM and ManLAM………………………………15

Figure 1.4. A putative mannose donor biosynthesis pathway…………………………...19

Figure 1.5. The placement (in context of the cell membrane) of the proposed catalytic

domains of M.tb ppm1 compared to Ms ppm1/2…………………………….24

Figure 1.6. Functional schematic of glucosamine-6-phosphate synthase with its substrates

and immediate products……………………………………………………...27

Figure 1.7 Alignment of protein sequences from M.tb Rv3256c and R. leguminosarum

NodM………………………………………………………………………..29

Figure 1.8. A phylogenetic tree of mycobacterium and related species…………………32

Figure 1.9. Gene deletion vector pML1611……………………………………………...38

Figure 2.1. A putative mannose donor biosynthesis pathway………..……… ………….45

Figure 2.3. Transcriptional profile of putative mannose donor pathway in broth….……52

Figure 2.4. Cartoon depiction of the early innate time points of macrophages….………54

Figure 2.5. Transcriptional profile of putative mannose donor pathway in MDMs…..…56

Figure 3.1. Analysis of mannolipids by SDS-PAGE and densitometry………..………..67

xii

Figure 3.2. 2D-TLC analysis of PIMs….………………………………………………..69

Figure 3.3. Survival assay for Rv3256c over-expression in MDMs……………………..70

Figure 3.4. Cell association assay in for over-expression of Rv3256c MDMs…………71

Figure 3.5. Western blot analysis of the Rv3256c His-tagged protein…………………..73

Figure 4.1. Visualization of cloning products of Rv3258c……………………………....83

Figure 4.2. Dose titration of anhydrotetracycline induction……………………………..84

Figure 4.3. Survival assay for over-expression of Rv3258c……………………………..86

Figure 4.4. SDS-PAGE analysis after heat induction of ppm1……………………….….87

Figure 4.5. Total sugar analysis of over-expressionof ppm1 by Mass spectrometry……89

xiii

List of Tables

Table 1. List of plasmids used including the key features……………………………….36

Table 2.1. DNA sequence identity of the mannose donor biosynthetic pathway relative to

M.tb H37RV ………………………………………………………………..…..43

Table 2.2. List of custom primers for RT-PCR………………………………………….50

Table 4.2. Total quantity of RNA extracted after induction……………………………..85

xiv

Chapter 1: Introduction

1.1 History of Tuberculosis

Mycobacterium tuberculosis (M.tb) is an ingenious, highly adaptable bacteria causing the disease Tuberculosis (TB) and has been an unremitting human companion going by the aliases; consumption, phthisis, Pott’s disease and the White Plague (25).

Detection of M.tb DNA in mummified remains from both new and old world gave us the first indication that this disease has long been part of our human history long before we diverged as a species. Recent genomic analysis has demonstrated that genomic expansion in the bacterial repertoire coincides with the expansion in human evolution as well as solidifying the idea that M.tb was not acquired as a zoonotic disease (19). These facts together provide evidence for the co-evolution and adaptation of the bacteria to become an exclusive human pathogen.

Not only is there scientific evidence pointing to this unwelcomed partnership, but much literary evidence as well. TB or Phthisis, first noted in written history by

Hippocrates around 460 BC in Book 1, of the Epidemics, as “the most considerable of the diseases which then prevailed, and the only one which proved fatal to many persons” (25,

71, 124). Hippocrates argued that this was not an affliction given as a punishment by the gods but rather it had something to do with our human biology, which he attributed to the

1 disease being due to a heritable trait. Although Hippocrates was incorrect about the heritability of TB, which Aristotle theorized was a contagion, he was not completely wrong in understanding that human characteristics also play into the disease outcome.

As human civilization became more sophisticated and hygienic towards the beginning of the 19th century, TB was still a major cause of death for seemingly healthy adults 18-35 taking an estimated 1 in 7 lives. It was so common and well-recognized that authors like Emily Brontë used it to take the lives of many of her fictional characters and ironically, it was the very thing that claimed her own life at 30. Many members of the

Brontë family including her mother, herself and all 5 siblings succumbed to TB earning it the moniker, the “family attendant”, providing further evidence to human genetic susceptibility and pathogenic adaption to the human host (25).

Recognition of TB as a contagious affliction after the Renaissance and into the industrial revolution, helped push the brightest scientists of the time to re-evaluate modern medicine. Robert Köch identified Mycobacterium tuberculosis as the causative agent of TB in his presentation of Die Aetiologie der Tuberculose in 1886 to the Berlin

Physiological Society. Here Köch demonstrated using a guinea pig that the passage of the etiological agent from a diseased animal to a healthy one resulted in the same disease state, earning him a Nobel Prize in medicine. Köch’s attempts at eliciting a self- protective response by injecting himself with small amounts of what he described as

“concentrated tuberculin” spurred the first idea of a vaccine for the prevention of TB. A more direct and effective approach was taken on by two young scientists named Albert

Calmette and Camille Guérin to attenuate a highly-related mycobacterium species that primarily infects cattle, M. bovis. The successful attenuation of this bacterium created the

2 most efficacious and still used TB vaccine, M. bovis BCG (Bacille Calmette-Guérin)

(112, 132).

The BCG vaccine is most effective in preventing pediatric TB, in particular, miliary TB and tuberculous meningitis (132). Although the prevention of disseminated TB in children was a great step scientifically, the world still needed effective treatments for the disease. In 1943, this finally seemed possible with the successful use of streptomycin against M.tb in animal infections. Streptomycin was administered to a critically ill TB patient and the effect was nearly immediate with his disease state getting visibly better and the clearance of bacteria from his sputum.

However, the celebration was short lived as patients began to show resistance to this mono therapy after only a few months. Thus began the golden age of antibiotics and multi-drug treatment regimens that if followed correctly, eliminated the bacterial infection.

1.2 M.tb pathogenesis and the host response

M.tb is a human respiratory pathogen, transmitted to a naïve individual via inhaled mucous droplets from an individual with an active pulmonary disease. After inhalation, the droplets are deposited in the distal alveoli where they encounter phagocytic cells, i.e. alveolar macrophages, and eventually disseminate through the lymphatics and blood (46, 102, 112). In a healthy human host, 90% of infected individuals remain healthy as a result of development of the protective immune response where bacteria are controlled within granulomas. These individuals are considered to

3 have latent TB infection (LTBI) and are thought to be at least partially protected from future encounters. Of the 10% that develop active disease, it is rarely due to a primary infection but rather a reactivation of infection from an established granuloma in a latently infected person. Latent individuals can remain symptom free for decades as the reactivation most commonly occurs when the person is immunocompromised as in old age or when co-infected with HIV (Figure 1.1).

Figure 1.1. Cartoon depiction of the fate of inhaled M.tb in context of the human host (112)

Throughout the bacteria’s adaptation to its human host, M.tb has adapted many ways of either eluding or manipulating its environment. In particular, the bacteria cell wall contains several immunomodulatory molecules; most notable here are the

4 mannosylated lipoglycans that can mediate entry of M.tb into host cells through C-type pattern recognition receptors (PRRs) and avoid host immune responses (131,132). C-type lectin PRRs are a class of innate receptors that have a Ca2+-dependent carbohydrate binding domains. These receptors play many roles in the host immune system including adhesion and immune surveillance (21). The most notable macrophage PRRs and phagocytic receptors for M.tb immune subversion are the mannose (C-type lectin) and complement receptors (MR and CRs) (117, 118). While the contribution of CRs to M.tb pathogenesis remains unclear, the engagement of mannosylated lipoglycans by the MR is associated with trafficking to the unique mycobacterial phagosome (its host cell niche) which is associated with M.tb survival (60). Once safely inside the macrophage phagosome, M.tb employs several mechanisms, like secreted effector molecules and mannosylated lipoglycans, to arrest phagosome-lysosome fusion, modifying the host response and remains unharmed. Once phagosomal arrest happens, M.tb replicates, with the eventual rupture and spread to other cells (dissemination), and ultimately resides within the granuloma (46).

An adaptive host immune response is necessary to clear (or control) a primary TB infection and gain immunity. The control of M.tb is in large part due to a robust CD4 and

CD8 T cell response and production of interferon-gamma (IFN) by engagement of HLA

Class I and II, and CD1b-restricted T cells (39, 129, 130). IFN is instrumental in activating infected macrophages to effectively fuse the phagosome with the lysosome, and enabling production of reactive oxygen and nitrogen species to kill the bacteria.

Dysregulation and/or evasion of the innate and adaptive immune system provide the opportunity for the establishment of a latent infection. Latent infections add an additional

5 treatment barrier in that the bacteria take on an altered metabolic and replicative state making themselves more resistant to antimicrobial insult. M.tb has the ability to lie in wait for a person’s lifetime with reactivation in 5-10% of latently infected individuals which is triggered by a weakened immune system. (46)

1.2.1 Drug resistance

Since the advent of the use of antibiotics and antimicrobial therapies to treat bacterial infections, the management of infectious diseases like TB has been possible.

The first line treatment for active TB involves taking 4 drugs: isoniazid, rifampicin, pyrazinamide and ethambutol, for the first two months, followed by isoniazid and rifampicin for an additional four months. After the 6 month treatment, patients are typically considered cured if they have drug susceptible strains of bacteria. Non- compliance and relatively limited availability of new treatments have generated multi- drug resistant (MDR) and extremely-drug resistant (XDR) M.tb strains creating a newer, more dangerous problem than before (143). MDR, defined by resistance to two or more of the first line drugs, is largely due to non-commitment to the drug regimen. The second line drugs are less effective, more toxic, and less available in developing countries where a majority of the instances of resistant infections have been reported (143). Treatment can last up to 2 years with reduced cure rates. The complexity and time course of the second line drug treatments along with non-compliance to these regimens, has led to XDR strains of M.tb, defined by resistance to both first and second line drugs (124, 143).

1.3 The mycobacterial cell wall

Mycobacterium species are compositionally unique organisms belonging to the phyla, Actinomycetes, which also includes Corynebacterium, Rhodococcus, Nocardia and Streptomyces. Although they are phylogenetically related to high G + C Gram- positive bacteria, they have been re-classified Acid-fast bacteria. Their cell envelope includes structural components of both Gram negative and Gram positive bacteria in addition to molecules exclusive to mycobacterium, like mycolic acids. This unique configuration provides M.tb with a physical barrier to environmental insults and is responsible for the variable retention of the Gram stain, giving it a previous moniker of

Gram-variable (Figure 1.2)

Figure 1.2. Mycobacterium tuberculosis cell wall depiction.

The uncommon cell envelope which shields M.tb from outside stresses, has also given researchers specific targets for elimination of the microbe. In fact, 3 of the 4 first line drugs target cell wall biosynthesis. Isoniazid blocks fatty acid synthase thereby inhibiting mycolic acid synthesis; ethambutol inhibits arabinogalactan synthesis by blocking an arabinosyl transferase; pyrazinamide indirectly blocks fatty acid synthesis by acidifying the cytosol; and rifampin targets the β subunit of RNA polymerase halting transcription (143).

1.3.1 Mycobacterium pseudocapsule

Many prokaryotes possess a layer of material, comprised mostly of polysaccharides, outside their designated cell wall. The capsule, which is mostly associated with Gram negative pathogenic bacteria, serves many functions including promoting bacterial adhesion and biofilm formation, inhibiting or manipulating phagocytosis by phagocytes, and/or serving as a nutrient reserve. For mycobacteria, the capsule is less well defined and a debated subject, but a comparison of mycobacterial species capsules showed a correlation between pathogenicity and the quantity and composition of extra cellular material (ECM) (67, 95). The data showed that large amounts of ECM that are carbohydrate belong to the pathogenic mycobacteria, which aid in the slower growth rate of these mycobacterium species. Smaller amounts of protein rich ECM were shown in the fast-growing non-pathogenic strains, again pointing to a potential reason for the varying growth rates among species.

The exact function and composition of the M.tb capsule was defined by a technique using 4mm glass beads and shaking, which releases the capsular material leaving the rest of the cell wall intact. The composition of the M.tb capsule was surprisingly virtually lipid free and made up of the neutral sugars glucan, arabinose and mannose. It has been demonstrated that loss of the capsule renders M.tb defective in binding to CR3 and entering the macrophage. (102) It is interesting to note that the removed capsular material showed similar structure via NMR and antigenic properties to those molecules collected from the culture filtrate. This result may point to M.tb actively sloughing off this material in an attempt to deceive the immune system into targeting non-essential outer structural molecules rather than crucial portions of the intact bacteria.

1.3.3: The Mycolyl-Arabinogalactan-Peptidoglycan Complex

The core of the mycobacterial cell wall is the covalently linked mycolyl- arabinogalactan-peptidoglycan (mAGP) complex. This complex is comprised of an arabinogalactan (AG) moiety anchored into a peptidoglycan (PG) layer unique to mycobacteria and esterified at the distal end by a dense layer of long chain mycolic acids

(14, 20, 57, 81, 91). All bacteria possess peptidoglycan as part of their cell wall to provide structural integrity as well as maintaining osmotic pressure but it is no surprise that mycobacteria have a slightly altered PG compared to most other bacteria. The peptidoglycan basic structure is repeating units of N-acetylglucosamine and N- acetyl/glycolylmuramic acid cross-linked by short peptides (24). The mycobacterial twist is that the N-acetyl group at carbon 2 of the muramic acid is preferentially hydroxylated

9 to an N-glycolyl group through the action of an enzyme called N-acetyl muramic acid hydroxylase, NamH (107). It has recently been shown by inactivation of the namH gene in a fast-growing surrogate host M. smegmatis, that this alteration is necessary to alter host cytosolic pathway binding, specifically to Nod-like receptor 2 (NOD2) and increasing TNFα production (50).

Another distinctive aspect of the mAGP is the AG which lacks repeating units and is instead made up of a distinct structural motif: The entire AG structure is tethered to the

PG at the C-6 position of the N-glycolylmuramic acid by a linker unit containing a diglycosylphosphoryl bridge, α-L-Rha-(1→3)-α-D-GlcNAc-(1→P), common among only Actinomycetes (16). The galactan component presents in the form of 30 alternating

β(1→5) and β(1→6) galactofuranose residues in a linear fashion. The arabinan component is similar with about 30 arabinofuranosyl residues linked to the linear C-5 of to some of the β(1→6) galactofuranose residues (16, 23, 79).

The last and seemingly most remarkable feature of the mycobacterial mAGP is the mycolic acids, which provide architecture and are the major reason for the cell wall impermeability. Mycolic acids are long-chain fatty acids, up to 90 carbon atoms long, that are α-branched and β-hydroxylated. These lipids make up 60% of the dry weight of

M.tb compared to 10% in most other bacteria and they are bound to the AG by esterification of a terminal pentaarabinofuranosyl (20, 24).

1.3.2 Cell wall-associated proteins

There are a variety of cell wall-associated proteins important in maintaining the structure, attaching to surfaces, inhibition or manipulation of phagocytosis, sensing environmental stresses, etc. Identification of the immunogenic molecules responsible for the efficacy of the BCG vaccine and positive reaction to the PPD skin test were first done in the 1930’s by Seibert and Munday (25). They demonstrated that the most immunogenic molecules, the “tuberculin” were found in the protein fraction of the bacteria lysate. Later work showed that these molecules were also found in culture filtrate obtained after bacterial growth in broth and even more recent work demonstrates the necessity of the molecules for initial and propagated bacilli growth in vivo (24). The list of cell wall proteins would be too much to consider, so the proteins discussed further will constitute some of the most studied immunomodulatory and immunogenic of these molecules in M.tb. They can be divided into two categories: membrane-bound or secreted.

1.3.2.1 Membrane-bound proteins

One membrane-bound protein not directly associated with pathogenicity is the porin. A porin is an arrangement of proteins that provide a water-filled passageway for small and hydrophilic molecules to pass through the membrane. In thinking of the mycobacterial cell wall containing so many hydrophobic lipids, these pores become a

11 crucial part in the surveillance of the environment and provide a way to maintain pH homeostasis and sense nutrient availability. Blockage of these channels renders the bacteria susceptible to killing in vivo, which potentially provides evidence for its contribution to virulence. Although it was previously thought that this is the mechanism by which drugs can be effectively effluxed from the bacteria, recent work shows that a separate efflux mechanism is responsible for this and the susceptibility of porin blocked bacteria is due to a decrease in internal pH (90, 100).

1.3.2.2 Secreted or unbound proteins

There are many known and unknown secreted proteins and ones that provoke the most immune response are antigen 85 complex, PE and PPE family of proteins, CFP-10 and its partner ESAT-6. The mechanism responsible for the secretion of proteins is the

ESX-1 system driven by a postulated Type VII secretion system which has yet to be defined in terms of machinery and function (76). Antigen 85 is a complex of three proteins involved in mycolyltransferase activity and is the predominant secreted product by M.tb. It binds to fibronectin on host cells, facilitating entry and dissemination (65,

137). The PE and PPE proteins comprise a 10% of the entire genome of TB and although many of the functions have yet to be elucidated, their role as immunogenic molecules is solid. These proteins have been shown to be integral in the movement out of the phagosome during the lysis of the host cell. (2, 27). Lastly, CFP-10 also gets its name in part due to protein mobility on by SDS-PAGE at 10 kDa. Early secretory antigenic target, ESAT-6 also got its name in part due to its migration at the 6 kDa position. These

12 molecules are highly immunogenic separately and when they form a heterodimeric complex which binds host cells and modifies the host response (30, 110, 124).

1.3.3 Peripheral lipids and carbohydrate-linked lipids

There are many types of lipids and carbohydrate-linked lipids that are not a permanent fixture but found throughout the M.tb cell envelope. Among the vast array of these nomadic, non-covalently associated molecules, the discussion will be narrowed to the groups common only to slow-growing mycobacterial species and shown to have an effect on virulence and/or pathogenicity and these include: trehalose dimycolate (TDM), sulfolipids (SL), phthiocerol dimycocerosate (PDIM) and mannosylated lipoglycans (7,

20, 53, 54). The most studied and still puzzling molecule in the list is TDM or cord factor. A seminal observation of M.tb by Robert Koch is that mycobacteria are pressed together and arranged in bundles (cords). Garnder Middlebrook first made the correlation between the cording phenotype and virulence by comparing M.tb H37Rv,

H37Ra and BCG. M.tb H37Ra and BCG are attenuated strains derived from M.tb H37Rv and M. bovis, respectively, and lack the cording phenotype (53, 54). The genetic basis for the attenuation of each organism is not related between the two strains, although the phenotypic loss of the cording effect due to disruption of the normal lipid synthesis is.

TDMs were first isolated Bloch in 1950 after a petroleum ether extraction and when injected into animals caused severe toxicity. It is the most abundant component of

M.tb, the largest naturally made lipid known and has two sets of surface-dependent

13 activities. The first activity is on the bacterial surface where TDM forms non-toxic micelles helping create a hydrophobic barrier to antimicrobials and prevention of host killing mechanisms. The second activity is initiated by contact with the host and involves a conformation change of the TDMs from micelles into a monolayer helping block phago-lysosome fusion and driving caseating necrosis. It has also been shown that M.tb produces and releases much more TDM during infection than is necessary for survival.

(12, 53, 54).

Although not as abundant as TDM but coming in at a close second place for quantity in the cell envelope is a group of sulfated trehalose esters called sulfolipids, namely sulfolipid-1 (SL-1), exclusive to pathogenic mycobacterial species and absent in avirulent strains. It has been 60 years since Bloch’s first described the extraction of M.tb cell wall lipids and made a correlation between them and the virulence of these bacteria.

Since then, researchers have been able to isolate each component of this extraction and implicate its role in pathogenesis but the role for sulfolipids remained elusive until later.

Purified SL-1 added to in vitro studies demonstrated that this molecule can alter host responses by inhibiting phagosome-lysosome fusion, modulating the reactive oxygen response and changing cytokine signaling (37). Unfortunately, M.tb strains lacking sulfolipids in mice and guinea pigs did not recapitulate the previous findings. However, it was recently shown that SL-1 mediates the susceptibility of M.tb to a human cationic antimicrobial peptide in vitro, suggesting that the non-effect in vivo in non-human models is due to a species-specific interaction (34, 120).

Another important lipid in M.tb pathogenesis is phthiocerol dimycocerosate or

PDIM. It begins with a lipid core containing two diols, phthiocerol and

14 phenolphthiocerol with multiple methyl-branched long chain mycocerosic acids and is regulated by a serine/threonine kinase (35). Although its implication in M.tb virulence seemed promising, the genes responsible for the biosynthesis of PDIMs remained elusive, thereby making its direct effect on the host a mystery until the disruption of the pps locus by Azad and coworkers in 1997 (8). In 2004 Rousseau et al. showed that DIM-deficient strains had impaired growth in mouse lungs and were sensitive to reactive nitrogen intermediates released by activated macrophages, implicating a role for PDIM in the early immune response to infection (114). Since then, the mechanism of host alteration has been found to be due to the insertion and disorganization of the host plasma membrane, modifying the biophysical properties to favor efficient receptor-mediated phagocytosis as well as contributing to the control of phagosomal pH (7).

1.3.4 Phosphatidyl-myo-inositol-based lipoglycans: Structure, biosynthesis and Effector function

Figure 1.3. Biosynthesis of phosphoinositol mannosides (PIMs), lipomannan and Mannosylated lipoarabinomannan beginning with an acylated, phosphoinositol ring.

1.3.4.1 Structure and biosynthesis

At the core of all mannosylated lipoglycan synthesis is phosphatidylinositol (PI) which is comprised of a myo-inositol ring with a phosphate at the C-1 position linking a diacylglycerol chain. The commitment step for phosphatidylinositol mannosides (PIMs) biosynthesis is catalyzed by the transfer of a mannose from GDP-mannose to the C-2 position of the PI anchor by PimA, generating the molecule PIM1. The next mannosyl residue addition is at the C-6 position of the inositol ring and results in the generation of

PIM2. Although the mechanism for addition of the third fatty acid is not completely understood, work using a PimB’ mutant suggests that the acylation with palmitate of

PIM2 at position 6 of the first Manp generating Ac1PIM2 occurs after the second mannosylation (14, 42, 64, 86, 87). Experimental evidence for additional mannosylation was done using a PimC over-expressing M. smegmatis strain in a cell-free assay that generated Ac1PIM3 (82). Subsequent studies deleting this gene in BCG had no effect on

PIM biosynthesis compared to WT M. bovis, suggesting the existence of a redundant gene or compensatory mechanism (64). As the step between AcPIM3 and AcPIM4 remains elusive, the last putative glycosyl transferase has been identified in this pathway,

PimE, resulting in AcPIM6 (87). PimE mutants in M. smegmatis showed an accumulation of AcPIM4 which had no effect on growth and viability, although changes in cell wall hydrophobicity suggested a role for this molecule in overall cell wall integrity. It is also of note that PimE is localized to a distinct cell membrane fraction enriched in AcPIM4-6 biosynthesis and was the first to show substrate specificity for

16 polyprenolphosphate mannose (PPM) contrary to PimB’ and PimC, specifically using

GDP-mannose as the donor (86, 127).

The generation of AcPIM4 becomes a biosynthetic pathway branching point since this molecule either gets further mannosylated by PimE described above or it becomes a substrate for the generation of a more complex intermediary lipoglycan, lipomannan

(LM). Derivation toward LM synthesis starts with another PPM substrate-specific mannosyltransferase, MptB, catalyzing the addition of 12-15 Manp at the proximal end with its counterpart, MptA, adding Manp to the distal end (79, 81, 82). Addition of 55-70 linear α(1→5) linked arabinose residues (Arap) to the LM core via the arabinosyltransferase, EmbC, is the first step in the generation of lipoarabinomannan

(LAM) (38, 91). Functional studies replacing an M.tb embC gene for a M. smegmatis homolog demonstrated that the size of the resulting LAM was dependent on the level of expression. Capping of these arabinan termini involves mannose, is only seen in pathogenic mycobacteria and the extent is specific to each species, and finally generates mannosylated LAM or ManLAM versus PILAM found in M. smegmatis (91).

1.3.4.2 Host cell effector functions

The terminal mannose cap structures of the high order PIMs (AcPIM4-6) and

ManLAM bind to the host macrophage via the MR, the predominant C-type lectin on alveolar macrophages, to enter the cell (118, 119). This is the first instance where we begin to see forms of host mimicry rather than being a structurally unique immune- modulator by mycobacteria. In vitro, these lipoglycans also play a key part remodeling

17 the phagocytic compartment, modulating host response and blocking phagosome- lysosome (PL) fusion, creating a niche for the bacteria to survive (58, 60). The basic structures of the mannosylated lipoglycans have been described but just like the diversity among animal populations, there is also much diversity among clinical isolates of M.tb

(128, 143).

1.3.5 Putative Mannose donor biosynthetic pathway

The synthesis of the lipoglycans is quite a lengthy pathway and difficult to study as a whole, so the emphasis of this introduction will deal with the synthesis of the mannose donor molecules. The terminal mannose caps as well as the mannan structures in the core of these molecules are synthesized through a variety of specific mannosyltransferases that use the donors GDP-mannose and polyprenyl phosphate mannose (PPM), that are products of the mannose donor biosynthesis pathway (49).

Functional studies to deteremine the function of these genes have been done exclusively in the surrogate organisms Ms and Cg and have yet to be studied directly in M.tb. The putative genes of this pathway in M.tb are orthologs of those in the attenuated vaccine strain M. bovis BCG and include manA,(99), manB (74), manC (92), and ppm1 (11, 48,

84, 106, 117). As genomic location can sometimes indicate function, there are several other neighboring genes, like whiB2 (62, 115), Rv3256c, Rv3258c and Rv3253c whose functions are unknown, but are potentially contributing members of the mannose donor biosynthesis pathway (Figure 4).

Figure 1.4. A putative mannose donor biosynthesis pathway. Depicted is a putative mannose donor biosynthesis pathway for the building of two key mannose donor molecules, GDP-mannose and polyprenyl monophosphate mannose (PPM) that serve as substrates for mannosyltransferases. The inset shows the genomic location and arrangement of key enzymes, manA, manB and manC, in the pathway as well as the other genes of interest with unknown functions, Rv3253c, Rv3256c, Rv3258c, used for the transcriptional expression study.

1.3.5.1: GDP-mannose pyrohoshorylase, manC

Although the functions of several of the genes proposed to be part of the mannose donor biosynthetic pathway have been elucidated in the related but non-pathogenic

Actinomycetes like M. smegmatis (Ms) and Corynebacterium glutamicum (Cg). The first

19 gene in this pathway to be analyzed was Rv3264c, or M.tb manC. This enzyme catalyzes the final rate limiting step in the generation of GDP-mannose from GTP and α-D- mannose-1-phosphate and has been shown to be essential for the growth of M.tb in vitro

(109, 117). The enzymatic function described was GDP-mannose pyrophosphorylase by isolation of the purified protein in a soluble Ms fraction (92). The Ms manC,

MSMEI_1784 shares an 80% identity to eukaryotic GDP-mannose pyrophosphorylases from pig liver and the yeast, Saccharomyces cerevesiae. Studies showed that unlike the eukaryotic enzymes with phophomannose/phosphoglucose isomerizing functions, the Ms

GDP-mannose pyrophosphorylase was limited to only the production of GDP-mannose.

The reaction was inhibited by GDP-glucose and glucose-1-phosphate providing evidence for its substrate specificity for mannose-1-phosphate. However, the DNA similarity between MSMEI_1784 and M.tb manC is 78%, so another functional study was done in the other surrogate host, Cg. Surprisingly the proposed GDP-mannose pyrophosphorylase in Cg, NCgl015 was not essential with the phenotype restored by complementing with M.tb manC (83). The cell wall of the Cg mutant was deficient in

PIMs, LM and LAM indicating the enzyme’s importance in generating mannosylated lipoglycans.

1.3.5.2: Phosphomannomutases, manB and pmmB

manB and pmmB belong to a class of enzymes called phophomannomutases

(PMM) that can interconvert mannose-6-phosphate and mannose-1-phosphate. PMMs are highly conserved enzymes and congenital deficiencies in their activity have been linked

20 to human diseases states ranging from hypoglycemia to severe mental psychomotor retardation and neural pathology (41). The most studied bacterial PMM belongs to

Pseudomonas aeruginosa and has been shown to be a critical enzyme in the biosynthesis of alginate and/or LPS via a common intermediate linker, GDP-mannose, and essential for virulence (141). To date, there is only one report describing a mycobacterial PMM in the context of virulence by over-expressing M.tb manB in Ms, resulting in increased PIM biosynthesis (74).

1.3.5.3: Phosphomannose isomerase, manA

Phosphomannose isomerase (PMI) is an enzyme that interconverts fructose-6- phosphate and mannose-6-phosphate and is a critical step of the of the mannose donor biosynthetic pathway in the absence of exogenous mannose. Evidence of mannose metabolism requirements for mycobacterial species was shown by deleting the Ms manA gene, MSMEI_1836, resulting in a mannose auxotroph. In the absence of exogenous mannose, the mutant displayed a hyperseptation phenotype and the synthesis of mannolipids and methylmannose polysaccharides synthesis was halted. To date, no work has been done to elucidate the M.tb manA gene or to associate it with virulence (99).

1.3.5.4: Neighboring genes and their postulated functions Rv3253c and whiB2

The genomic location of the mannose donor biosynthetic genes also includes some neighboring, yet uncharacterized genes transcribed in the same direction. Since

21 genes that function similarly or together tend to be located as such, the transcription of these genes were studied in the context of this thesis to identify any potential differential regulation associated with the mannose donor biosynthetic pathway. DNA alignment tools have postulated that the function of Rv3253c involves the translocation of substrate across the membrane but no indication that it is involved in the mannose donor biosynthetic pathway (18). WhiB-like genes are a class of transcriptional regulators that are unique to Actinomycetes that serve to regulate a vast array of fundamental cell processes like septation, environmental stress, nutrient stress, etc. WhiB2 has been studied in M.tb and Ms, and the current published work indicates that it may play an integral chaperone function associated with cell septation (62, 105, 115).

1.3.5.5: Rv3258c

Rv3258c is a very small hypothetical protein with generated by a gene with a 492 bp ORF located approximately 125bps downstream of whiB2 and 100bps upstream of manB in M.tb (18). Bioinformatic analysis gives no indication of the function of the enzyme but it is highly conserved among pathogenic mycobacteria species. In a TraSH mutagenesis library, interruption of the ORF resulted in slow growth in vitro and was essential for establishment of infection in mouse spleens (109).

1.3.5.6: PPM synthase, ppm1

As stated above, PPM synthase is an enzyme that catalyzes the transfer of mannose from GDP-mannose to PPM. PPM is an important mannose sugar donor molecule in that unlike GDP-mannose which remains cytosolic, PPM has the ability to cross the membrane and be utilized by mannosyltransferases in the cell wall. It is believed that the larger cell wall-associated molecules like LM, ManLAM and high-order

PIMs are constructed and terminally mannosylated here where PPM serves as the only mannose donor for the GT-C superfamily (integral membrane proteins) of glycosyltransferases that have substrate specificity for PPM. Conditional deletion of the catalytic portion of M. smegmatis PPM synthase resulted in the loss of high-order PIMs and LM, and although other PPM synthase putative homologues are present in M. smegmatis, they were unable to compensate, thus proving an essential role for PPM synthase (117).

M.tb ppm1 shares ~70% identity with Ms ppm1 and Ms ppm2 (Ms ppm1/2) but these molecules differ in their genetic organization. M.tb ppm1 is transcribed as one gene comprised of two distinct domains. Ms ppm1/2 is organized on an operon with a 9bp spacer transcribing as two separate genes with a functional association. M.tb PPM synthase possesses a unique protein structure with one domain spanning the plasma membrane and a second domain in the outer membrane containing the catalytic subunit.

While the function of PPM synthase has been described in part in non-pathogenic mycobacterium strains, the secondary structure of the protein has not been elucidated.

There are two theories pertaining to the position of the catalytic subdomain. The first is that the mannose donor for PPM synthase is GDP-mannose which is unable to cross the plasma membrane, placing the catalytic portion on the cytoplasmic face of the membrane.

The second is that that the building of high order PIMs, LM and LAM occur on the extracytoplasmic side of the plasma membrane, which would place the second catalytic subdomain on the extracytoplasmic side (11, 33, 106) (Figure 1.5).

Figure 1.5. Box 1 Images depict the placement (in context of the cell membrane) of the proposed catalytic domains of M.tb ppm1 compared to Ms ppm1/2 (5). Box 2 is the predicted transmembrane regions of the PPM synthase by SOSUI.

Recent studies using the non-pathogenic related organism C. glutamicum (Cg) have shed some light on the function of this enzyme. Cg PPM synthase genes are organized genetically like those of M. smegmatis, in an operon with two distinct ORFs:

Cg-ppm1 and Cg-ppm2, and are analogous to Ms-ppm1 and Ms-ppm2 for this study. The role for Ms or Cg-ppm1 corresponding the M.tb ppm1/Domain 2, is most likely cytosolic

24 and responsible for the transfer of mannose from GDP-mannose to PPM, but an earlier study using a bacterial-two hybrid system showed that Ms or Cg-ppm2, corresponding to

M.tb ppm1/Domain 1, serves to stabilize the synthase in the membrane with no indication of function (11, 49). Recent studies have shown that the ppm2 ORF is responsible for the

N-acylation of LppX in M. smegmatis. For the first time, there is a common biosynthetic pathway in which lipoprotein N-acylation and glycosylation are tightly coupled (84).

Studying this pivotal enzyme in the mannose donor biosynthesis pathway provides a greater understanding of the pathogenicity of M.tb and can help us define novel pathways.

1.3.5.7: Bioinformatics and putative function of Rv3256c

Sequencing of the M.tb genome in 1998 gave us the first insight into a largely uncharacterized chromosome based on DNA homology to known genes (18). However, mycobacterium differs widely from the reference bacteria, like E.coli, used in the sequencing studies. It was initially found that a large portion of the genome is devoted to lipid and carbohydrate metabolism with little indication about the actual function of putative enzymes. As the databases for protein function grew and the mutagenesis tools improved so did our knowledge of just how different mycobacterium species are.

Transposon situ hybridization (TraSH) in M. bovis BCG allowed for a quick identification of genes required for growth under different environmental conditions

(111). High-density insertional mutagenesis coupled with microarray mapping of mutant pools is a powerful method for categorizing genes allowing for solid verification (or not)

25 of the proposed functions provided by homology alone (116). Further analysis using

M.tb TraSH libraries in mice narrowed the list of genes to those that are required for growth in macrophages (109). Platform technologies like mutant libraries and microarrays help take the putative list of more than 4000 genes down to distinct pathways and helps give researchers a solid starting point for characterization.

Initial work on Rv3256c in the mannose donor pathway provided us the first indication of its role in pathogenesis. Rv3256c was found to not be required for growth in broth but was required for growth in macrophages (109). These data, along with DNA alignments predicting its function to be that of an isomerase, suggested its possible role in the mannose donor biosynthetic pathway. Since the initial gene annotations were done by DNA homology referencing very different bacteria like E. coli, it has become increasingly apparent with M.tb that a gene proposed to have one function that often turns out to have another (18). This may indeed be the case with Rv3256c, whose function has remained unknown until the current work. Protein fold prediction software (Phyre2) predicts with 100% confidence over 84% of the protein in the proposed catalytic regions that it is most likely a glucosamine-6-phosphate synthase which would include the originally annotated isomerase activity along with two more activities: glutaminase and synthase activities, thus potentially generating a much different end product than originally designated.

1.3.5.8: Rv3256c as a potential glucosamine-6-phosphate synthase, glmS

Glucosamine-6-phosphate synthases (GlmS) are a unique group of an amidotransferase subfamily of enzymes that does not display ammonia-dependent activity. This enzyme catalyzes the first committed step in the eventual formation of uridine 5’-diphospho-N-acetyl-D-glucosamine (UDP-GlcNAc) which becomes an important structural cellular component in all domains of life but is particularly crucial in the biosynthesis of bacterial and fungal cell walls (78, 88). The molecular mechanism of the reaction catalyzed by GlmS is complex and involves both an amino transfer and sugar isomerisation (Figure 1.6). This unique enzyme functions as a dimer by first engaging fructose-6-phosphate near the C-terminal isomerase domain producing glucose-6- phosphate which causes a conformational change in the N-terminus of the enzyme, allowing for the specific binding of L-glutamine as an amide donor. The amide donation and subsequent synthase activity that forms glucosamine-6-phosphate are the rate- limiting and irreversible parts of the reaction (88).

Figure 1.6. Functional schematic of glucosamine-6-phosphate synthase with its substrates and immediate products

glmS is an essential gene in prokaryotes and yeast, and dysfunction of human glucosamine-6-phosphates or GFATs have been implicated in cancer and insulin resistance in diabetes. It would then come as no surprise that some organisms utilize gene redundancy and, in fact, this phenomenon has been seen in mice and in a plant symbiote, Rhizobium leguminosarum. R. leguminosarum is a nitrogen-fixing bacterium that lives at the root of leguminous plants and induces the plant to produce sac-like pockets called nodules where they live safely and provide the plant with nitrogen which is especially important for the plant in nitrogen-poor environments. This bacterium has 2 functionally homologous glucosamine homologues, glmS and nodM, and alteration of

28 either enzyme results in poor nodule formation and/or greatly reduced nitrogen fixation

(73).

Although mycobacterium already has a glmS (Rv3436), annotated by DNA homology, there have not been any glucosamine-6-phosphate synthases described for mycobacterium or any related species in the entire order of Actinomycetales. Other enzymes associated with the UDP-GlcNAc pathway like GlmU and GlmM have been identified in mycobacterium, although the organization and regulation are quite different than its E. coli counterpart (69, 142). E. coli UDP-GlcNAc pathway enzymes are part of an operon that is tightly regulated by riboswitches, common in low-GC organisms, and metabolite flux. Protein alignment of Rv3256c with R. leguminosarum NodM by Clustal

Omega shows an appreciable degree of similarity but most importantly they are nearly identical in the proposed catalytic regions designated within the alignment (Figure 1.7).

Therefore we hypothesized that Rv3256c is a glmS homologue essential for the entry and acclimation of the macrophage and address this possibility this in the context of this thesis.

Figure 1.7 Alignment of protein sequences from M.tb Rv3256c and R. leguminosarum NodM using Clustal Omega. The highlighted portions represent identical, strongly similar and weakly similar amino acids. The X near the N-terminus represents the proposed glutaminase binding site and the + near the C-terminus represents the proposed catalytic sites for the isomerase activity. Proposed catalytic sites were generated by Phyre2

1.3.5.9: Sugar donor molecules, GDP-mannose and polyprenolphosphate mannose

The mannose donor biosynthetic pathway results in the formation of two carrier molecules that serve as substrates for mannosylation; guanosine diphosphate mannose

(GDP-mannose) and polyprenol phosphate mannose (PPM). GDP-mannose is a ubiquitous mannosyl donor and remains cytosolic. PPM is generated from GDP- mannose via an enzyme called polyprenyl phosphate mannose synthetase encoded by the ppm1 gene. PPM is able to cross the plasma membrane and is utilized by substrate specific mannosyltransferases for the terminal mannosylation of cell envelope associated lipoglycans. To date most of the functional studies of PPM synthetase have been done in

Ms which encodes this enzyme by two separate ORFs, ppm1 and ppm2. The gene products work together to synthesize PPM from GDP-mannose and has recently been shown to have acyltransferase activity as well from ppm2 domain in Cg (84).

1.4: Model organisms

M.tb is one of the slowest growing bacteria known, yet it is by far the most successful human pathogen in history. In optimal growth conditions, M.tb is able to duplicate every 18-24 hours which is difficult enough to keep aseptic and, in addition, aerosol pathogens like M.tb must be handled in a contained facility. Because of these difficulties, use of surrogate organisms have facilitated the progress in basic research.

Recent comparative analysis of mycobacterium and related Actinomycetes demonstrates the genomic divergence of mycobacterial pathogens in context of each other and their relationship to distant relatives. The analysis of protein evolution revealed that fatty acid metabolism, DNA repair, small non-coding RNAs and molybdopterin biosynthesis were

31 greatly expanded in M.tb and are associated with the adaptation from a soil saprophyte to an obligate human pathogen (76).

Figure 1.8. A phylogenetic tree of mycobacterium and related species (76).

One heavily used surrogate is the least related organism to M.tb, Cg (Figure 1.8).

Cg belongs to the same order of bacteria, Actinomycetales, and shares the common ancestor of mycobacteria and streptomyces. Cg is a slow growing, non-pathogenic bacteria mostly used industrially to produce amino acids and enzymes. It tolerates the expression of M.tb genes without any codon bias and the resulting proteins made have not been shown to be toxic in this setting as well. Although the constituents of the cell wall are similar, the ultrastructure depth is less than M.tb due to shorter Cg mycolic acids

(140). Cg has facilitated the initial work on mannosylated lipoglycan pathways since a large portion of those associated enzymes are essential in M.tb.

The most widely used M.tb surrogate is the fast-growing, non-pathogenic, soil- dwelling M. smegmatis Ms is a closer relative genetically and structurally to M.tb when compared to Cg with a few exceptions such as Ms LAM being terminally decorated with phosphoinositol instead of mannose (49). Ms is the most used M.tb surrogate and was used to develop the first transfection system in mycobacterium (56). The rapid growth of this organism was exploited to generate shuttle plasmids containing origins of replication for both E. coli and Mycobacterium species (125). The generation of such plasmids allowed for the expression and construction of vectors in E. coli and terminal expression in Mycobacterium species. But for all of the advantages Ms affords us, it is still distinctly different as exemplified by the genome of Ms being 1.5 times larger than that of M.tb.

The last and certainly not least of all the surrogate organisms used is the highly related M. bovis BCG, which is the attenuated vaccine strain of M. bovis. BCG was generated by serial passages of an isolate from cattle by French scientists Albert Calmette and Camille Guerin in the early part of the 20th Century, thereby generating the vaccine

33 name, Bacille de Calmette et Guerin (BCG). After many rounds of replicate plating,

Calmette and Guerin noticed one bacterial colony that was morphologically distinct from the others and administration of it to guinea pigs provided protection against subsequent challenges with M.tb. Within a decade of the published results, BCG vaccination was disseminated worldwide and is still the most widely used TB vaccine with variable efficacy (112, 143). M. bovis BCG is also a slow-grower and is 99% genetically identical to M.tb making it the closest relative organism for study. Although they are nearly identical organisms, an example of a distinct difference between M.tb and BCG is the exclusion of a Region of Difference 1 (RD1) in the latter that contains an important secretion system and virulence factors for M.tb (104). A second difference is in the overall structure of ManLAM. In BCG it is shorter and more branched making it less prominently exposed on the cell wall versus that of the ManLAM of M.tb (103).

Transcription studies performed in this thesis have demonstrated significant differences in expression between the two organisms which points out that even though they are 99% genetically identical, they are not completely interchangeable and that work performed in surrogate organisms should be verified in M.tb.

1.5: Genetic systems in mycobacterium

Mycobacterial DNA has been detected in the oldest mummified human remains from Egypt to Peru providing direct evidence that the disease was with us well before our species began to diverge and populate the world (5, 25, 101). Therefore it is not surprising that the pathogenesis of M.tb is multi-faceted and that the unique complexity

34 and uniqueness of the cell wall plays a large part. The 60% lipid cell wall provides a physical barrier to antimicrobial insults and its components also modulate the host immune system allowing M.tb to remain a highly successful human pathogen.

1.5.1: Plasmids used in mycobacterial research

Although advantageous for the bacteria, the lipid rich cell wall of mycobacterium makes it very difficult to transform DNA and this is in large part responsible for the considerable delay in the development of genetic tools to manipulate this genus. M.tb was considered to be a genetically intractable organism until 1970 when the first transductions in M. smegmatis provided a way to introduce DNA in mycobacterium, although attempts to transduce M.tb were unsuccessful (51). A very elegant shuttle phasmid genetic system was developed that exploited the replication and manipulation of plasmids in E. coli and also served as phages to introduce foreign DNA into the mycobacterial chromosome (125). It is remarkable to consider that although M.tb was the first disease-causing microbial agent to be isolated, thus proving Köch’s Postulate, it is one of the very last to be able to satisfy Köch’s Molecular Postulate.

Development of the phage-based system ushered in a new era of genetic tools for

M.tb. The first generation of plasmids contained an oriM and oriE, and the MCS under the control of the hsp60 promoter. As the genetic tools were beginning to be refined, our knowledge of the advantages and disadvantages of the plasmid components became clear.

The hsp60 promoter is excellent for constitutive expression but since it is a heat shock protein promoter, its expression greatly increases during cell stress, making overall

35 expression variable (97). The tet-on/off system developed by Ehrt and Schnappinger uses the tetracycline operator site to tightly regulate expression. This does solve the problem of variable constitutive expression but does not come without drawbacks. Even on one of the most tightly regulated plasmids, some leaky transcription occurs and increasing amounts of tetracycline can have effects on the host cell as well. A complete list of the plasmids used in this thesis with their key features is listed in Table 1.

Table 1. List of plasmids used including the promoter, origin of replication for E. coli and Mycobacterium (oriE/OriM), size in base pairs (bp), copy number in E. coli and Mycobacterium (E.coli/Myco), and the key features.

Vector Promoter oriE/ Size (bp) Copy Key Features oriM number E.coli/Myco pMV261 hsp60 Y/Y 4488 High/Low Constitutive expression pMV361 hsp60 Y/Y 5641 High/Low Contains an integrative attP site for insertion into the genome pSE100 Hybrid of Y/Y 5538 High/Low Inducible by anhydrotetracycline myc and tetO pTrcHis2- Hybrid of Y/N 4381 High/NA Inducible with IPTG TOPO trpB and C-terminal 6X His tag lacUV5 pJV53 acetamidase Y/Y 8812 High/High Che9c genes 60-61 recombinases Inducible by 2% acetamide pNIT PnitA Y/Y 6208 High/Low Nitrile-inducible gene expression pJAM2 acetamidase Y/Y 9810 High/Low C-terminal 6X His tag Inducible by 2% acetamide pML1611 myc Y/Y 11651 High/Low HR substrate using gfp-hyg interruption cassette flanked with loxP sites for unmarked deletion sacB, xylE counterselectable markers pNILRB5 Promoterles Y/N 6547 High/NA lacZ and sacB for screening s vector will fuse with pGOAL19 and serve as HR substrate as part of LIC pGOAL19 sacB under Y/N 10435 High/NA lacZ screen, sacB counterselection hsp60 will fuse with pGOAL19 and serve as HR substrate as part of LIC

1.5.2: Gene deletion systems in Mycobacterium

The simplest way to evaluate a particular genetic element in an experimental setting is to eliminate it from the system, observe the change and then replace the element to restore the original phenotype. The same is true in microbial genetics, satisfying

Köch’s Molecular Postulate. The most common way to delete or interrupt a gene is by exploiting the most common mechanism of DNA repair, homologous recombination

(HR), and selective pressure usually via antibiotics. The majority of the systems developed that have these features are done in E. coli and extrapolation to other prokaryotes is done fairly easily. One exception to this is Mycobacterium species. To date the most efficient methods of gene deletion are phage-based, which were not attempted in context of this thesis. Instead, attempts at gene deletion were done using a plasmid donated by Michael Neiderweis (UAB), pML1611, that is an elegantly designed plasmid with markers for screening, selection and counter selection. (Figure 1.9) The plasmid relies on a gfp-hygromycin cassette flanked with loxP sites to interrupt the ORF of the gene in question, uses antibiotic pressure for selection and once the clone is verified, uses cre recombinase to unmark the mutant. Unfortunately, all attempts to transform and get double cross-over events with the plasmid alone failed.

Figure 1.9. Gene deletion vector pML1611. Depicted is the arrangement of genes and restrictions sites for the replicative plasmid used in the improved gene deletion method (Neiderweis, UAB, unpublished).

An alternative strategy attempted in this thesis was the recombineering method based on the E. coli Lambda Red system, which requires the prior transformation into

M.tb of a plasmid (pJV53) that expresses mycobacteriophage recombinases to facilitate specific allelic exchange in an otherwise naturally homologous recombination-poor environment (135). This recombineering system also requires the construction of a linear allelic exchange substrate (AES) that consists of a hygromycin cassette (hyg) flanked by

~ 500bp regions of the gene of interest. The pJV53 plasmid was successfully transformed into M.tb to make a recombineering strain which was subsequently transformed with our AES construct of interest to achieve enhanced homologous recombination. This method has proven unsuccessful so far, even after acetamide induction of the recombinases was confirmed by Western blotting.

Subsequently, another deletion method was developed in Ms using Ligation

Independent Cloning. which employs two plasmids, pNILRB5 and pGOAL19.

Constructs were made easily following a protocol provided by R. Balhana. Construction of the plasmids was straight forward and for the first time, we began to see single cross- over events giving hope for the complete deletion during the second cross-over. But the only flaw in the system was no selectable marker for the last cross-over and every clone evaluated eventually reverted back to WT phenotype.

The world of TB genetics is still a burgeoning field providing many new possibilities which come as a double-edged sword. It was demonstrated by Glickman’s group that mycobacteria do not rely solely on HR as their major DNA repair mechanism and that they employ a non-homologous end-joining (47) that had previously only been seen in Archaea and Eukaryotes (47). This group has also shown that along with HR and

NHEJ, Mycobacterium has a third, unique single stranded break repair (ssr) mechanism.

These findings along with the fact that there is no evidence of horizontal gene acquisition and that all antimicrobial resistance demonstrated by the organism is acquired through mutation in a bacteria with a very low mutation rate indicates a more elegant and directed

DNA repair system than previously thought (75). Until we understand this DNA repair system more clearly, gene deletion will continue to be a challenge in mycobacterium.

1.6: Animal models for TB

There are several animal models like mice, rabbits and non-human primates to help assess virulence but these have various limitations and do not fully recapitulate events in humans (136). These issues must be kept in mind when comparing in vitro data

39 with host primary cells to in vivo data with animal models so that virulence factors that are host specific (particularly in humans) are not dismissed prematurely (25).

Summary

M.tb is one of the oldest and most successful pathogens in human history. M.tb has been with us from the first migrations out of Africa and as human populations expanded, diversification of the bacterial genome in parallel gives us proof of the amazing adaptability of this organism (19). M.tb has developed a plethora of ways to subvert the host immune response and make a home out of what should be its undoing, the phagosome. It is no accident that the bacterium coats itself in an abundant host- mimicking sugar, mannose, when it is most likely to encounter cells like alveolar macrophages that have up regulated C-type lectins such as the MR which recognizes mannans. Since it has adapted to happily reside within us for a lifetime, M.tb has also become romanticized in our culture claiming the lives of some of our most beloved characters. Although we understand that the disease is caused solely by this unicellular organism, the interaction between the host and M.tb is incredibly complex. A more complete understanding of the entry and acclimation of M.tb in host cell macrophages could possibly allow us to prevent infection, limit dissemination during primary infection and provide better immune protection.

Specific Aims

Mannosylated lipoglycans in the M.tb cell wall modulate the host cell niche in vitro and have long been considered major M.tb virulence factors. Putative enzymes involved in the mannose donor biosynthetic pathway have been analyzed in surrogate hosts but their expression levels and functions have yet to be determined in M.tb.

Therefore, we hypothesize:

The regulation and expression of genes in the mannose donor biosynthetic pathway

contribute to entry and acclimation in the host macrophage

To examine this hypothesis we pursued the following Aims:

1. Establish a transcriptional profile of the family of genes including manA, manB, manC, whiB2, Rv3253c, Rv3256c, Rv3258c, pmmB and ppm1 in M.tb and BCG in vitro and within human macrophages

2. Express and characterize the function of Rv3256c and Rv3258c, genes up-regulated upon M.tb entry in macrophages, and ascertain their contribution to macrophage infection.

3. Characterize the role of ppm1 in the mannose biosynthetic pathway and in macrophages

Chapter 2: Transcriptional Profiling of the Putative Mannose Donor Biosynthetic Pathway in Mycobacterium

2.1: Introduction of the Putative Mannose Donor Biosynthetic Pathway

Tuberculosis (TB) kills nearly 2 million people each year and has become the leading cause of death among HIV patients. Although treatments have been available for more than 80 years now, inconsistent completion of antibiotic courses has led to resistance to all current anti-TB drugs among different bacterial isolates (143). The emergence of MDR and XDR strains of TB, especially in HIV patients, poses a serious threat to the control of TB worldwide. Mycobacterium tuberculosis (M.tb) is the causative agent of TB and possesses a compositionally unique cell wall which has been the target of several TB therapeutics like isoniazid and ethambutol. The last new drug marketed for TB was in 1963 (124) and coupled with the development of resistance, emphasizes a critical need for the discovery of new drug targets in M.tb.

It has long been thought that M.tb has co-evolved with its human host which is the only known reservoir (124). The complex pathogenicity of this bacterium in the context of its preferred niche is only partially understood. Comparative genomics between virulent strains of M.tb and the attenuated vaccine strain Mycobacterium bovis BCG

(BCG) have revealed gene deletions in BCG providing insight into some key

determinants of virulence within the Regions of Difference (RD) (38). It was thought that

complementation of the RDs into BCG would restore complete virulence but the result

was only a partial restoration (31, 104). This clearly indicates the complexity of M.tb

pathogenesis and opens the door to investigate other factors such as the regulation of

transcription that contribute to its success as a pathogen.

Table 2.1. DNA sequence identity of the mannose donor biosynthetic pathway relative to M.tb H37RV

Identity to H37RV DNA sequence Gene M. bovis M. M.tb M. M. C. M. BCG smegmat H37Ra marinum leprae glutamic avium is TN um ATCC 13032 manA 100 74 100 82 *NS 69 82 manB 99 77 100 98 97 66 97 manC 100 78 100 81 *NS 72 83 whiB2 100 80 100 84 81 78 84 ppm1 100 80 100 80 75 67 86 pmmB 99 70 100 78 79 79 79 Rv3253c 100 75 100 81 75 86 83 Rv3256c 100 69 100 78 77 92 75 Rv3258c 100 78 100 85 88 72 87 *NS denotes No sequence similarity found

The M.tb cell wall contains the highly mannosylated cell wall components

phosphatidyl-myo-inositol mannosides (PIMs), lipomannan (LM) and mannose-capped

lipoarabinomannan (ManLAM) (14, 42, 43, 129) which are important in TB

immunopathogenesis. The terminal mannose cap structures of higher-order PIMs and

ManLAM bind to the host macrophage mannose receptor in a form of host molecular

mimicry (119, 128). The terminal mannose caps as well as the mannan structures in the

43 core of these molecules are synthesized through a variety of specific mannosyltransferases that use the donors GDP-mannose and polyprenyl phosphate mannose (PPM) that are products of the mannose donor biosynthesis pathway (49)

(Figure 2.1). The putative genes of this pathway in M.tb are orthologs with 99-100% sequence identity to those in the attenuated vaccine strain BCG and include manA (an isomerase) (99), manB (a phosphomannomutase) (74), manC (a GDP-pyrophosphorylase)

(92), and ppm1 (a polyprenyl-phosphate mannose synthase) (11, 48, 84, 106, 117) (Table

2.1). Additionally, there are several other neighboring genes, like whiB2 (an Fe-S clustering molecule and transcriptional regulator) (62, 115), Rv3256c, Rv3258c and

Rv3253c (hypothetical proteins) whose functions are unknown, but are potentially contributing members of the mannose donor biosynthesis pathway. Such knowledge will have direct relevance to the availability of mannose donors for building the cell wall mannosylated components.

Figure 2.1. A putative mannose donor biosynthesis pathway. Depicted is a putative mannose donor biosynthesis pathway for the building of two key mannose donor molecules, GDP-mannose and polyprenyl monophosphate mannose (PPM) that serve as substrates for mannosyltransferases. The inset shows the genomic location and arrangement of key enzymes, manA, manB and manC, in the pathway as well as the other genes of interest with unknown functions, Rv3253c, Rv3256c, Rv3258c, used for the transcriptional expression study.

Expression differences have been described between wild-type and attenuated vaccine strains of M. bovis and it is thought that these differences may have a major impact on antigenic profiles (31). Mycobacterial gene expression work to date has been largely performed using platform technologies like microarrays, and while this approach provides an indication of large global expression changes, it lacks specificity for particular pathways postulated to be attributed to virulence (111). The regulation of

45 expression of these genes in vitro and within macrophages has not been determined in virulent M.tb nor compared with BCG. A more in-depth look at individual pathways will provide us with greater insight into the complexities of transcriptional regulation as a major means of regulating virulence. Here we compared the regulation of expression of genes known or likely to be involved in the mannose donor biosynthesis pathway for generating mannosylated cell wall molecules between M.tb and BCG grown in vitro and in human monocyte-derived macrophages (MDMs). We show significant expression differences of identical genes in the putative mannose donor biosynthesis pathway between these two mycobacterial species. An extended discussion of the function of these genes and analysis of the putative pathway was described in detail in Chapter 1.

2.4: Materials and Methods

2.4.1: Mycobacterial strains and growth media

M. tuberculosis H37Rv (ATCC 27294) and M. bovis BCG-Pasteur (ATCC 35734) were grown either on Middlebrook 7H11 agar (Difco, Franklin Lakes, NJ) with 10% oleic acid, dextrose and catalase (OADC) enrichment for 9-14 days (for macrophage experiments) or in Middlebrook 7H9 broth (Difco) + 10% OADC and 0.05% Tween-80 with magnetic stirring for up to 24 days (138). Single cell suspensions were obtained as previously described (119). Growth curves were determined by taking an OD600 reading

46 of stirred cultures grown in 7H9 + tween 80 every 24 hours and performed in triplicate for each strain.

2.4.2: Isolation of human monocyte-derived macrophages (MDMs)

MDMs were obtained from peripheral blood mononuclear cells (PBMCs) as previously described (118). Briefly, heparinized blood was obtained by venipuncture from purified protein derivative (PPD) negative donors using an approved protocol by

The Ohio State University Institutional Review Board. PBMCs were separated on a

Ficoll cushion and were cultured for 5 days in RPMI medium containing 20% autologous serum in Teflon wells at 37°C with 5% CO2. After 5 days of growth in the Teflon wells,

MDMs were adhered to 100 x 150mm tissue culture dishes with medium in the presence of 10% autologous serum for 2 h, non-adherent cells were removed by washing with pre- warmed RPMI and MDMs were cultured for an additional 7 days in 20% autologous serum, allowing a total of 12 days of growth (94). At day 12, the MDMs were infected at a multiplicity of infection (MOI) of 5:1 in the presence of serum for 2 h at 37°C with 5%

CO2. Counting of CFUs was performed in parallel on agar plates to verify the inoculum used and thus confirm the MOI.

2.4.3: Bacterial lysis, RNA isolation and Real-time PCR.

47 zirconia/silica beads (Biospec Products) coupled with DNase I (Qiagen, Valencia, CA) treatment. Isolation of bacteria from within infected macrophages was achieved by using a guanidinium thiocyanate (GTC)-based differential lysis solution as previously described (85). Bacterial RNA from within macrophages was processed by the procedure as described above. RNA was reverse transcribed to cDNA using 500 U of Superscript II reverse transcriptase with 10 mM dNTPs, 10 U RNase inhibitor, 0.1M DTT, and 3 μg of random hexamers (all from Invitrogen) for 120 min at 42°C, followed by inactivation with 1N NaOH at 65°C for 10 minutes. Control reactions were performed in parallel without reverse transcriptase to verify the absence of DNA contamination. PCR was performed on the resulting cDNA using 300 mM custom-made primers with iQ SYBR

Green Master mix (Bio-Rad) and 4% dimethyl sulfoxide (DMSO). All samples were run

-∆∆C in triplicate and analyzed using the 2 T method and expression was determined relative to the housekeeping gene rpoB (70). GAPDH primers were used to identify contaminating eukarytotic RNA [5’ACTTTGCTATCGTGGAAGGACT 3’ (forward) and

5’GTAGAGGCAGGGATGATGTTCT 3’ (reverse)]. Bacterial gene-specific custom primers are listed in Table 2.2.

2.5: Results

Using basic bioinformatics as well as previous work done in M. smegmatis for manA, manB, manC and ppm1, we constructed a putative mannose donor biosynthesis pathway (Figure 2.1) to depict the building of the two mannose donor molecules: GDP- mannose and PPM. Open reading frames (ORFs) for manA, manB, pmmB (another

48 phosphomannomutase gene), manC, and ppm1 orthologues of M.tb H37Rv were obtained using the Tuberculist. Located near manA, manB and manC in the M.tb genome (Figure

2.1, see inset) were several genes of unknown function (Rv3253c, Rv3256c and Rv3258c) as well as whiB2 which has been recently described as a transcriptional regulator that is embedded in the genomic region of interest (Figure 2.1 inset) (115). Further bioinformatic analysis using basic local alignment search tool (BLAST) from NCBI

(http://blast.ncbi.nlm.nih.gov/Blast.cgi) revealed that all of the genes of interest for our transcriptional profiling study had 99-100% sequence identity with BCG. The locus tag and names of all of these candidate genes in M.tb and BCG are listed in Table 2.2.

2.5.1: Transcriptional profiling of mannose donor biosynthesis genes of M.tb and

BCG grown in broth culture

Growth curves of M.tb and BCG were generated by growing the bacteria under identical culture conditions. Comparable time points were chosen for these two strains to analyze gene expression at different growth phases. Custom primers were designed for each gene listed in Table 2.2, and because of the identical sequences between M.tb and

BCG genes, the same primer sets were used for both strains in all conditions. All expression analyses were determined relative to the housekeeping gene rpoB which remained constant in all conditions for both strains.

Table 2.2. List of custom primers for RT-PCR. Table 1: List of specific primers used for RT-PCR. Locus Gene Primer sequence (5’ to 3’) Rv0667, BCG_0716 rpoB CCTGGAAGAGGTGCTCTACG (forward) GGGAAGTCACCCATGAACAC (reverse)

Rv2051c, ppm1 TGGTTGAAGTCGATCCTTCC (forward) BCG_2070c GCGAACAAGACCAGGCATATG (reverse)

Rv3253c, 53c CCAACTACTCGCCGTTCATT (forward) BCG_3282c GCAGTTGGGTGTATGGAACC (reverse)

Rv3255c, manA GTTCACCACCTGGATTACCG (forward) BCG_3284c AACCCTCGGTGCATAACAAG (reverse)

Rv3256c, 56c CTGACGAGTTCGGGTTGTC (forward) BCG_3285c ACCAGATAGGCGTCATCGAG (reverse)

Rv3257c, manB GATCACGTTGTGGATGATGG (forward) BCG_3286c GTGGATCTGCAGGCCTATGT (reverse)

Rv3258c, 58c GTTGGTGCACGATAGCCTTT (forward) BCG_3287c ACACACAGATCCCACGAATG (reverse)

Rv3260c, whiB2 CCATTCGAGGAACCTCTGC (forward) BCG_3289c CAGGGCGTACTCCAGACACT (reverse)

Rv3264c, manC ACATCGCCGTTAAACACCAT (forward) BCG_3293c GTTCCTCACCCATCTGCTGT (reverse)

Rv3308, BCG_3373 pmmB ATACAGATCACGGCGTCACA (forward) CGCTGGATATAACGGTCGAT (reverse)

*GAPDH primers included as an internal eukaryotic control **All genes listed share a 100% nucleotide sequence identity between strains and the same primers were optimized in each organism separately.

Our results for the transcriptional profile in broth show that there are differences in expression of certain genes between M.tb and BCG under the same growth conditions

50 despite their identical sequences. In M.tb (Figure 2.3A), whiB2 has relatively high expression compared to the other genes in the study. As the growth continues, the expression level of whiB2 decreases during the exponential phase but then increases as growth continues and peaks at late stationary phases. In BCG (Figure 2.3B), whiB2 begins with a relatively high expression, peaks at early log phase and decreases as growth continues, a trend which is opposite to that seen in M.tb for the same gene. During the early log phase of growth of BCG, Rv3253c expression is higher than the other genes and decreases to baseline as the growth continues. The most notable differences in expression between the strains were observed during the stationary phase of growth. In

M.tb, manC expression spikes during this growth phase and manB increases in expression. In BCG, manA spikes during this growth phase and Rv3258c and manB show increased expression.

Figure 2.3. Transcriptional profile of putative mannose donor biosynthesis genes of M.tb (A) and BCG (B) grown in broth medium. cDNA was made from RNA extracted from both strains at their different growth phases and subjected to Real-Time PCR for determining the amounts of transcripts of each gene as a measure of its expression. The Y-axis represents expression of target genes relative to that of the housekeeping gene, rpoB, and X-axis represents optical density (OD)600 of the culture corresponding to the time points (in days) used for generating growth curves. Each data point in the graph is the mean ± SD of values from triplicate wells. Shown are graphs plotted from representative experiments (n=3).

2.5.2: Transcriptional profiling of mannose donor biosynthesis genes of M.tb and

BCG in macrophages

Since human macrophages are the host cell niche for M.tb, we next infected these cells with M.tb and BCG and developed a reliable assay for recovering intact intracellular mycobacteria for RNA extraction. Using a specialized lysis buffer (GTC), we were able to stabilize bacterial RNA, and keep the bacteria intact while completely lysing the monolayer. Optimization required an MOI of 5:1 (bacteria per MDM) for the macrophage infection which allowed for the infected cell monolayer to remain intact for at least 5 days post-infection while still being able to generate enough bacterial RNA for cDNA synthesis, especially at early time points. Analysis of the purity of RNA was tested by no RT in the cDNA synthesis parallel to experimental samples, assuring no contaminating genomic DNA was present, while the overall integrity of the RNA was tested by analyzing the 16S and 23S ribosomal RNA with the Experion Automated

Electrophoresis Station

The time points chosen correspond to three different aspects of the macrophage infection and are categorized as: Entry at the 2 hour time point, Acclimation at the 24 and

48 hour time points (the bacterium has a doubling time of approximately 24 hours) and

Proliferation at the 72, 96 and 120 hour time points. Figure 2.4 is a cartoon representation of an M.tb infection in human macrophages.

Figure 2.4. Cartoon representation of the early time points of macrophage infection from the bacteria’s perspective. Entry is represented at the 2 hour time point. Acclimation is represented at the 24 and 48 hour time points. Proliferation is represented at the 72, 96 and 120 hour time points.

In order to carefully differentiate unique expression changes in bacteria residing within macrophages from those in the “input” bacteria for the experiment, a single cell suspension was prepared from plate-grown bacteria and analyzed by qRT-PCR. Results of input bacteria expression profiles show similar trends for both bacteria with whiB2

54 being highly expressed relative to all other genes in the set (Figure 2.5A). Results from intra-macrophage bacteria indicate that whiB2 is consistently highly expressed in both

M.tb and BCG throughout the course of infection. In M.tb (Figure 2.5B), there is an initial peak in expression for Rv3258c, Rv3256c and ppm1 after 2 hours of infection compared to input bacteria and then expression levels for all three genes steadily decrease over time. manC and manB are moderately expressed during the course of infection. Results for BCG (Figure 2.5C) show little to no change in expression of all genes in the study with the exception of whiB2 which remains high throughout the course of infection.

Figure 2.5. Transcriptional profile of putative mannose donor biosynthesis genes in M.tb and BCG from infected MDMs. Single suspensions of M.tb grown on 7H11 agar plates were used to infect MDM monolayers at an MOI of 5:1. GTC lysis buffer was used to extract RNA at 24, 48, 72 and 120 hour time points. cDNA was synthesized from RNA and subjected to Real-Time PCR for determining the amounts of transcripts of each gene as a measure of its expression. (A) Represents the transcriptional profile of the single cell suspension bacteria prior to infection. (B) and (C) represent the transcriptional profiles of M.tb and BCG, respectively, harvested from infected macrophages. Each data point in the graph is the mean ± SD of values from triplicate wells. Shown are graphs plotted from representative experiments (n=5).

2.6: Discussion

In the present study we show that there are differences in the transcriptional expression profiles of identical mannose donor biosynthetic genes between M.tb and

BCG strains grown in broth culture as well as within human macrophages. One of the most noticeable differences in expression in broth between these two mycobacterial strains is that of the Fe-S containing transcriptional regulator, WhiB2 (Figure 2.3). The functions of the WhiB-like molecules in the context of Mycobacterium species have been described to be involved in crucial cellular processes like cell division, nutrient starvation, stress, antibiotic resistance and pathogenesis (105, 116). The only trend difference seen for whiB2 between M.tb and BCG was during growth in broth. Since no significant difference in expression of whiB2 was observed between M.tb and BCG in macrophages during the infection period as well as in the input bacteria (Figure 2.5A), this suggests that it may not play a direct role in pathogenesis. However, it will be interesting to know whether whiB2 has any effect on mycobacterial mannosylation because it is located in the neighborhood of the mannose donor biosynthesis pathway (see

Figure 2.1, inset) and known to be essential for mycobacterial growth (105).

M.tb in broth culture during its late log to early stationary growth phase showed relatively increased expression for two genes, manB and manC (Figure 2.3A), encoding a phosphomannomutase and a GDP-mannose pyrophosphorylase, respectively, which are key enzymes in the mannose donor biosynthesis pathway. In BCG during the similar growth phases, the genes that are increased in expression are manA, which

57 encodes a phosphomannose isomerase, and Rv3258c, a hypothetical gene of unknown function (Figure 2.3B). This differential regulation of expression of mannose donor- related genes between these two mycobacteria could contribute to the known structural differences in mannosylated molecules such as ManLAM. The structure of BCG

ManLAM differs from M.tb by having a shorter mannan backbone with highly branched

3,5-linked D-Araf residues and the mannose-capping motifs in BCG are dominated by monocaps as opposed to those in M.tb which are dominated by di- or tri-caps (103). The goal of future research will be to link changes in gene expression for the mannose donor biosynthetic enzymes with the level of biosynthesis of mannose-containing cell wall molecules produced during infection.

Gene expression differences between M.tb and BCG in broth may relate to the fact that BCG has multiple deletions and accumulated many mutations in the genome that have led to adaptation to growth on glycerol-containing media (31). Our current work shows that BCG has reduced expression of several genes both in broth (Figure 2.3B) and within macrophages (Figure 2.5C).

The expression profile of M.tb and BCG mannose donor biosynthesis genes in macrophages is of particular importance not only because macrophages are the natural host cell niche for M.tb but also because the expression profiles are highly reproducible among different donors. Because humans are a heterogeneous population, it is often speculated that donor-to-donor variation might alter bacterial gene expression which was not found to be the case with our study. The transcriptional expression profiles of M.tb and BCG genes from bacteria within MDMs are representative of 5 independent experiments with 5 different donors, all generating similar transcriptional profiles. These

58 reproducible data argue that the mycobacterial transcriptional events are required for bacterial acclimation, survival and proliferation in the host cell. In human macrophages, the distinct differences in the pattern of expression of the majority of the genes under study between M.tb (Figure 2.5B) and BCG (Figure 2.5C) suggest a possible link to the virulence of M.tb. It is of particular interest that the genes Rv3256c, Rv3258c and ppm1 were highly expressed in M.tb 2 hours post-infection of macrophages and then gradually decreased. This suggests that the expression of these genes during the early period of infection is important for the entry and/or acclimation of the bacterium to its preferred host cell environment. Furthermore, the elevated expression of manB and manC at 48 and

72 hours post-infection (Figure 2.5B) suggests the possibility of an increased need for mannose donor molecules as the bacteria begin to replicate intracellularly after the initial adaptation period. In BCG, most of the identical genes under study were expressed at low levels (Figure 2.5C) in macrophages, suggesting a lesser need of these genes for the purpose of infection and adaptation by BCG. The results thus obtained from human macrophages in our study give us new insight in the regulatory nature of transcription potentially related to virulence and provide us with an increased understanding for rational selection of bona fide drug targets for TB.

Chapter 3: Elucidation of Rv3256c, a Putative Enzyme Involved in the Mannose

Donor Biosynthetic Pathway in Mycobacterium

3.1: Introduction

Further understanding the dynamics and structure of the unique mycobacterium cell envelope provides us with many potential targets for new therapies. For instance, cell wall mannosylated lipoglycans have been shown to be important in the entry and acclimation of M.tb to the macrophage. Rather than evaluate the huge array of enzymes involved in building the complete lipoglycan structures, we chose to focus on a specific mannose donor biosynthetic pathway and generation of the two mannose donor molecules, GDP-mannose and PPM. Transcriptional profiling of genes postulated to be involved in this pathway in human macrophages allowed us to better ascertain the changes in transcription for early time points of infection and helped us identify previously uncharacterized neighboring genes that may contribute to virulence.

The transcriptional profile of M.tb from within MDMs in Figure 2.5B provided us with information about the regulation of the mannose donor biosynthetic genes in the early aspects of macrophage infection. Along with the putative genes in the pathway, some neighboring genes (Rv3256c and Rv3258c) that are transcribed in the same

60 direction were included in the profile. Interestingly, they were up-regulated at the Entry time point of 2 hours and decreased as the infection progressed and returned to baseline copy numbers in the bacterial proliferative phase. This led us to a more narrowed hypothesis that Rv3256c and Rv3258c are important in the entry and acclimation of M.tb to its host cell, the macrophage. Rv3258c will be discussed in detail in the next chapter.

The initial annotation of the M.tb genome by DNA homology attributed an isomerase function to the ORF of RV3256c. Secondary structure based on the protein sequence detailed an additional two functions: potential glutaminase and synthase activities, changing the putative enzyme function from just an isomerase to a glucosamine-6-phosphate synthase (GlmS). Although mycobacterium already has a glmS

(Rv3436), annotated by DNA homology, there have not been any glucosamine-6- phosphate synthases described for mycobacterium or any related species in the entire order of Actinomycetales. Also of note is that redundancy for this enzyme has been described for a plant symbiote in which the second copy is used to induce nodule formation by its accompanying host. Therefore we hypothesized that Rv3256c is a glmS homologue essential for the entry and acclimation to the macrophage which is the focus of this chapter.

3.4: Materials and Methods

3.4.1: Mycobacterial strains, plasmids and growth media

M. tuberculosis H37Rv (ATCC 27294) and M. smegmatis (ATCC 700084) were grown either on Middlebrook 7H11 agar (Difco, Franklin Lakes, NJ) with 10% oleic acid, dextrose and catalase (OADC) enrichment for 9-14 days (for macrophage experiments) or in Middlebrook 7H9 broth (Difco) + 10% OADC and 0.05% Tween-80 and 20ug/ml kanamycin (138). Single cell suspensions used for infections were obtained as previously described (118). A modified Sauton’s media was prepared minus glycerol and Tween-80, supplemented with 2% D-glucose as the sole carbon source and

0.00025% Tyloxapol as a detergent.

Plasmids: pSMT3 is a lab stock plasmid with a MCS under the control of the constitutively expressing promoter, hsp60 (52). The ORF of Rv3256c was amplified from M.tb H37Rv genomic DNA for insertion into the multiple cloning site with a HindIII recognition site added to the 5’ end and an EcoRV recognition site added to the 3’ end. pJAM2 was a gift from Katarina Mikusova (Comenius University, Bratislava, Slovakia)

It contains a 6 x Histidine tag on the C-terminus and is under the tight regulation of the acetamide inducible promoter. The ORF of Rv3256c was amplified from M.tb H37Rv genomic DNA for insertion into the multiple cloning site with ScaI recognition sites added to the 5’ and 3’ ends. As a vector control, M. smegmatis strains harboring pSMT3 and pJAM2 plasmids with no insertion in the MCS were used in parallel to all experiments.

Induction of M. smegmatis clones with pJAM56c was done using a 2% acetamide in the modified Sauton’s media listed above, shaking for 24 hours with 20ug/ml

Kanamycin. Cells were lysed using a Fast-prep-24 instrument and lysing matrix B beads

(116911100). His-tagged proteins were isolated using a HisPur Ni-NTA purification kit with the accompanying protocol, visualized by Coomassie stained SDS-PAGE gels and quantified using a BCA assay (131). All plasmid constructs were verified by amplifying over the MCS of the plasmid and will be further tested by sequencing.

3.4.2: Isolation of human monocyte-derived macrophages (MDMs)

The Ohio State University Institutional Review Board. PBMCs were separated on a

Ficoll cushion and were cultured for 5 days in RPMI medium containing 20% autologous serum in Teflon wells at 37°C with 5% CO2. After 5 days of growth in the Teflon wells,

MDMs were adhered in 24 well tissue culture plates with medium in the presence of 10% autologous serum for 2 h. Non-adherent cells were removed by washing with pre- warmed RPMI and MDMs were cultured for an additional 7 days in 20% autologous serum, allowing a total of 12 days of growth (94).

3.4.3: Enumeration of bacterial CFUs and cell association with macrophages

At day 12, the MDMs were used to analyze survival by a CFU assay using a multiplicity of infection (MOI) of 5:1 (bacteria:MDM); cells were treated with DNAse I, lysed with 0.025% SDS and enriched with 0.05% BSA. Samples were quickly vortexed with glass beads to obtain a single cell suspension and serial dilutions in PBS were plated on 7H11 agar with supplements described above and allowed to incubate at 37°C for up

14 days 6 weeks. Enumeration of the bacteria for CFUs was performed using sample triplicates, each plated in duplicate per dilution for each time point. The MOI for each strain and condition was verified by serial dilution, plating and enumeration.

Washed MDMs were adhered to chromerge-cleaned glass coverslips in 24 well tissue culture plates, repleted with 20% autologous serum and allowed to incubate for an additional 7 days to obtain day 12 MDMs for infection, staining and enumeration. At day

12, wells containing MDM adhered coverslips were washed and repleted with RPMI and either 2.5% autologous serum or with RPMI 10 mM HEPES and 1 mg ml HSA (UniZLB

Bioplasma AG, Berne, Switzerland) which will be referred to as RHH medium. A subset of experimental sample wells were pre-incubated with mannan (Sigma-Aldrich) at

250mg/ml per well for 30 minutes at 37°C prior to infection to assess macrophage MR involvement since this has been shown to effectively inhibit binding to the MR in this assay (118). MDMs were incubated with each strain of bacteria at an MOI of 10:1 and then incubated shaking for 30 minutes and resting for an additional 90 minutes at 37°C,

5% CO2. After 2 hours, cells were washed three times with PBS and fixed with 2% paraformaldehyde. MDM-associated bacteria were stained with TB Auramine- rhodamine T primary stain following the manufacturer’s instructions (BD). The average

64 number of bacteria per cell was determined by counting 100 cells per coverslip in triplicate using an Olympus BX51 flourescent microscope using the UPlanFI 100x/1.30

Ph3 oil immersion objective and a FITC wide blue filter.

3.4.4: Western blotting and silver staining

Purified protein samples were normalized to 20ug/ lane by protein content determined by BCA assay and separated by SDS-PAGE using a 15% gel. Western blots were achieved using a primary monoclonal mouse IgG anti-6xHis antibody (Invitrogen cat# 37-2900) and a secondary donkey anti-mouse IgG (sc-2314) and visualized using the

ECL detection system (GE Healthcare RPN 2106) and X-ray film. Direct visualization in the gel was done using Coomassie Gel Code Blue (cat# 24592) or silver staining modified with periodic acid to preferentially detect sugars.

3.4.5: 2 Dimensional Thin Layer Chromatography

For analysis of phosphoinositol mannosides (PIMs), cells were harvested after 24 hours growth in the modified Sauton’s from Ms clones harboring pSMT3 containing the

Rv3256c ORF and its vector control counterpart in parallel. Lipid extraction based on protein content (200 μg/sample), was performed by washing the lysates first with chloroform:methanol (2:1), then with chloroform:methanol (1:2) and finally with chloroform:methanol:water (10:10:3). The resulting total crude lipid extract was spotted at the origin of 10cm x 10cm silica gel 60 plated, aluminum-backed thin layer sheets

(Merck, Darmstadt, Germany). Samples were allowed to migrate in 2 dimensions with

65 the first direction using chloroform:methanol:water (60:30:6) and in the second direction using chloroform:aceticacid:methanol:water (40:25:3:6) until the solvent front reaches the edge of the plate (13). Plates were dried and lightly sprayed with a solution of 10% concentrated sulfuric acid and α-napthol in absolute ethanol and heated at 110°C until lipids appeared.

3.4.5: Bioinformatics

To search nucleotide and protein sequence libraries, the Basic Local Alignment

Tool or BLAST supported by the National center for biotechnology information (NCBI) was used. Nucleotide and sequence alignments were done using Clustal Omega, maintained by the European Molecular Biology Laboratory that is affiliated with the

European Bioinformatics Institute (EMBL-EBI). Secondary protein structure predictions were made using Protein homology/analogy Recognition Engine version 2.0 or Phyre2 which is maintained as part of the Genome3D project by the Structural Bioinformatics group at the Imperial College of London. ImageJ program is supported by the NIH and was used for densitometry calculations.

3.5: Results

Rv3256c is situated in the center of a putative 3 gene operon downstream of manB (Rv3257c) and upstream of manA (Rv3255c). Over-expression of the manB gene in M. smegmatis showed an increase in LAM, lipomannan and PIMs compared to the WT counterpart and this translated to a better association with the macrophage (74). It has

66 also been demonstrated that the extent of terminal mannosylation plays a key role in pathogenesis as the bacteria will coat itself with this sugar to bind the mannose receptor and enter the macrophage in a way that promotes its pathogenicity (128). For all of these reasons, up-regulated expression of Rv3256c at early time points would lead us to hypothesize that the constitutive over-expression of this gene in M. smegmatis would result in increased mannosylated products.

Figure 3.1. Analysis of mannolipids by SDS-PAGE and densitometry. (A) vector control (VC) and the Rv3256c over-expressor were normalized to 20μg by protein content, diluted serially 2-fold in PBS and loaded into lanes of a 15% SDS-PAGE gel and visualized by a modified silver stain to preferentially show lipoglycans. (B) and (C) are the densitometric analysis of each band using ImageJ. Results are representative of 3 experiments

3.5.1: Analysis of mannolipids

The Rv3256c constitutively over-expressing strain of M. smegmatis (OE) and its vector control (VC) were grown in a modified Sauton’s media with a defined carbon source for 24 hours, shaking at 37°C. Whole cell lysates were obtained by a Fast-prep ultra-vortexer using Lysing matrix B columns and protein content was obtained by BCA assay. The resulting lysates starting at 20μg were diluted serially 2-fold and loaded on to an SDS-PAGE gel. After mobilization, the gels were stained with a modified silver stain to preferentially stain for sugars and the densitometry measurements were obtained by

ImageJ. Duplicate gels were stained with Coomassie Blue dye to ensure equal loading.

In contrast to the hypothesis, we observed a decrease in the overall quantity of ManLAM and LM in the OE compared to the VC as depicted in the stained gel (Figure 3.1A) and in the densitometric analysis (Figure 3.1B, 3.1C).

PIMs are low molecular weight glycolipids (~12kDA) that migrate anomalously as one large band by SDS-PAGE. To more accurately depict the individual species,

PIMs are analyzed by 2D TLC. Crude lipid extracts were generated using 200μg by protein content for each sample and the material was spotted at the origin of each plate.

As shown in, there was an overall decrease in PIM species for the OE compared to its VC

(Figure 3.2).

Figure 3.2. 2D-TLC analysis of Rv3256c over-expression (OE) compared to the vector control (VC). PIMs were extracted using a crude lipid extract protocol starting with 200μg by protein content for each sample. After extraction, dried samples were re- suspended in chloroform/methanol/water (10:10:3) and then spotted at the origin of each plate. Following separation, plates were sprayed with a solution of 10% concentrated sulfuric acid and α-napthol in absolute ethanol, and heated at 110°C until lipids appeared. Results are representative of 3 experiments.

3.5.2: Effect of Rv3256c over-expression on bacterial survival and association with macrophages

Based upon the unexpected experimental results showing a decrease in the overall quantity of mannosylated lipoglycans in the OE compared to the VC, we hypothesized that there would be a decrease in survival of bacteria with macrophages. Day 12 MDMs were infected with M.tb OE or VC in the presence of 2.5% autologous serum in RPMI or in RHH for different time intervals. Sample wells in triplicate were lysed, diluted, plated in duplicate and allowed to grow for 14-21 days at 37°C in order to assess survival. Once again in contrast to the hypothesis, the OE survival was significantly greater at the 2 hour

69 time point than the VC in serum conditions (Figure 3.3A). Even more striking was the survival outcome in RHH media lacking serum opsonins such as complement. The survival of OE was significantly greater at the 2 and 24 hour time points indicating a new non-serum dependent route of entry and acclimation into the macrophage (Figure 3.3B).

Figure 3.3. Survival assay for enumerating bacterial colony forming units (CFUs). Time points represent post-infection intervals. After cell lysis, triplicate samples were diluted and plated in duplicate on 7H11 + 10% OADC and bacteria allowed to grow for 14-21 days. (A) Bacterial CFU values from the test group in which bacteria were incubated with cells in 2.5% autologous serum in RPMI. (B) Bacterial CFU values from the test group in which bacteria were incubated with cells in RHH. P values were generated by the Student’s t-Test. The data are representative of 3 independent experiments with different donors, all exhibiting statistical significance at the 2 hour time point.

Based on the above result, we next asked whether the OE associated better with the macrophage at the 2 hour time point which was the prediction based on the CFU data.

We incubated bacteria with MDMs on glass coverslips in the presence and absence of serum or mannan (to block the mannose receptor) for 2h, washed and fixed the cells with paraformaldehyde, and stained and enumerated the number of bacteria per cell.

Consistent with the CFU data, the OE associated with MDMs significantly greater than the VC in all four conditions; Serum, Serum + mannan, RHH and RHH + mannan

(Figure 3.4 A, B, C, D). These data indicate that the increased association and survival of bacteria with macrophages in a serum-independent manner.

Figure 3.4. Cell Association assay. MDMs were seeded on glass coverslips and then incubated with M.tb OE or VC for 2 hours at an MOI (10:1) in RPMI in 4 different conditions; with 2.5% autologous serum, 2.5% autologous serum + mannan, with HEPES Buffer and Human Serum Albumin in RPMI (RHH), RHH + mannan. The cells were washed and fixed with 2% paraformaldehyde, stained with Auramine/Rhodamine T and enumerated in triplicate. (A) Represents the average number of bacteria/MDM in serum conditions. (B) Represents the average number of bacteria/MDM in RHH. (C,D) are representative microscopic images of Auramine/Rhodamine T stained infected macrophages for VC and OE in RHH conditions (magnification at 100x) ). Data are representative of 2 experiments.

3.5.3: Cloning of Rv3256c and protein production

We next sought to determine the specific function(s) of Rv3256c in order to correlate its function(s) with the macrophage data obtained. The first step was to purify the protein from both E. coli and M. smegmatis The ORF of Rv3256c was successfully cloned into the pJAM2 plasmid that is under the control of the inducible acetamide promoter and has a 6 X Histidine tag at the C-terminus and electroporated into in M. smegmatis. The bacteria were grown in modified Sauton’s adding kanamycin, 2% D- glucose as the sole carbon source and 2% acetamide for 24 hours, shaking at 37°C as described (138). After 24 hours, whole cell lysates were added to Ni-NTA columns, washed and eluted as described in the accompanying protocol. The protein quantity was measured using a BCA assay, separated by SDS-PAGE using a 15% SDS-PAGE gel and bands visualized by Coomassie stain and Western Blot against the His tag. The expected result based on the 363 amino acid sequence is a monomer at 36kDa as seen when expressed in E. coli (Figure 3.5C). The surprising result with M. smegmatis was a band migrating above 64kDa and below 80kDa visualized in the stained gel (Figure 3.5A) as well as in the WB (Figure 3.5B).

Figure 3.5. 15% SDS-PAGE with 20μg of purified protein in each lane in duplicate. (A) Coomassie-stained gel with a strong band migrating just above 64kDa. (B) Western Blot on samples purified from M. smegmatis using a 1° mouse IgG monoclonal anti-6X His tag antibody and a secondary HRP-tagged antibody, and visualized on X-ray film. A band is seen just above the 64kDa. (C) Western Blot on samples purified from E. coli using a 1°mouse IgG monoclonal anti-6X His tag antibody and a secondary HRP-tagged antibody, and visualized on X-ray film. A band is seen at 37kDa. Representative of 2 experiments.

3.6: Discussion

In the present study we have shown that the over-expression of M.tb Rv3256c in

M. smegmatis (OE) results in a decrease in the total amount of M.tb mannolipids while at the same time increasing the survival and association with macrophages compared to the

VC. Although our lab has shown that mannosylated lipoglycans are important for entry through the mannose receptor and modifying the phagosome, our results suggest that over-expression of Rv3256c in M. smegmatis (OE) leads to an altered bacterial surface coat, compared to VC strains, which results in a fundamentally new way for the bacteria

73 to enter and grow (at least early on) in its host cell niche, the macrophage (60). We propose here that in addition to the importance of terminal mannosylation, there are other bacterial cell wall molecules that can be recognized by the macrophage which are magnified with over-expression of Rv3256c. The idea that cell wall ultrastructure is static is a misnomer and it has been shown by many Gram – pathogens like Salmonella typhi, that LPS charge modification is a critical bacterial response to host environments allowing them to inhibit the binding of antimicrobial cations. These modifications are tightly regulated by a PmrA/PmrB two-component system and the quantity of transcripts dictates the extent of the modification, so it is tailored to the environment (17). This phenotypic cell wall changes compared to VC make it plausible that a stealth pathogen like M.tb could dynamically change the cell wall to accommodate its entry into host cells.

It also appears from both the transcriptional analysis of M.tb in macrophages

(chapter 2) and the survival assay, that the expression kinetics of Rv3256c in response to human macrophages is important. In chapter 2 Figure 4A, the transcriptional profile of

Rv3256c in rich media conditions shows little to no expression above the baseline for the time points analyzed and this was also true for the input bacteria prior to macrophage infection. It is only in context of early macrophage infection that we see up-regulation of the transcript which steadily declines as the infection progresses. This was our first indication that the regulation in expression of Rv3256c was important in the context of infection. Further support comes from a TraSH library in which Rv3256c was not essential in broth culture but was essential for establishing infection in mice (109). The

CFU survival assay and association analyses in Figures 3.3 and 3.4 demonstrate that in fact, the OE strain gets in to macrophages and acclimates better than the VC strain. These

74 results mirror the transcriptional data starting to decrease at 24 hours post infection, when the transcripts return to null, the significant difference in CFU is also lost.

Together, these results provide evidence that up-regulation of Rv3256c plays a role during entry and acclimation in the macrophage but plays a lesser role as the infection progresses. In fact if one looks carefully at the slope of the growth of the OE strain in the macrophages (Figure 3.3), it is flatter than the vector control strain suggesting that the new pathway for entry of the OE strain may not provide a survival advantage for the bacterium.

If the bioinformatics are correct and Rv3256c is in fact a glmS homologue, it would be the smallest one reported at 347 amino acids and the only one described in the entire order of Actinomycetales. The second GlmS in R. leguminosarum, NodM, induces nodule formation in plants (73). If Rv3256c is a glmS homologue, perhaps its role is in part to enable M.tb to regulate the host response by modifying its nascent phagosome, akin to the process of nodule formation in plants. One could further speculate that the production of more GlcNAc by the action of Rv3256c would provide a more rigid cell wall since GlcNAc is a precursor to PG and AG, elements of the M.tb cell wall skeleton.

A more rigid cell wall would be more resistant to host environmental insults and also could impact the location of molecules on the outer cell envelope, exposing them for interaction with macrophage receptors.

First attempts at functional assays using HPLC and LC/MS have not yielded definitive results and some the answers to all of the above propositions will become clearer as purification of a His-tagged Rv3256c is optimized and ready to use. The ORF of Rv3256c is 1041 bps and should produce a 36kDa protein but the Coomassie stained

75 gels and WBs of the OE strain in M. smegmatis reveal a band that migrates closer to 70 kDa (Figure 3.5B). In contrast, in E. coli a band at 36 kDa is seen (Figure 3.5C). If the band in M. smegmatis turns out to be Rv3256c, then these results suggest that the protein is either post-translationally modified in M. smegmatis or runs as a non-reducible dimer

(we have not been able to create a 36 kDa band with different approaches to reduce the protein). It is of interest, that the enzyme is reported to function as a dimer (88). Our next efforts will be to sequence the band via mass spectrometry. If the band is found to be

Rv3256c, we will initiate enzyme functional assays. Although the exact function(s) of

Rv3256c remains unclear at present, we have been able to show that its over-expression appears to be linked to entry and acclimation of M.tb in the host cell macrophage.

Chapter 4: Partial Characterization of ppm1 and Rv3258c in the Putative Mannose

Donor Biosynthetic Pathway in Mycobacterium

4.1: Introduction

Our transcriptional profile of the putative mannose donor pathway genes for M.tb in macrophages included a small solitary neighboring gene transcribed in the same direction, Rv3258c, and a non-neighboring but still critical gene, ppm1, responsible for the synthesis of the mannose donor molecule polyprenyl monophosphomannose (PPM).

Rv3258c was included, along with a few other genes of unknown function, with the idea that genes located together and transcribed in the same direction often have similar or related functions. PPM synthase, ppm1 or Rv2051c, is of particular interest in this pathway because it catalyzes the transfer of mannose from GDP-mannose to PPM which then serves as a donor molecule in the biosynthesis of high-order PIMs, LM and LAM.

Although these genes are of particular interest to us, difficulty in cloning and/or lack of a phenotype slowed our progress. In this chapter we present partial characterization of these two genes.

4.2: Materials and Methods

4.2.1: Mycobacterial strains, plasmids and growth media

M. tuberculosis H37Rv (ATCC 27294) and M. smegmatis (ATCC 700084) were grown on either Middlebrook 7H11 agar (Difco, Franklin Lakes, NJ) with 10% oleic acid, dextrose and catalase (OADC) enrichment for 9-14 days (for macrophage experiments) or in Middlebrook 7H9 broth (Difco) + 10% OADC and 0.05% Tween-80 and 20ug/ml kanamycin (138). Single cell suspensions used for infections were obtained as previously described (118).

Plasmids: pSMT3 is a lab stock shuttle plasmid with a MCS under the control of the constitutively expressing promoter, hsp60 (52). The regulated tetracycline inducible shuttle plasmid pSE110 (Addgene plasmid 17972) was obtained via the non-profit plasmid repository Addgene, deposited by Sabine Ehrt (45). The ORF of Rv3258c or ppm1 was amplified from M.tb H37Rv genomic DNA for insertion into the MCS of each plasmid construct. Plasmids were transformed into mycobacterial strains made competent using the standardized method described (55). As a vector control, M. smegmatis strains harboring pSMT3 and pSE100 plasmids with an empty MCS were used in parallel to all experiments.

4.2.2: Bioinformatics

To search nucleotide and protein sequence libraries, the Basic Local Alignment

Tool or BLAST supported by the National center for biotechnology information (NCBI)

78 was used. Nucleotide and sequence alignments were done using Clustal Omega, maintained by the European Molecular Biology Laboratory that is affiliated with the

University that helps predict secondary structure of proteins with transmembrane loops.

4.2.3: Isolation of human monocyte-derived macrophages (MDMs)

The Ohio State University Institutional Review Board. PBMCs were separated on a

Ficoll cushion and were cultured for 5 days in RPMI medium containing 20% autologous serum in Teflon wells at 37°C with 5% CO2. After 5 days of growth in the Teflon wells,

At day 12, the MDMs were used to analyze survival of mycobacteria by a CFU assay using a multiplicity of infection (MOI) of 5:1, cells were treated with DNAse I, lysed with 0.025% SDS and enriched with 0.05% BSA. Samples were quickly vortexed

79 with glass beads to obtain single cell suspensions and serial dilutions in PBS were plated on 7H11 agar with supplements described above and allowed to incubate at 37°C for up to 6 weeks. Enumeration of the bacteria for CFUs represent sample triplicates each plated in duplicate per dilution for each time point. MOI for each strain and condition was verified by serial dilution, plating and enumeration.

4.2.4: Coomassie staining

Samples were loaded into a 15% SDS-PAGE gel (20ug/lane by protein content determined by BCA assay) and separated. Direct visualization in the gel was done using Coomassie Gel Code Blue (cat# 24592).

4.2.5: Bacterial lysis, RNA isolation and Real-time PCR

Samples from broth grown bacterial cultures were taken at pre-determined growth phases and pelleted by centrifugation at 10,000 x g. Total RNA was extracted and purified by using a RNeasy Mini Column (Qiagen, Valencia, CA) and 0.1mm zirconia/silica beads (Biospec Products) coupled with DNase I (Qiagen, Valencia, CA) treatment. Isolation of bacteria from within infected macrophages was achieved by using a guanidinium thiocyanate (GTC)-based differential lysis solution as previously described (85). Bacterial RNA from within macrophages was processed by the procedure as described above. RNA was reverse transcribed to cDNA using 500 U of Superscript II reverse transcriptase with 10 mM dNTPs, 10 U RNase inhibitor, 0.1M DTT, and 3 μg of random hexamers (all from Invitrogen) for 120 min at 42°C, followed by inactivation

80 with 1N NaOH at 65°C for 10 min. Control reactions were performed in parallel without reverse transcriptase to verify the absence of DNA contamination. PCR was performed on the resulting cDNA using 300 mM custom-made primers with iQ SYBR Green Master mix (Bio-Rad) and 4% dimethyl sulfoxide (DMSO). All samples were run in triplicate

-∆∆C and analyzed using the 2 T method and expression was determined relative to the housekeeping gene rpoB (70).

4.2.6: Total carbohydrate analysis

Whole cell lysates were normalized by protein content via BCA assay and subjected to Proteinase K digestion and were converted to alditol acetates using scyllo- inositol as an internal standard and analyzed by gas chromatography (GC) as previously described (127, 77). GC of alditol acetates was performed using the ThermoQuest Trace

Gas Chromatograph 2000 connected to a GCQ/Polaris MS detector (ThermoQuest,

Austin, TX) at an initial temperature of 50°C for 1 min, increasing to 170°C at 30°C/min and finally to 270°C at 5°C/min.

4.3: Results

4.3.1: Cloning of Rv3258c

We elected to study Rv3258c based upon its genomic location, i.e., proximal to the mannose biosynthetic pathway manB operon as well as the results with the TraSH assay (109). The first experiments included cloning the ORF into a constitutively expressing plasmid, pSMT3, and transforming the completed construct into the fast- growing surrogate M. smegmatis. Successful transformants were screened using primers extending over the MCS, generating an approximately 1400bp fragment with empty vectors generating an approximately 900bp fragment (Figure 4.1A). An initial experiment using the Rv3258c over-expressor (58OE) strain and vector control (VC) strain, showed that the OE was slow growing relative to the VC. When efforts to reproduce these data were unsuccessful, it was discovered that the plasmid had been lost from the OE strains (Figure 4.1B).

Figure 4.1. 2% Agarose gels for visualization of PCR products used to verify M. smegmatis clones harboring pSMT3- Empty (VC) and pSMT3-58c (OE). Primers used to screen transformants were designed over the MCS of pSMT3. Empty plasmids generate an ~ 900bp fragment, positive transformants for pSMT3-58c generate an~ 1400bp fragment and Ms gDNA generates no fragment. M is for the molecular weight marker in both gels (1Kb plus DNA ladder, Invitrogen) (A) Confirmation of the initial transformants; Lane 1- VC, Lane 2- OE clone#14, Lane 3- OE clone#16, Lane 4-Ms gDNA and Lane 5- Empty purified plasmid. (B) Loss of pSMT3-58c in Ms clones; Lane 1- OE clone#3, Lane 2- OE clone#11, Lane 3- OE clone#14, Lane 4- OE clone#16, Lane 5- OE clone#20, Lane 6- VC, Lane 7- VC, Lane 8-Ms gDNA, Lane 9- purified plasmid pSMT3-58c, Lane 10- purified plasmid pSMT3.

Further attempts at introducing constitutively expressing plasmids of Rv3258c in

M. smegmatis resulted in the same loss of plasmid as before. The next step was to clone the Rv3258c ORF into a tightly regulated plasmid so as to control expression and stabilize the M. smegmatis recipient cells. pSE100 is a commercially available shuttle plasmid with low copy number in mycobacterium and regulated by the myc1tetO promoter. Anhydrotetracycline was used to bind to the operator site and induce

83 transcription in concentrations ranging from 0-100ng, and cell growth and viability were evaluated by measuring OD600 over time. The result based on OD600 revealed no significant difference in the growth after 24 hours of the OE compared to the VC although growth was diminished for both strains at the highest concentration of inducer used (Figure 4.2 A,B). However, when cells were lysed for RNA analysis, it was noted that there was almost 4 times as much total RNA in the VC compared to OE (Figure 4.2,

Table 4.1). Further analysis of the quantity of Rv3258c transcripts by RT-PCR revealed that there was no detectable mRNA for this gene after 60 min induction with 20ng anhydrotetracycline, thus implying degradation of the transcript.

Figure 4.2. Dose titration of anhydrotetracycline to induce transcription for plasmid pSE100 containing the ORF of Rv3258c. OD 600 readings were taken at 0, 6 and 24 hour time points to assess viability of Ms expressing Rv3258c and the VC after induction. (A) VC (B) Ms expressing Rv3258c.

Table 4.1. The total quantity of RNA extracted after 60 minutes hours induction with anhydrotetracycline from both strains.

RNA sample Quantity ng/ul RNA sample Quantity ng/ul

Ms-58c, S1 21.49 Ms-VC, S1 91.19

Ms-58c, S2 25.87 Ms-VC, S2 95.22

Ms-58c, S3 29.06 Ms-VC, S3 64.15

Contrary to over-expression in M. smegmatis, the ORF of Rv3258c cloned into pSMT3 was very stable when transformed and expressed into M.tb competent cells.

RNA was extracted from the clones of M.tb retaining the plasmid and elevated Rv3258c transcripts were detected compared the VC. Analysis of ManLAM, LM and PIMs by

SDS-PAGE between the two strains revealed no difference (not shown). Because the gene was up-regulated at 2 hours in the MDM transcriptional profile, we proposed that there would be an effect on bacterial growth in macrophages (Figure 4.3). However, we observed no phenotypic difference between M.tb 58OE and the VC.

Figure 4.3. CFU analysis of M.tb containing either pSMT3 empty (VC) or pSMT3-58c (OE) in macrophages. Twelve day MDMs were infected with the M.tb strains at an MOI of 5:1, lysed post infection at time points indicated and plated for growth of the bacteria on 7H11 + 10% OADC plates for enumeration. N= 2.

4.3.2: Polyprenol monophosphomannose synthase, ppm1

Growth curves were done with the WT Ms, Ms containing the empty pSMT3 plasmid (VC), and Ms with pSMT3-ppm1 (OE). OD600 was taken every 2 hours for 30 hours and showed no difference in growth rates among all strains (not shown). To determine the most appropriate growth time point in which to harvest bacteria in order to perform biochemical analyses of the cell wall, WT Ms RNA was extracted from lag, log, and stationary time points (time points were taken when the OD600 = 0.2, 0.6, 0.8, 1.2, 2.0 and 2.4). Real-Time PCR analysis showed that there was nearly baseline expression, relative to the housekeeping gene Ms rpoB, until the stationary phase where there was a large spike in gene expression.

Figure 4.4. SDS-PAGE gel (10%) analysis of Ms whole cell lysates after 30 min of heat shock at 45°C. Lanes 1 and 2: soluble fractions from the Ms Empty vector control. Lanes 3 and 4: insoluble fractions from the Empty vector control. Lanes 5 and 6: soluble fractions from Ms over-expressing M.tb ppm1. Lanes 7 and 8: insoluble fractions from the M.tb ppm1 -expressing Ms strain. M is the marker. Band of interest is denoted by an arrow at ~ 94 kDa. Results do not show any significant changes in protein bands for the M.tb ppm1- expressing Ms strain compared to Empty controls. Shown are duplicate samples from a representative experiment (n=3).

It has been shown that not only does the well-established M. bovis hsp60 promoter normally act in a constitutive manner, but is also inducible at 45°C (138). Ms cultures were heat-shocked for 15 and 30 min at 45°C and then allowed to recover at

37°C for 1 hour. Real-time PCR analysis showed an increase varying from 20 to 200

RCN after 30 minutes of heat shock. After heat-shock for 30 min, we analyzed soluble whole cell lysates as well as insoluble fractions for proteins by 10% SDS-PAGE followed by Coomassie Blue staining (Figure 4.4). We expected to see a protein band of 94 kDa, but the gel showed no differences between the pSMT3-ppm1 Ms strain and its empty vector control. Total carbohydrate analysis was also performed using the Ms strains and although the total quantity of sugars was different between strains (Figure 4.5A), the

87 ratios of mannose to arabinose, mannose to myoinositol, and mannose to glucose were not significantly different (Figure 4.5B) between the empty vector and the ppm1 expressing strain.

Figure 4.5. Total sugar analysis by GCMS of Ms VC (Empty) or Ms- M.tb ppm1 OE (ppm1). (A) Shows all sugars analyzed compared to a scyllo-inositol internal standard after Proteinase K digestion and Alditol acetate preparation. (B) Shows the ratio of mannose/arabinose, mannose/myoinositol and mannose/glucose.

4.4: Discussion

4.4.1: Rv3258c

Rv3258c became of interest to characterize based on our own transcriptional profile from within MDMs showing its up regulation at early time points post infection along with data demonstrating that inactivation of this gene resulted in slowed growth in vitro and its essentiality in establishing infection in mice (109). Bioinformatic analysis of the ORF for DNA and protein sequences gives little indication of function, so its function during the early course of M.tb infection remains unclear. What is interesting to note is that this gene is conserved at a >85% identity in pathogenic mycobacterial species indicating a more generalized pathogenic function rather than being specific to M.tb and its human host.

One of the most challenging but intriguing aspects of working with this gene was its instability in the fast-growing surrogate host, M. smegmatis. Dose titration of anhydrotetracycline (0-100ng) for both strains clearly indicates that there is a dose- dependent effect on the growth of the cells after 24 hours (Figure 4.2 A, B). What is particularly of interest is the difference between the VC and OE for 0 ng at 24 hours in that the VC was a denser cell population compared to the OE strain. This indicated that there may be some leakiness to the tightly regulated plasmid, pse100, and that even a miniscule amount of Rv3258c is toxic to M. smegmatis. Also noted in Table 4.1 is the total amount of RNA extracted between strains although the OD600 was similar and resulted in considerably lower RNA quantity in the OE compared to the VC suggesting

90 degradation. This is further indication that expression of Rv3258c in M. smegmatis is highly toxic, so it is not surprising that the plasmid was continuously lost from the clones.

It was also a surprise that the very same constitutively expressing plasmid transformed into M.tb had no discernible phenotype. We expected that it would based on our data regarding its up-regulation during early points of macrophage infection (chapter

2) and the work in mice that there may be a phenotype (109). CFU analysis demonstrated that there was no effect on growth, suggesting that this gene product may be regulated in M.tb and or other pathogenic mycobacterium species in ways that we do not understand yet, and that there is a lack of such regulation in a surrogate organism making its expression toxic to the organism. Further characterization of Rv3258c is warranted to further our understanding its role in pathogenesis and its function.

4.4.2: PPM synthase

To date, all data generated about M.tb PPM synthase have been done in the surrogate organisms M. smegmatis or C. glutamicum by deleting endogenous genes and extrapolating results to M.tb because of similarity in sequence or by replacing endogenous surrogate genes with M.tb ppm1. While these organisms have provided a great deal of information on the functions of M.tb ppm1, there are fundamental differences in the enzyme between these organisms that should be considered. First, the genomic organization is different in that Cg and Ms PPM synthases ares transcribed in an operon of two genes corresponding to two different domains of a single ORF in M.tb. It

91 has been modeled that Domain 2 remains cytosolic while Domain 1 serves as a membrane anchor (11).

It was recently shown that the Domain 1 is actually an active apolipoprotein N- acyltransferase suggesting that PPM synthase is a bi-functional enzyme but the work was done by replacing Ms ppm1/2 with Cg ppm1/2 and referring to them as M.tb homologues

(48). It was also recently shown that deletion of the Ms PPM operon is lethal and essential for glycosylation and acylation in Cg, supporting the bi-functional work in Ms

(84). This is in contrast to the data showing that M.tb ppm1 is not essential in vivo or in vitro in mice (109). These results provide evidence for some redundancy in the M.tb genome for this enzyme or a difference in functionality between the organisms. Clearly, more work needs to be done with the endogenous gene in M.tb strains.

Chapter 5: Synthesis and Future Directions

M.tb is one of the oldest and most successful pathogens in human history. It has long been theorized that M.tb co-evolved with humans and recent genetic analysis has shown this to be true with expansions in the genetic diversity of the organism coinciding with expansions in human populations. It is no surprise then that M.tb has developed a plethora of ways to subvert the host immune response and make a home out of what should be its undoing, its host cell niche, the phagosome. From an evolutionary standpoint then it is no accident that the bacterium coats itself in an abundant host- mimicking sugar, mannose, when it is most likely to encounter cells in the lung like alveolar macrophages that have an up regulated C-type lectins such as the MR which recognizes mannans.

The M.tb cell wall structurally unique features have not only been a problem for the host to combat but have also provided barriers to analysis by researchers. Until the

1970’s, M.tb was considered an intractable organism for molecular biology manipulation and although the genetic tools are more available, the majority of the structural, biochemical and virulence research continues to be done with models. Faster-growing mycobacterial species like Ms or avirulent and attenuated relatives like Cg and M. bovis

BCG provide TB researchers with preliminary answers to important questions regarding virulence but just as man is no mouse, these organisms are not M.tb.

The transcriptional profiling work detailed in Chapter 2 demonstrated that the presence of 99-100% identical genes in different organisms are not expressed or utilized in the same fashion. We have shown that in rich growth conditions, the manB operon

(manB-Rv3256c-manA) is a single 3 Kb transcript, yet in context of the human macrophage they are differentially regulated with the expression of Rv3256c up- regulated at early time points. Work on N. meningitidis showed a similar phenomenon in context of the human host for a small operon that was accomplished with small non- coding RNAs and alternate sigma factors (50). Recent genomic analysis has revealed that small non-coding RNAs are more common than previously thought in M.tb. This revelation along with the fact that M.tb has the choice of 13 mycobacterial sigma factors hints at the idea of transcriptional regulation in this respect.

Targeting ppm1 for analysis was based on our putative mannose donor biosynthetic pathway and the idea that it was the key step in moving mannose as a substrate across the plasma membrane. The gene coding for this enzyme is 2.7 Kb, large for a prokaryote, and proved difficult in cloning both for expression and deletion.

Transposon mutants showed its non-essentiality in mice (109). This fact along with our observation that there was no differential regulation in our transcriptional profiling

(Chapter 2) made characterizing this enzyme of less importance to us and the project was halted. It is very interesting to note that ALL work done characterizing this enzyme has used surrogates and to date, no characterization of the enzyme function or its role in

94 pathogenesis has been addressed in M.tb. Such research should prove fruitful in the future.

On the other hand, cloning the 492 bp ORF of Rv3258c was fairly straightforward and interest to characterize this gene was supported by our own transcriptional profile. The up-regulation of Rv3258c during early time points of infection coupled with the transposon mutagenesis work in mice (109) gave this project great appeal. However, bioinformatics gave little to no indication of the function of this gene, expression (regulated or constitutive) in Ms was toxic and over-expression in M.tb gave no positive results. Based on this, we concluded that the expression of this gene is highly regulated in the M.tb and toxic if not. Although this gene showed much promise, the genetic approaches attempted yielded no phenotype and was halted. Newer approaches being developed in the field should enable more positive results in the future and htus this gene requires more follow up.

The third and last gene that was up-regulated early during M.tb infection of macrophages was Rv3256c. Chapter 3 provides a detailed analysis of the over- expression of this gene showing alteration in the appearance of cell wall determinants, as well as changes in M.tb association with and survival in macrophages. The proposed function of the gene based solely on DNA homology is an isomerase but further secondary predictions suggest the possibility of additional glutaminase and synthase functions, like that of glucosamine-6-phosphate synthases or GlmS. This is of great interest as no such synthases have been described for mycobacterium or other actinomycetes. The biggest remaining question for this portion of the thesis is the true function of Rv3256c. To answer this question requires the expression and isolation of the

95 purified enzyme. The Rv3256c ORF was cloned into expression vectors with 6 X His tags and purified from E. coli and Ms. E. coli purified enzyme migrated under reduced and denaturing conditions as the predicted 36kDa monomer in SDS-PAGE gels. In contrast, the Ms purified enzyme migrated at ~ 70kDa, indicating a possible dimer. All attempts to reduce this “dimer” were unsuccessful, suggesting to a post-translational modification that leads to a non-reducible dimer (assuming the band is indeed Rv3256c).

In this respect, it remains intriguing that the E. coli GlmS was shown to be functional only as a dimer (88). Attempts to quantify the substrate and subsequent products of this putative enzyme have been very difficult thus far. Quantifying the potential substrate, fructose-6-phosphate, or the immediate product, glucosamine-6-phosphate, in whole cell lysates from over-expressing strains has not been definitive by radiolabeled TLCs, HPLC or LC/MS. Isolation of the purified enzyme in cell-free assays will allow us to narrow the experimental parameters and give some indication of function. This is the essential next step and work is in progress.

In conclusion, we have shown that M.tb pathogenesis relies on the expression and regulation of specific sets of bacterial genes during infection of macrophages. We have also shown that identical ORFs in closely related organisms are not expressed identically in human macrophages, pointing to the regulation of expression being both bacterial species and host specific. The likely toxic effect of expression of M.tb Rv3258c in Ms also supports this idea of specificity. As we develop more tools to evaluate the finer points of M.tb pathogenesis, we must also gravitate away from using surrogate organisms and animals to define this complex interaction between host and microbe. Understanding

96 the early events of infection should provide potential new drug of vaccine targets that either inhibit initial infection or limit dissemination during primary infection.

References

1. Afonso-Barroso A, Clark SO, Williams A, Rosa GT, Nóbrega C, Silva-Gomes

S, Vale-Costa S, Ummels R, Stoker N, Movahedzadeh F, van der Ley P, Sloots

A, Cot M, Appelmelk BJ, Puzo G, Nigou J, Geurtsen J, Appelberg R. 2012.

Lipoarabinomannan mannose caps do not affect mycobacterial virulence or the

induction of protective immunity in experimental animal models of infection and

have minimal impact on in vitro inflammatory responses. Cell. Microbiol. 15:660–

674.

2. Akhter Y, Ehebauer MT, Mukhopadhyay S, Hasnain SE. 2012. The PE/PPE

multigene family codes for virulence factors and is a possible source of

mycobacterial antigenic variation: perhaps more? Biochimie 94:110–6.

3. Alderwick LJ, Seidel M, Sahm H, Besra GS, Eggeling L. 2006. Identification of

a novel arabinofuranosyltransferase (AftA) involved in cell wall arabinan

biosynthesis in Mycobacterium tuberculosis. J. Biol. Chem. 281:15653–61.

4. Alibaud L, Pawelczyk J, Gannoun-Zaki L, Singh VK, Rombouts Y, Drancourt

M, Dziadek J, Guérardel Y, Kremer L. 2014. Increased Phagocytosis of

Mycobacterium marinum Mutants Defective in Lipooligosaccharide Production: A

STRUCTURE-ACTIVITY RELATIONSHIP STUDY. J. Biol. Chem. 289:215–28.

5. Anastasiou E, Mitchell PD. 2013. Palaeopathology and genes: investigating the

genetics of infectious diseases in excavated human skeletal remains and mummies

from past populations. Gene 528:33–40.

6. Andersen P, Askgaard D, Ljungqvist L, Bennedsen J, Heron I. 1991.

tuberculosis during growth . Proteins Released from Mycobacterium tuberculosis

during Growth 59.

7. Astarie-Dequeker C, Le Guyader L, Malaga W, Seaphanh F-K, Chalut C,

Lopez A, Guilhot C. 2009. Phthiocerol dimycocerosates of M. tuberculosis

participate in macrophage invasion by inducing changes in the organization of

plasma membrane lipids. PLoS Pathog. 5:e1000289.

8. Azad a. K. 1997. Gene Knockout Reveals a Novel Gene Cluster for the Synthesis

of a Class of Cell Wall Lipids Unique to Pathogenic Mycobacteria. J. Biol. Chem.

272:16741–16745.

9. Balhana R, Stoker N, Sikder M. 2010. Rapid construction of mycobacterial

mutagenesis vectors using ligation-independent cloning. J. Microbiol. Methods

83:34–41.

10. Batoni F, Maisetta F, Florio W, Freer G, Campa M, Senesi S. 1998. Analysis of

the Mycobacterium bovis hsp60 promoter activity in recombinant Mycobacterium

avium. FEMS Microbiol. Lett. 169:117–124.

11. Baulard AR, Gurcha SS, Engohang-Ndong J, Gouffi K, Locht C, Besra GS.

2003. In vivo interaction between the polyprenol phosphate mannose synthase

Ppm1 and the integral membrane protein Ppm2 from Mycobacterium smegmatis

revealed by a bacterial two-hybrid system. J. Biol. Chem. 278:2242–8.

12. Brennan P. 2003. Structure, function, and biogenesis of the cell wall of

Mycobacterium tuberculosis. Tuberculosis 83:91–97.

13. Brennan PJ, Heifets M, Ullom BP. 1982. lipid antigens as a means of identifying

nontuberculous Thin-Layer Chromatography of Lipid Antigens as a Means of

Identifying Nontuberculous Mycobacteria. J. Clin. Microbiol. 15:447.

14. Briken V, Porcelli S a, Besra GS, Kremer L. 2004. Mycobacterial

lipoarabinomannan and related lipoglycans: from biogenesis to modulation of the

immune response. Mol. Microbiol. 53:391–403.

15. Chatterjee D, Hunter SW, Mcneil M, Brennans J. 1992. Lipoarabinomannan:

Multiglycosylated Form of the Mycobacterial Mannosylphosphatidylinositols. J.

Biol. Chem. 267:6228–6233.

16. Chem JB, Brennan J. 1990. Evidence for the nature of the link between the

arabinogalactan and peptidoglycan of mycobacterial cell walls . Alerts : • When this

article is cited Evidence for the Nature of the Link between Peptidoglycan of

Mycobacterial Cell Walls * the Arabinogalact.

17. Chen HD, Groisman E a. 2013. The biology of the PmrA/PmrB two-component

system: the major regulator of lipopolysaccharide modifications. Annu. Rev.

Microbiol. 67:83–112.

18. Cole, Stewart T, Brosch, R., Parkhill, T., Garnier, C., Churcher, D. et al. 1998.

Deciphering the biology of Mycobacterium tuberculosis form the complete genome

sequence 396.

19. Comas I, Coscolla M, Luo T, Borrell S, Holt KE, Kato-Maeda M, Parkhill J,

Malla B, Berg S, Thwaites G, Yeboah-Manu D, Bothamley G, Mei J, Wei L,

100

Bentley S, Harris SR, Niemann S, Diel R, Aseffa A, Gao Q, Young D, Gagneux

S. 2013. Out-of-Africa migration and Neolithic coexpansion of Mycobacterium

tuberculosis with modern humans. Nat. Genet. 45:1176–82.

20. Crick DC, Brennan PJ, McNeil MR. 1994. Chapter 9, the cell wall of

Mycobacterium tuberculosis, p. 135–147. In Rom, W., Garay, SM, Bloom, BR

(eds.), Tuberculosis: Pathogenesis, Protection and Control, 2nd ed.

21. Cummings RD, McEver RP. 2009. C-type lectins, p. . In Varki A, Cummings RD,

Esko JD, Freeze HH, Stanley P, Bertozzi CR, Hart GW, Etzler ME (ed.), Essential

in Glycobiology, 2nd ed. Cold Spring Harbor, Cold Spring Harbor.

22. Cywes C, Hoppe HC, Ehlers MR. 1997. Nonopsonic binding of Mycobacterium

tuberculosis to complement receptor type 3 is mediated by capsular polysaccharides

and is strain dependent . Nonopsonic Binding of Mycobacterium tuberculosis to

Complement Receptor Type 3 Is Mediated by Capsular Polysac.

23. Daffe M, Brennan PJ, McNeil M. 1990. Predominant structural features of the cell

wall arabinogalactan of Mycobacterium tuberculosis as revealed through

characterization of oligoglycosyl alditol fragments by gas chromatography/mass

spectrometry and by 1H and 13C NMR analyses. J. Biol. Chem. 265:6734–43.

24. 2099. The mycobacterial cell envelope, 1st ed. ASM Press.

25. Daniel TM. 2006. The history of tuberculosis. Respir. Med. 100:1862–70.

26. Davies J, Davies D. 2010. Origins and evolution of antibiotic resistance. Microbiol.

Mol. Biol. Rev. 74:417–33.

101

27. Deng W, Xie J. 2012. Ins and outs of Mycobacterium tuberculosis PPE family in

pathogenesis and implications for novel measures against tuberculosis. J. Cell.

Biochem. 113:1087–95.

28. Driessen NN, Stoop EJM, Ummels R, Gurcha SS, Mishra AK, Larrouy-

Maumus G, Nigou J, Gilleron M, Puzo G, Maaskant JJ, Sparrius M, Besra GS,

Bitter W, Vandenbroucke-Grauls CMJE, Appelmelk BJ. 2010. Mycobacterium

marinum MMAR_2380, a predicted transmembrane acyltransferase, is essential for

the presence of the mannose cap on lipoarabinomannan. Microbiology 156:3492–

502.

29. Durand P, Golinelli-Pimpaneau B, Mouilleron S, Badet B, Badet-Denisot M-A.

2008. Highlights of glucosamine-6P synthase catalysis. Arch. Biochem. Biophys.

474:302–17.

30. Ganguly N, Siddiqui I, Sharma P. 2008. Role of M. tuberculosis RD-1 region

encoded secretory proteins in protective response and virulence. Tuberculosis

(Edinb). 88:510–7.

31. Garcia Pelayo MC, Uplekar S, Keniry A, Mendoza Lopez P, Garnier T, Nunez

Garcia J, Boschiroli L, Zhou X, Parkhill J, Smith N, Hewinson RG, Cole ST,

Gordon S V. 2009. A comprehensive survey of single nucleotide polymorphisms

(SNPs) across Mycobacterium bovis strains and M. bovis BCG vaccine strains

refines the genealogy and defines a minimal set of SNPs that separate virulent M.

bovis strains and M. bovis BCG strains. Infect. Immun. 77:2230–8.

32. Geiman DE, Raghunand TR, Agarwal N, Bishai WR. 2006. Differential gene

expression in response to exposure to antimycobacterial agents and other stress

102

conditions among seven Mycobacterium tuberculosis whiB-like genes. Antimicrob.

Agents Chemother. 50:2836–41.

33. Gibson KJC, Eggeling L, Maughan WN, Krumbach K, Gurcha SS, Nigou J,

Puzo G, Sahm H, Besra GS. 2003. Disruption of Cg-Ppm1, a polyprenyl

monophosphomannose synthase, and the generation of lipoglycan-less mutants in

Corynebacterium glutamicum. J. Biol. Chem. 278:40842–50.

34. Gilmore S a, Schelle MW, Holsclaw CM, Leigh CD, Jain M, Cox JS, Leary J a,

Bertozzi CR. 2012. Sulfolipid-1 biosynthesis restricts Mycobacterium tuberculosis

growth in human macrophages. ACS Chem. Biol. 7:863–70.

35. Gómez-Velasco A, Bach H, Rana AK, Cox LR, Bhatt A, Besra GS, Av-Gay Y.

2013. Disruption of the serine/threonine protein kinase H affects phthiocerol

dimycocerosates synthesis in Mycobacterium tuberculosis. Microbiology 159:726–

36.

36. Gordon S V, Bottai D, Simeone R, Stinear TP, Brosch R. 2009. Pathogenicity in

the tubercle bacillus: molecular and evolutionary determinants. Bioessays 31:378–

88.

37. Goren MB, Brokl O, Schaefer WB. 1974. Lipids of Putative Relevance to

Virulence in Mycobacterium tuberculosis : Correlation of Virulence with

Elaboration of Sulfatides Lipids of Putative Relevance to Virulence in

Mycobacterium tuberculosis : Correlation of Virulence with Elaboration of Sulfat.

38. Goude R, Amin AG, Chatterjee D, Parish T. 2008. The critical role of embC in

Mycobacterium tuberculosis. J. Bacteriol. 190:4335–41.

103

39. Green AM, Difazio R, Flynn JL. 2013. IFN-γ from CD4 T cells is essential for

host survival and enhances CD8 T cell function during Mycobacterium tuberculosis

infection. J. Immunol. 190:270–7.

40. Griffin JE, Gawronski JD, Dejesus M a, Ioerger TR, Akerley BJ, Sassetti CM.

2011. High-resolution phenotypic profiling defines genes essential for

mycobacterial growth and cholesterol catabolism. PLoS Pathog. 7:e1002251.

41. Grunewald S. 2002. Congenital Disorders of Glycosylation: A Review. Pediatr.

Res. 52:618–624.

42. Guerin ME, Kaur D, Somashekar BS, Gibbs S, Gest P, Chatterjee D, Brennan

PJ, Jackson M. 2009. New insights into the early steps of phosphatidylinositol

mannoside biosynthesis in mycobacteria: PimB’ is an essential enzyme of

Mycobacterium smegmatis. J. Biol. Chem. 284:25687–96.

43. Guerin ME, Korduláková J, Alzari PM, Brennan PJ, Jackson M. 2010.

Molecular basis of phosphatidyl-myo-inositol mannoside biosynthesis and

regulation in mycobacteria. J. Biol. Chem. 285:33577–83.

44. Guirado E, Schlesinger LS. 2013. Modeling the Mycobacterium tuberculosis

Granuloma - the Critical Battlefield in Host Immunity and Disease. Front.

Immunol. 4:98.

45. Guo X V, Monteleone M, Klotzsche M, Kamionka A, Hillen W, Braunstein M,

Ehrt S, Schnappinger D. 2007. Silencing Mycobacterium smegmatis by using

tetracycline repressors. J. Bacteriol. 189:4614–23.

104

46. Gupta A, Kaul A, Tsolaki AG, Kishore U, Bhakta S. 2012. Mycobacterium

tuberculosis: immune evasion, latency and reactivation. Immunobiology 217:363–

74.

47. Gupta R, Barkan D, Redelman-Sidi G, Shuman S, Glickman MS. 2011.

Mycobacteria exploit three genetically distinct DNA double-strand break repair

pathways. Mol. Microbiol. 79:316–30.

48. Gurcha SS, Baulard AR, Kremer L, Locht C, Moody DB, Muhlecker W,

Costello CE, R DCC, R PJB, Besra GS. 2002. Ppm1, an novel polyprenol

monophosphomannose synthase from Mycobcterium tuberculosis. J. Biochem.

450:441–450.

49. Guy MR, Illarionov P a, Gurcha SS, Dover LG, Gibson KJC, Smith PW,

Minnikin DE, Besra GS. 2004. Novel prenyl-linked benzophenone substrate

analogues of mycobacterial mannosyltransferases. Biochem. J. 382:905–12.

50. Hansen JM, Golchin SA, Domenech P, Boneca IG, Azad AK, Rajaram MVS,

Schlesinger LS, Divangahi M, Reed MB, Behr MA, Avenue C, Bact I, F- P,

Avenir G, Biology MI, Infection M, International M, Centre TB,

Correspondence C. 2013. cr ipt ce pt us cr ipt Ac ce pt us 1–30.

51. Hatfull G, Jacobs WR. 2000. Molecular genetics of Mycobacteria, 1st ed. ASM

Press, Washington, DC.

52. Herrmann JL, O’Gaora P, Gallagher a, Thole JE, Young DB. 1996. Bacterial

glycoproteins: a link between glycosylation and proteolytic cleavage of a 19 kDa

antigen from Mycobacterium tuberculosis. EMBO J. 15:3547–54.

105

53. Hunter RL, Armitige L, Jagannath C, Actor JK. 2009. TB research at UT-

Houston--a review of cord factor: new approaches to drugs, vaccines and the

pathogenesis of tuberculosis. Tuberculosis (Edinb). 89 Suppl 1:S18–25.

54. Hunter RL, Olsen MR, Jagannath C, Actor JK. 2006. Review : Multiple Roles

of Cord Factor in the Pathogenesis of Primary , Secondary , and Cavitary

Tuberculosis , Including a Revised Description of the Pathology of Secondary

Disease 36:371–386.

55. Jacobs WR, Cole, Stewart T, Brosch R, Gordon S, Eiglmeier K, Garnier T,

Tekaia F, Yeramian E, Cole ST, Hatfull G, Pashley C, Stoker N. 2000.

Molecular genetics of mycobacteria, 1st ed. ASM Press, Washington, DC.

56. Jacobs WR, Tuckman M, Bloom BR. 1987. Induction of foreign DNA into

mycobacteria using a shuttle phasmid. Nature 327:532–535.

57. Jankute M, Grover S, Rana AK, Besra GS. 2012. potential as drug targets 129–

147.

58. Józefowski S, Sobota A, Kwiatkowska K. 2008. How Mycobacterium

tuberculosis subverts host immune responses. Bioessays 30:943–54.

59. Jung S, Yang C, Lee J, Shin A, Jung S, Son JW, Harding C V, Kim H, Park J,

Paik T, Song C, Jo E. 2006. The Mycobacterial 38-Kilodalton Glycolipoprotein

Antigen Activates the Mitogen-Activated Protein Kinase Pathway and Release of

Proinflammatory Cytokines through Toll-Like Receptors 2 and 4 in Human

Monocytes The Mycobacterial 38-Kilodalton Glycolipoprotei.

60. Kang PB, Azad AK, Torrelles JB, Kaufman TM, Beharka A, Tibesar E,

DesJardin LE, Schlesinger LS. 2005. The human macrophage mannose receptor

106

directs Mycobacterium tuberculosis lipoarabinomannan-mediated phagosome

biogenesis. J. Exp. Med. 202:987–99.

61. Kaur D, Obregón-Henao A, Pham H, Chatterjee D, Brennan PJ, Jackson M.

2008. Lipoarabinomannan of Mycobacterium: mannose capping by a

multifunctional terminal mannosyltransferase. Proc. Natl. Acad. Sci. U. S. A.

105:17973–7.

62. Konar M, Alam MS, Arora C, Agrawal P. 2012. WhiB2/Rv3260c, a cell

division-associated protein of Mycobacterium tuberculosis H37Rv, has properties

of a chaperone. FEBS J. 279:2781–92.

63. Korduláková J, Gilleron M, Mikusova K, Puzo G, Brennan PJ, Gicquel B,

Jackson M. 2002. Definition of the first mannosylation step in phosphatidylinositol

mannoside synthesis. PimA is essential for growth of mycobacteria. J. Biol. Chem.

277:31335–44.

64. Kremer L, Gurcha SS, Bifani P, Hitchen PG, Baulard A, Morris HR, Dell A,

Brennan PJ, Besra GS. 2002. involved in phosphatidylinositol trimannoside

biosynthesis in Mycobacterium tuberculosis 447:437–447.

65. Kuo C-J, Bell H, Hsieh C-L, Ptak CP, Chang Y-F. 2012. Novel mycobacteria

antigen 85 complex binding motif on fibronectin. J. Biol. Chem. 287:1892–902.

66. Labigne S. 1997. The Helicobacter pylori ureC gene codes for a The Helicobacter

pylori ureC Gene Codes for a Phosphoglucosamine Mutase 179.

67. Lemassu A, Lanéelle MA, Bardou F, Silve G, Lemassu A, Lane M. 1996.

Identification of the surface-exposed lipids on the cell envelopes of Mycobacterium

tuberculosis and other mycobacterial species . Identification of the Surface-Exposed

107

Lipids on the Cell Envelopes of Mycobacterium tuberculosis and Other

Mycobacterial Spe.

68. Lemassu a, Ortalo-Magné a, Bardou F, Silve G, Laneélle M a, Daffé M. 1996.

Extracellular and surface-exposed polysaccharides of non-tuberculous

mycobacteria. Microbiology 142 Pt 6:1513–20.

69. Li S, Kang J, Yu W, Zhou Y, Zhang W, Xin Y, Ma Y. 2012. Identification of M.

tuberculosis Rv3441c and M. smegmatis MSMEG_1556 and essentiality of M.

smegmatis MSMEG_1556. PLoS One 7:e42769.

70. Livak KJ, Schmittgen TD. 2001. Analysis of relative gene expression data using

real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25:402–

71. Major R. 1945. Classical descriptions of disease, 3rd ed. Springfield.

72. Makinoshima H, Glickman MS. 2005. Regulation of Mycobacterium tuberculosis

cell envelope composition and virulence by intramembrane proteolysis. Nature

436:406–9.

73. Marie, C, Barny, M.A. and Downie J. 1992. Rhizobium leguminosarum has two

glucosamine syntheses , GImS and NodM , required for nodulation and

development of nitrogen-fixing nodules 6:843–851.

74. McCarthy TR, Torrelles JB, MacFarlane AS, Katawczik M, Kutzbach B,

Desjardin LE, Clegg S, Goldberg JB, Schlesinger LS. 2005. Overexpression of

Mycobacterium tuberculosis manB, a phosphomannomutase that increases

phosphatidylinositol mannoside biosynthesis in Mycobacterium smegmatis and

mycobacterial association with human macrophages. Mol. Microbiol. 58:774–90.

108

75. McGrath M, Gey van Pittius NC, van Helden PD, Warren RM, Warner DF.

2014. Mutation rate and the emergence of drug resistance in Mycobacterium

tuberculosis. J. Antimicrob. Chemother. 69:292–302.

76. McGuire AM, Weiner B, Park ST, Wapinski I, Raman S, Dolganov G,

Peterson M, Riley R, Zucker J, Abeel T, White J, Sisk P, Stolte C, Koehrsen

M, Yamamoto RT, Iacobelli-Martinez M, Kidd MJ, Maer AM, Schoolnik GK,

Regev A, Galagan J. 2012. Comparative analysis of Mycobacterium and related

Actinomycetes yields insight into the evolution of Mycobacterium tuberculosis

pathogenesis. BMC Genomics 13:120.

77. Mcneil BM, Chatterjee D, Hunter SWU, Brennan PJ. 1989. Glycolipids of

Mycobacterium. Methods Enzymol. 179:215–242.

78. Milewski S. 2002. Glucosamine-6-phosphate synthase--the multi-facets enzyme.

Biochim. Biophys. Acta 1597:173–92.

79. Mishra AK, Alderwick LJ, Rittmann D, Tatituri RV V, Nigou J, Gilleron M,

Eggeling L, Besra GS. 2007. Identification of an alpha(1-->6)

mannopyranosyltransferase (MptA), involved in Corynebacterium glutamicum

lipomanann biosynthesis, and identification of its orthologue in Mycobacterium

tuberculosis. Mol. Microbiol. 65:1503–17.

80. Mishra AK, Alderwick LJ, Rittmann D, Wang C, Bhatt A, Jacobs WR,

Takayama K, Eggeling L, Besra GS. 2008. Identification of a novel alpha(1-->6)

mannopyranosyltransferase MptB from Corynebacterium glutamicum by deletion

of a conserved gene, NCgl1505, affords a lipomannan- and lipoarabinomannan-

deficient mutant. Mol. Microbiol. 68:1595–613.

109

81. Mishra AK, Driessen NN, Appelmelk BJ, Besra GS. 2011. Lipoarabinomannan

and related glycoconjugates: structure, biogenesis and role in Mycobacterium

tuberculosis physiology and host-pathogen interaction. FEMS Microbiol. Rev.

35:1126–57.

82. Mishra AK, Krumbach K, Rittmann D, Appelmelk B, Pathak V, Pathak AK,

Nigou J, Geurtsen J, Eggeling L, Besra GS. 2011. Lipoarabinomannan

biosynthesis in Corynebacterineae: the interplay of two α(1→2)-

mannopyranosyltransferases MptC and MptD in mannan branching. Mol.

Microbiol. 80:1241–59.

83. Mishra AK, Krumbach K, Rittmann D, Batt SM, Lee OY-C, De S, Frunzke J,

Besra GS, Eggeling L. 2012. Deletion of manC in Corynebacterium glutamicum

results in a phospho-myo-inositol mannoside- and lipoglycan-deficient mutant.

Microbiology 158:1908–17.

84. Mohiman N, Argentini M, Batt SM, Cornu D, Masi M, Eggeling L, Besra G,

Bayan N. 2012. The ppm operon is essential for acylation and glycosylation of

lipoproteins in Corynebacterium glutamicum. PLoS One 7:e46225.

85. Monahan IM, Mangan JA, Buthcer PD. 2001. Mycobacterium tuberculosis

protocols, p. 31–42. In Parish, T, Stoker, N (eds.), , 3rd ed. Humana Press, Totowa.

86. Morita YS, Fukuda T, Sena CBC, Yamaryo-Botte Y, McConville MJ,

Kinoshita T. 2011. Inositol lipid metabolism in mycobacteria: biosynthesis and

regulatory mechanisms. Biochim. Biophys. Acta 1810:630–41.

87. Morita YS, Sena CBC, Waller RF, Kurokawa K, Sernee MF, Nakatani F,

Haites RE, Billman-Jacobe H, McConville MJ, Maeda Y, Kinoshita T. 2006.

110

PimE is a polyprenol-phosphate-mannose-dependent mannosyltransferase that

transfers the fifth mannose of phosphatidylinositol mannoside in mycobacteria. J.

Biol. Chem. 281:25143–55.

88. Mouilleron S, Badet-Denisot M-A, Pecqueur L, Madiona K, Assrir N, Badet B,

Golinelli-Pimpaneau B. 2012. Structural basis for morpheein-type allosteric

regulation of Escherichia coli glucosamine-6-phosphate synthase: equilibrium

between inactive hexamer and active dimer. J. Biol. Chem. 287:34533–46.

89. Mukhopadhyay S, Balaji KN. 2011. The PE and PPE proteins of Mycobacterium

tuberculosis. Tuberculosis (Edinb). 91:441–7.

90. Niederweis M. 2003. Mycobacterial porins - new channel proteins in unique outer

membranes. Mol. Microbiol. 49:1167–1177.

91. Nigou J, Gilleron M, Puzo G. 2003. Lipoarabinomannans: from structure to

biosynthesis. Biochimie 85:153–66.

92. Ning B, Elbein a D. 1999. Purification and properties of mycobacterial GDP-

mannose pyrophosphorylase. Arch. Biochem. Biophys. 362:339–45.

93. O’Garra A, Redford PS, McNab FW, Bloom CI, Wilkinson RJ, Berry MPR.

2013. The immune response in tuberculosis.Annual review of immunology.

94. Olakanmi O, Britigan BE, Schlesinger LS. 2000. Gallium disrupts iron

metabolism of mycobacteria residing within human macrophages. Infect. Immun.

68:5619–27.

95. Ortalo-Magné a, Dupont M a, Lemassu a, Andersen a B, Gounon P, Daffé M.

1995. Molecular composition of the outermost capsular material of the tubercle

bacillus. Microbiology 141 ( Pt 7:1609–20.

111

96. Ott CM, Lingappa VR. 2002. Integral membrane protein biosynthesis: why

topology is hard to predict. J. Cell Sci. 115:2003–9.

97. Pandey AK, Raman S, Proff R, Joshi S, Kang C-M, Rubin EJ, Husson R,

Sassetti CM. 2009. Nitrile-inducible gene expression in mycobacteria.

Tuberculosis 89:12–16.

98. Papa F, Cruaud P, Luquin M, Thorel MF, Goh KS, David HL. 1993. Isolation

and characterization of serologically reactive lipooligosaccharides from

Mycobacterium tuberculosis. Res. Microbiol. 144:91–9.

99. Patterson JH, Waller RF, Jeevarajah D, Billman-Jacobe H, McConville MJ.

2003. Mannose metabolism is required for mycobacterial growth. Biochem. J.

372:77–86.

100. Perera AS, Wang H, Basel MT, Pokhrel MR, Gamage PS, Kalita M, Wendel S,

Sears B, Welideniya D, Liu Y, Turro C, Troyer DL, Bossmann SH. 2013.

Channel blocking of MspA revisited. Langmuir 29:308–15.

101. Petronis A, Kennedy JL, Jl K, Genetic PA, Nerlich AG, Haas CJ, Zink A,

Szeimies U. 1997. Molecular evidence for tuberculosis in an ancient Egyptian

mummy Contribution of job control to social gradient in 350:1997.

102. Philips J a, Ernst JD. 2012. Tuberculosis pathogenesis and immunity. Annu. Rev.

Pathol. 7:353–84.

103. Prinzis S, Chatterjee D, Brennan PJ. 1993. Structure and antigenicity of

lipoarabinomannan from Mycobacterium bovis BCG. J. Gen. Microbiol. 139:2649–

58.

112

104. Pym AS, Brodin P, Brosch R, Huerre M, Cole ST. 2002. Loss of RD1

contributed to the attenuation of the live tuberculosis vaccines Mycobacterium

bovis BCG and Mycobacterium microti. Mol. Microbiol. 46:709–17.

105. Raghunand TR, Bishai WR. 2006. Mapping essential domains of Mycobacterium

smegmatis WhmD: insights into WhiB structure and function. J. Bacteriol.

188:6966–76.

106. Rana AK, Singh A, Gurcha SS, Cox LR, Bhatt A, Besra GS. 2012. Ppm1-

encoded polyprenyl monophosphomannose synthase activity is essential for

lipoglycan synthesis and survival in mycobacteria. PLoS One 7:e48211.

107. Raymond JB, Mahapatra S, Crick DC, Pavelka MS. 2005. Identification of the

namH gene, encoding the hydroxylase responsible for the N-glycolylation of the

mycobacterial peptidoglycan. J. Biol. Chem. 280:326–33.

108. Ren H, Dover LG, Islam ST, Alexander DC, Chen JM, Besra GS, Liu J. 2007.

Identification of the lipooligosaccharide biosynthetic gene cluster from

Mycobacterium marinum. Mol. Microbiol. 63:1345–59.

109. Rengarajan J, Bloom BR, Rubin EJ. 2005. Genome-wide requirements for

Mycobacterium tuberculosis adaptation and survival in macrophages. Proc. Natl.

Acad. Sci. U. S. A. 102:8327–32.

110. Renshaw PS, Lightbody KL, Veverka V, Muskett FW, Kelly G, Frenkiel T a,

Gordon S V, Hewinson RG, Burke B, Norman J, Williamson R a, Carr MD.

2005. Structure and function of the complex formed by the tuberculosis virulence

factors CFP-10 and ESAT-6. EMBO J. 24:2491–8.

113

111. Rohde KH, Abramovitch RB, Russell DG. 2007. Mycobacterium tuberculosis

invasion of macrophages: linking bacterial gene expression to environmental cues.

Cell Host Microbe 2:352–64.

112. Rook G a W, Dheda K, Zumla A. 2005. Immune responses to tuberculosis in

developing countries: implications for new vaccines. Nat. Rev. Immunol. 5:661–7.

113. Rousseau C, Turner OC, Rush E, Bordat Y, Sirakova TD, Kolattukudy PE,

Orme IM, Gicquel B, Jackson M, Ritter S. 2003. Sulfolipid Deficiency Does Not

Affect the Virulence of Mycobacterium tuberculosis H37Rv in Mice and Guinea

Pigs Sulfolipid Deficiency Does Not Affect the Virulence of Mycobacterium

tuberculosis H37Rv in Mice and Guinea Pigs.

114. Rousseau C, Winter N, Pivert E, Bordat Y, Neyrolles O, Avé P, Huerre M,

Gicquel B, Jackson M. 2004. Production of phthiocerol dimycocerosates protects

Mycobacterium tuberculosis from the cidal activity of reactive nitrogen

intermediates produced by macrophages and modulates the early immune response

to infection 6:277–287.

115. Rybniker J, Nowag A, van Gumpel E, Nissen N, Robinson N, Plum G,

Hartmann P. 2010. Insights into the function of the WhiB-like protein of

mycobacteriophage TM4--a transcriptional inhibitor of WhiB2. Mol. Microbiol.

77:642–57.

116. Sassetti CM, Boyd DH, Rubin EJ. 2001. Comprehensive identification of

conditionally essential genes in mycobacteria. Proc. Natl. Acad. Sci. U. S. A.

98:12712–7.

114

117. Scherman H, Kaur D, Pham H, Skovierová H, Jackson M, Brennan PJ. 2009.

Identification of a polyprenylphosphomannosyl synthase involved in the synthesis

of mycobacterial mannosides. J. Bacteriol. 191:6769–72.

118. Schlesinger LS. 1993. Macrophage Phagocytosis of Virulent but Not Attenuated

Strains of 150:2920–2930.

119. Schlesinger LS, Bellinger-Kawahara CG, Payne NR, Horwitz M a. 1990.

Phagocytosis of Mycobacterium tuberculosis is mediated by human monocyte

complement receptors and complement component C3. J. Immunol. 144:2771–80.

120. Seeliger JC, Holsclaw CM, Schelle MW, Botyanszki Z, Gilmore S a, Tully SE,

Niederweis M, Cravatt BF, Leary J a, Bertozzi CR. 2012. Elucidation and

chemical modulation of sulfolipid-1 biosynthesis in Mycobacterium tuberculosis. J.

Biol. Chem. 287:7990–8000.

121. Seidel M, Alderwick LJ, Birch HL, Sahm H, Eggeling L, Besra GS. 2007.

Identification of a novel arabinofuranosyltransferase AftB involved in a terminal

step of cell wall arabinan biosynthesis in Corynebacterianeae, such as

Corynebacterium glutamicum and Mycobacterium tuberculosis. J. Biol. Chem.

282:14729–40.

122. Singh A, Crossman DK, Mai D, Guidry L, Voskuil MI, Renfrow MB, Steyn

AJC. 2009. Mycobacterium tuberculosis WhiB3 maintains redox homeostasis by

regulating virulence lipid anabolism to modulate macrophage response. PLoS

Pathog. 5:e1000545.

123. Singh S, Saraiva L, Elkington PTG, Friedland JS. 2014. Regulation of matrix

metalloproteinase-1, -3, and -9 in Mycobacterium tuberculosis-dependent

115

respiratory networks by the rapamycin-sensitive PI3K/p70S6K cascade. FASEB J.

28:85–93.

124. Smith I. 2003. Mycobacterium tuberculosis Pathogenesis and Molecular

Determinants of Virulence Mycobacterium tuberculosis Pathogenesis and

Molecular Determinants of Virulence 16.

125. Snapper SB, Lugosi L, Jekkel a, Melton RE, Kieser T, Bloom BR, Jacobs WR.

1988. Lysogeny and transformation in mycobacteria: stable expression of foreign

genes. Proc. Natl. Acad. Sci. U. S. A. 85:6987–91.

126. Sonnenberg MG, Belisle JT. 1997. Definition of Mycobacterium tuberculosis

culture filtrate proteins by two-dimensional polyacrylamide gel electrophoresis , N-

terminal amino acid sequencing , and electrospray mass Definition of

Mycobacterium tuberculosis Culture Filtrate Proteins by Two-D.

127. Torrelles JB, DesJardin LE, MacNeil J, Kaufman TM, Kutzbach B, Knaup R,

McCarthy TR, Gurcha SS, Besra GS, Clegg S, Schlesinger LS. 2009.

Inactivation of Mycobacterium tuberculosis mannosyltransferase pimB reduces the

cell wall lipoarabinomannan and lipomannan content and increases the rate of

bacterial-induced human macrophage cell death. Glycobiology 19:743–55.

128. Torrelles JB, Schlesinger LS. 2010. Diversity in Mycobacterium tuberculosis

mannosylated cell wall determinants impacts adaptation to the host. Tuberculosis

(Edinb). 90:84–93.

129. Torrelles JB, Sieling P a, Zhang N, Keen M a, McNeil MR, Belisle JT, Modlin

RL, Brennan PJ, Chatterjee D. 2012. Isolation of a distinct Mycobacterium

116

tuberculosis mannose-capped lipoarabinomannan isoform responsible for

recognition by CD1b-restricted T cells. Glycobiology 22:1118–27.

130. Torrelles JB, Sieling P a, Zhang N, Keen M a, McNeil MR, Belisle JT, Modlin

RL, Brennan PJ, Chatterjee D. 2012. Isolation of a distinct Mycobacterium

tuberculosis mannose-capped lipoarabinomannan isoform responsible for

recognition by CD1b-restricted T cells. Glycobiology 22:1118–27.

131. Triccas, J.A., Parish, T., Britton, W. J. and Gicquel B. 1998. pJAM2 purification

paper.pdf. FEMS Microbiology Letters.

132. Trunz BB, Fine P, Dye C. 2006. Effect of BCG vaccination on childhood

tuberculous meningitis and miliary tuberculosis worldwide: a meta-analysis and

assessment of cost-effectiveness. Lancet 367:1173–80.

133. Usha V, Lloyd AJ, Lovering AL, Besra GS. 2012. Structure and function of

Mycobacterium tuberculosis meso-diaminopimelic acid (DAP) biosynthetic

enzymes. FEMS Microbiol. Lett. 330:10–6.

134. van der Woude AD, Sarkar D, Bhatt A, Sparrius M, Raadsen S a, Boon L,

Geurtsen J, van der Sar AM, Luirink J, Houben ENG, Besra GS, Bitter W.

2012. Unexpected link between lipooligosaccharide biosynthesis and surface

protein release in Mycobacterium marinum. J. Biol. Chem. 287:20417–29.

135. Kessel JC Van, Hatfull GF. 2007. Recombineering in Mycobacterium

tuberculosis. Nat. Methods 4:147–152.

136. Wallis RS, Kim P, Cole S, Hanna D, Andrade BB, Maeurer M, Schito M,

Zumla A. 2013. Tuberculosis biomarkers discovery: developments, needs, and

challenges. Lancet Infect. Dis. 13:362–72.

117

137. Warrier T, Tropis M, Werngren J, Diehl A, Gengenbacher M, Schlegel B,

Schade M, Oschkinat H, Daffe M, Hoffner S, Eddine AN, Kaufmann SHE.

2012. Antigen 85C inhibition restricts Mycobacterium tuberculosis growth through

disruption of cord factor biosynthesis. Antimicrob. Agents Chemother. 56:1735–43.

138. Wayne LG. 1994. Tuberculosis: Pathogenesis, protection and control, p. 73–83. In

Bloom, BR (ed.), First. ASM Press.

139. Wolucka B a, de Hoffmann E. 1998. Isolation and characterization of the major

form of polyprenyl-phospho-mannose from Mycobacterium smegmatis.

Glycobiology 8:955–62.

140. Yang Y, Shi F, Tao G, Wang X. 2012. Purification and structure analysis of

mycolic acids in Corynebacterium glutamicum. J. Microbiol. 50:235–40.

141. Ye RW, Zielinski N a, Chakrabarty a M. 1994. Purification and characterization

of phosphomannomutase/phosphoglucomutase from Pseudomonas aeruginosa

involved in biosynthesis of both alginate and lipopolysaccharide. J. Bacteriol.

176:4851–7.

142. Zhang W, Jones VC, Scherman MS, Mahapatra S, Crick D, Bhamidi S, Xin Y,

McNeil MR, Ma Y. 2008. Expression, essentiality, and a microtiter plate assay for

mycobacterial GlmU, the bifunctional glucosamine-1-phosphate acetyltransferase

and N-acetylglucosamine-1-phosphate uridyltransferase. Int. J. Biochem. Cell Biol.

40:2560–71.

143. Zumla A, Nahid P, Cole ST. 2013. Advances in the development of new

tuberculosis drugs and treatment regimens. Nat. Rev. Drug Discov. 12:388–404.

118