UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of the ESX-1 (Snm) secretion system in mycobacteria Lalani, Shifana 2009

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2009_spring_lalani_shifana.pdf [ 2.78MB ]
Metadata
JSON: 24-1.0067093.json
JSON-LD: 24-1.0067093-ld.json
RDF/XML (Pretty): 24-1.0067093-rdf.xml
RDF/JSON: 24-1.0067093-rdf.json
Turtle: 24-1.0067093-turtle.txt
N-Triples: 24-1.0067093-rdf-ntriples.txt
Original Record: 24-1.0067093-source.json
Full Text
24-1.0067093-fulltext.txt
Citation
24-1.0067093.ris

Full Text

CHARACTERIZATION OF THE ESX-1 (SNM) SECRETION SYSTEM IN MYCOBACTERIA  by Shifana Lalani  B. Sc., University of British Columbia, 2006  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  The Faculty of Graduate Studies (Biochemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2009  © Shifana Lalani, 2009  ABSTRACT Tuberculosis is a common and deadly disease caused by the Gram-positive bacteria Mycobacterium tuberculosis. Due to the unique nature of the extracytoplasmic localization of proteins in mycobacteria, specialized secretion systems are essential. Recent studies have identified a novel secretion pathway termed the ESX-1 (or Snm) secretion system. This system is known to secrete two virulence factors, ESAT-6 and CFP-10. The ESX-1 system is comprised of at least three core-components, Rv3870, Rv3871 and Rv3877. Rv3877 is an integral membrane protein, and possibly creates a translocation channel within the membrane. Rv3870 and Rv3871 have sequence homology to AAA-ATPases. Together, Rv3870, Rv3871 and Rv3877 may work to propel ESAT-6 and CFP-10 across the bacterial membrane. These predictions are primarily based on the primary sequences of the proteins, and no other biochemical information exists. Thus, it is important to determine how the secretion system components interact with one another. To investigate these interactions, homologues of Rv3870 (Snm1), Rv3871 (Snm2), and Rv3877 (Snm4) were cloned from the Mycobacterium smegmatis genome and expressed in bacterial expression systems. All three components were successfully expressed, and Snm2 was purified in the presence of adenosine tri-phosphate. Investigation of purified Snm2 revealed the presence of a dimer complex in solution. Snm2 was tested for binding interaction with the CFP10 and EAST-6 complex using gel filtration chromatography, native gel analysis, radiolabeling techniques, and crosslinking experiments. Through the assistance of a crosslink, CFP-10 was shown to interact with Snm2.  ii  This is the first biochemical characterization of the ESX-1 components. The data from these results will provide additional information on the specialized secretion systems of mycobacteria.  iii  TABLE OF CONTENTS ABSTRACT ........................................................................................................................ ii TABLE OF CONTENTS ....................................................................................................... iv LIST OF TABLES ............................................................................................................... vi LIST OF FIGURES............................................................................................................. vii ACKNOWLEDGEMENTS ................................................................................................... ix DEDICATION .................................................................................................................... x CHAPTER ONE INTRODUCTION ......................................................................................... 1 1.1 Preface............................................................................................................................. 1 1.2 The BCG vaccine and the Region of Difference RD1 ....................................................... 1 1.3 Dissection of RD1 ............................................................................................................ 4 1.4 Protein Secretion in Mycobacteria................................................................................ 10 1.5 ESAT-6/CFP-10 ............................................................................................................... 12 1.6 Conserved Components of the ESX-1 Secretion System............................................... 14 1.7 Protein-Protein Interactions between ESX-1 Secretion System Components.............. 17 1.8 C-terminal Signal Sequence of CFP-10 .......................................................................... 21 1.9 Thesis Investigation ....................................................................................................... 27 CHAPTER TWO MATERIALS AND METHODS .................................................................... 28 2.1 Cloning and Strains ........................................................................................................ 28 2.2 Polymerase Chain Reaction Conditions ........................................................................ 28 2.3 Purification of Amplified DNA from Polymerase Chain Reaction ................................. 30 2.4 Plasmid and DNA fragment Restriction Digests ............................................................ 30 2.5 Ligation Reaction and Sequencing ................................................................................ 31 2.6 E. coli and Rhodococcus jostii RHA1 Competent Cells .................................................. 31 2.7 Transformation of E. coli and RHA1 .............................................................................. 32 2.8 Expression of Gene Fragments in E. coli and RHA1 ...................................................... 33 2.9 Cell Lysis ........................................................................................................................ 33 2.10 Purification of Snm2 expressed in E. coli .................................................................... 34 2.11 Purification of Snm4 Expressed in E. coli .................................................................... 35 2.12 Expression of Snm4 in Inner Membrane Vesicles ....................................................... 36 2.13 Purification of CFP-10 and CFP10:ESAT6 Expressed in E. coli ..................................... 36 2.14 Purification of Snm2 expressed in RHA1 ..................................................................... 37 iv  2.15 Sodium Dodecyl Sulphate Polyacrylamide Gel Electrophoresis (SDS-PAGE) .............. 37 2.16 Western Blot ............................................................................................................... 38 2.17 Gel Filtration Chromatography ................................................................................... 38 2.18 Clear Native and Blue Native PAGE Analysis ............................................................... 39 2.19 Radiolabeling of CFP-10 and CFP10:ESAT6 ................................................................. 39 2.20 Crosslinking of Snm2 with CFP-10 or CFP10:ESAT6 .................................................... 40 CHAPTER THREE RESULTS ............................................................................................... 41 3.1 Expression and Purification of ESX-1 Secretion System Components in a Gramnegative System .................................................................................................................. 41 3.1.1 Expression of Snm1 ................................................................................................ 41 3.1.2 Expression and Purification of Snm2 ...................................................................... 45 3.1.3 Expression and Purification of Snm4 ...................................................................... 50 3.1.4 Snm4 Expressed in Inner Membrane Vesicles........................................................ 53 3.1.5 Expression and Purification of ESAT-6 and CFP-10 ................................................ 55 3.1.6 Formation of the CFP10:ESAT6 Heterodimeric Complex ....................................... 59 3.2 Expression and Purification of ESX-1 Secretion System Components in a Gram-positive System ................................................................................................................................. 61 3.2.1 Expression and Purification of Snm2 ...................................................................... 61 3.2.2 Gel Filtration and Native PAGE Analysis of Soluble Snm2 ...................................... 65 3.3 Binding Interactions Between Snm2 and CFP10:ESAT6 ................................................ 71 3.3.1 Gel Filtration Chromatography of Snm2 with CFP-10 and CFP10:ESAT6 ............... 71 3.3.2 Native-PAGE Analysis of Snm2 and CFP10:ESAT6 .................................................. 75 3.3.3 Binding Interaction of Snm2 and CFP-10 using the Radioisotope I125.................... 78 3.3.4 Interactions between Snm2 and CFP10:ESAT6 using Crosslinking ........................ 79 CHAPTER FOUR .............................................................................................................. 83 4.1 Discussion and Conclusion ............................................................................................ 83 4.2 Future Directions ........................................................................................................... 87 REFERENCES................................................................................................................... 88 APPENDIX ...................................................................................................................... 92  v  LIST OF TABLES Table 2.1 - Genetic constructs, primers, vectors and restriction sites………………………..…..…..29 Table 2.2 - Polymerase Chain Reaction Program………………………………………….…………………….30  vi  LIST OF FIGURES Figure 1.1 - The Extended Region of Difference (RD1) of the M. tuberculosis Genome…………3 Figure 1.2 - Comparison of the ESAT-6 Clusters in M. tuberculosis and M. smegmatis……………………………….…………………………….........................……………………………..…..7 Figure 1.3 - Genes Involved in the ESX-1 Secretion System…………………………………………………9 Figure 1.4 - Schematic Representation of the Cell Wall of M. tuberculosis…………………………11 Figure 1.5 - The Solution Nuclear Magnetic Resonance Structure of CFP-10 and ESAT-6……13 Figure 1.6 - Localization of Conserved ESX-1 Secretion System Components……………….…..16 Figure 1.7 - Yeast Two Hybrid Analysis of ESX-1 Secretion System Components……………….18 Figure 1.8 - Proven Protein-Protein Interactions of the ESX-1 Secretion System Components……………………………………………………………………………………………………………………...20 Figure 1.9 - Rv3871 Interaction with CFP-10………………………………………………………………………22 Figure 1.10 - In vitro Analysis of C-terminal Signal Sequence of CFP-10…………….……………….24 Figure 1.11 - C-terminal Signal Sequence of CFP-10 Fused to Yeast Ubiquitin………….…………26 Figure 3.1 - Genetic Constructs of Snm1………….…………………………………………………………………43 Figure 3.2 - Expression of Snm1 in E. coli……………………………………………………………………………44 Figure 3.3 - Expression of Snm2 in E.coli…………………………………………………………………………….46 Figure 3.4 - IMAC Batch Purification of N-terminal His-tagged Snm2 from E. coliI BL21………48 Figure 3.5 - Gel Filtration of Snm2 Expressed in E. coli…………………………………………………….…49 Figure 3.6 - Western Blot Expression Test of Snm4 in E. coli……………………………………….………51 Figure 3.7 - IMAC and Ion Exchange Chromatography of N-terminal His-tagged Snm4………52 Figure 3.8 - Western Blot of Snm4 Enriched in Inner Membrane Vesicles of E. coli……….……54 Figure 3.9 - Genetic Constructs of CFP-10 and ESAT-6………………………………………………………..56 Figure 3.10 - Expression of CFP-10 and ESAT-6 in E. coli BL21…………………………………………….56 Figure 3.11 – IMAC of CFP-10 and CFP10:ESAT6…………………………………………………………………58 vii  Figure 3.12 – Gel Filtration Chromatography and CN-PAGE Analysis of CFP-10 and CFP10:ESAT6………………………………………………………………………………………………………………………60 Figure 3.13 – Expression of Snm2 in RHA1……………………………………….…………………………………63 Figure 3.14 – IMAC Purification of Snm2……………………………………………….……………………………64 Figure 3.15 - Gel Filtration Chromatography of Snm2…………………………………………………………66 Figure 3.16 – Gel Filtration Chromatography of Snm2 with Various ATP Concentrations……69 Figure 3.17 – Native-PAGE Analysis of Snm2………………………………………………………………………70 Figure 3.18 – Gel Filtration Chromatography of Binding Interactions Between Snm2 and CFP10……………………………………………………………………………………………………………………………………….73 Figure 3.19 - Gel Filtration Chromatography of Binding Interactions Between Snm2 and CFP10:ESAT6………………………………………………………………………………………………………………………74 Figure 3.20 – CN-PAGE Analysis of Binding Interactions between Snm2 and CFP10:ESAT6……………………………………………………………………………………………………………………...76 Figure 3.21 – Binding Interaction of Radiolabeled CFP-10 with Snm2…………………………………77 Figure 3.22 – Crosslinking Between Snm2 and CFP-10……………………………………………………….81 Figure 3.23 – Crosslinking Between Radiolabeled and Non-Radiolabeled Snm2 and CFP10……………………………………………………………………………………………………………………………………….82  viii  ACKNOWLEDGEMENTS I would like to thank above all my supervisor, Dr. Franck Duong, for his patience in teaching me the basics of protein biochemistry and scientific writing. I would also like to thank my lab members, Dr. M. Alami, Dr. Antoine Malliard, Kush Dalal, Huan Bao, Cathy Zhang, Jean-Francois Montariol and all the undergraduate students for their support during the various stages of my degree. I am grateful for the financial support from the Duong Laboratory funded by the Natural Sciences and Engineering Research Council of Canada, as well as the Canadian Institutes of Health Research. I am also grateful to my supervisory committee, Dr. Eltis and Dr. Strynadka, for their input and direction on my research project. A special thanks to Sachi Okamoto and Dr. Eltis, who provided me with the Rhodococcus jostii RHA1 strain, and helped me with subsequent experiments. I would like to give thanks to my family and friends for their support and inspiration.  ix  DEDICATION  To my family  x  CHAPTER ONE  INTRODUCTION 1.1 Preface Tuberculosis is a common and deadly disease caused by the Gram-positive bacteria Mycobacterium tuberculosis. This organism infects one third of the world’s population, and causes 2-3 million deaths annually despite available treatments [1]. Drug resistant strains are emerging and pose a potential health threat. Therefore, it is necessary to understand M. tuberculosis molecular mechanisms of virulence to facilitate novel therapeutics for the related disease.  1.2 The BCG vaccine and the Region of Difference RD1 The live-attenuated vaccine, or BCG, derived from Mycobacterium bovis was developed by Albert Calmette and Camille Guerin in 1921 [1]. Analysis of the M. tuberculosis sequenced genome revealed that a 9.5 kilobase region (termed RD1) was absent in BCG (Figure 1.1) [1]. Genes within and flanking RD1 were also absent in another avirulent strain, Mycobacterium microti. Therefore, to investigate the importance of RD1, genes within and surrounding the region were inserted into the BCG genome by recombination [2, 3]. This complemented BCG strain, BCG, and M. tuberculosis were all tested for virulence by intravenous injection in mice. The complemented BCG persisted and multiplied more than BCG but less than the wild type M. tuberculosis in the lungs and spleen of the tested mice [3]. This study showed that presence of RD1 in the complemented BCG strain improved its survival within the host. 1  In a separate study, the RD1 sequence was deleted from the genome of M. tuberculosis [4]. The deletion mutant and M. tuberculosis were separately injected into individual human macrophage cells. M. tuberculosis multiplied ten-fold within four days and destroyed the cell monolayer, while the RD1 deletion strain multiplied only three-fold and did not damage the cell monolayer [4]. Therefore, RD1 is necessary for bacterial survival within the macrophage.  2  Figure 1.1-The Extended Region of Difference (RD1) of the M. tuberculosis Genome. The genes shown represent RD1 and surrounding region on the chromosome of M. tuberculosis. The exact area deleted from the BCG vaccine is represented by ∆RD1bcg. The genes esxB and esxA represent cfp-10 and esat-6 respectively.  3  1.3 Dissection of RD1 Before the availability of the sequenced genome, a secreted T-cell protein antigen was isolated and purified from the culture medium of M. tuberculosis. The protein was termed the 6-kDa early secretory antigenic target (ESAT-6, Rv3875), because it is recognized by immune cells during the early phase of M. tuberculosis infection. The gene sequence was obtained by screening expressed proteins from genomic libraries with an antibody targeted to ESAT-6. Western blots indicated this protein was absent in BCG [5, 6]. Later, when the M. tuberculosis genome sequence became available, the gene was mapped within RD1. The promoter region was cloned and characterized, and a gene was found to co-transcribe with ESAT-6. This gene situated next to ESAT-6 encoded for a protein identified as the 10-kDa culture filtrate protein (CFP-10, Rv3874) [7]. Both ESAT-6 and CFP-10 are secreted outside the bacteria into the culture medium, otherwise known as the culture filtrate. When purified ESAT-6 or CFP-10 was incubated with human blood samples, a T-cell host immune response was stimulated, quantified by the release of the cytokine gamma interferon (IFN-γ) *8+. However, both proteins are synthesized without obvious secretion signal sequences. This evidence suggested secretion from a Sec-independent pathway. During genome sequencing, eleven additional genes were found to share sequence similarity to esat-6. Also, several genes with similarity to cfp-10 were found situated directly adjacent to esat-6. These genes were grouped as the esat-6 family genes. The originally annotated esat-6 and cfp-10 were found to lie in a gene cluster duplicated along the genome. In silico analysis identified five independent ESAT-6 gene clusters in M. tuberculosis genomic DNA [9]. Throughout the clusters, conserved genes were revealed when compared to RD1 and the 4  surrounding region, termed the extended RD1 (Figure 1.2) [9]. These included putative ABC transporters (integral inner-membrane proteins), serine proteases, and adenosine triphosphate hydrolyzing proteins (ATPases). Gene clusters with homologues of the conserved genes found in M. tuberculosis were identified in the related organism Mycobacterium smegmatis (Figure 1.2) [9]. These genes encode proteins that are often required in protein secretion. In gram negative bacteria, other protein secretion systems involved in virulence such as type III or type IV secretion, are often found in gene clusters adjacent to the substrates they secrete. This led to the hypothesis that genes surrounding esat-6 may be involved in its secretion [9]. Two different approaches were used to prove this hypothesis. One approach introduced genes from extended RD1 of M. tuberculosis into the BCG genome. A clone containing only esat-6/cfp-10 was inserted. Both antigens were expressed in the cytosol, but neither was secreted in amounts comparable to M. tuberculosis [2]. Surprisingly, when the entire extended RD1 of M. tuberculosis was inserted into BCG, a significant amount of ESAT-6 and CFP-10 was exported and recovered in the culture filtrate [2]. This experimental evidence proved that RD1 was essential for export of ESAT-6. Another approach deleted specific genes within RD1, to establish their effect on ESAT-6 export. Initially, three genes neighboring esat-6 termed rv3870, rv3871 and rv3877 were deleted from M. tuberculosis using transposon mutagenesis [10]. The genes of rv3870 and rv3871 code for ATPases, and rv3877 encodes for an integral membrane protein. All of these genes are conserved in other ESAT-6/CFP-10 clusters, and were deemed potential components of the secretion system. Each of the transposon mutant strains failed to export ESAT-6 and CFP-  5  10 into the culture filtrate when compared with M. tuberculosis [10]. This result was not due to a polar effect as complementation of the deleted genes enabled ESAT-6/CFP-10 export out of the bacterium.  6  Figure 1.2 - Comparison of the ESAT-6 Clusters in M. tuberculosis and M. smegmatis. The genes represented within the extended RD1 region are present multiple times along the genome of M. tuberculosis and homologue genes are present multiple times along the genome of M. smegmatis. The clusters are compared, with an emphasis on the conserved genes throughout the clusters. Gaps between genes do not represent physical gaps between genes on the genome. Gene families were named arbitrarily according to their position in RD1. Genes encoding for homologous ESAT-6 proteins are represented by Family H (dark blue). Genes encoding for homologous CFP-10 proteins are represented by family G (light blue). The two genes encoding for an ATPase are represented by family D (purple). The amino-terminal transmembrane protein is represented by family C (orange). The integral membrane protein is represented by family J (green). Finally, the mycosin substilisin-like protease is represented by family K (red). These conserved genes suggest their importance in secretion of ESAT-6 and CFP-10 [9].  7  Although three genes were identified as potential components of the ESAT-6 secretion system, the extended RD1 contains numerous genes. A study created a series of modified cosmids where each gene within M. tuberculosis extended RD1 was mutated by in vitro transposon mutagenesis, insertion of an antibiotic cassette, or by small deletions [11]. The cosmids were used to complement various M. tuberculosis models, including M. tuberculosis H37Rv strain with extended RD1 deleted, BCG, and Mycobacterium microti. The strains were analyzed for their ability to secrete ESAT-6/CFP-10. Various mutations in the genes rv3864rv3867, rv3876, rv3878 and rv3879 still produced ESAT-6 in the culture filtrate and therefore were not essential. However, the genes rv3868-rv3871, rv3874, rv3875 and rv3877 were essential, as deletion mutants prevented CFP-10 or ESAT-6 secretion [11]. By transposon insertions into the genes rv3872 (PE35) or rv3873 (PPE68), no detectable expression of ESAT-6 or CFP-10 was observed, suggesting these genes regulate antigen expression within this region of M. tuberculosis [11]. PE and PPE families are named for the conserved proline (P) and glutamate (E) residues near the N-terminal region of the encoded proteins. Together, these genetic studies established the exact genes required for ESAT-6/CFP-10 secretion and created a model for a specialized secretion system important for M. tuberculosis (Figure 1.3).  8  Figure 1.3 - Genes Involved in the ESX-1 Secretion System. Plus or minus signs below the coding sequences show the involvement of each gene in: ESAT-6 secretion in species belonging to the M. tuberculosis complex (Mtb); haemolysis and ESAT-6 secretion in Mycobacterium marinum (Mm); ESAT-6 secretion in M. smegmatis. A question mark indicates inconclusive results, and a blank indicates not enough information is available. The arrows represent the different coding sequences and the direction of their transcription. The numbers above the arrows represent the respective gene names in M. tuberculosis [12].  9  1.4 Protein Secretion in Mycobacteria After genetic analysis of RD1 from M. tuberculosis, the genes identified as essential to secrete ESAT-6 and CFP-10 were grouped as part of a secretion system for these virulence factors, termed the early secretory antigenic target 6 system 1 (ESX-1) or the type VII secretion system [12]. The specialized substrates require an alternative secretion system for passage through the three-layered cell envelope of M. tuberculosis [12]. The first layer is the inner membrane surrounding the bacterium. On the outer surface of this membrane, peptidoglycan and arabinogalactan form a covalently linked network. This network is also covalently linked to extremely long hydroxylated branched-chain fatty acids termed mycolic acids. Free lipids such as trehalose dimycolate and a variety of phospholipids and glycolipids intercalate with the mycolic acids, forming a very rigid hydrophobic layer [13, 14]. Due to the perpendicular arrangement of the lipids and mycolic acids to the surface of the inner membrane viewed by electron micrographs, this layer is termed the ‘mycomembrane’. Finally, the outer surface of the mycolic acids is coated by polysaccharide molecules. This is otherwise known as the S-layer and constitutes up to 90% carbohydrates (Figure 1.4) [14].Due to this complex arrangement substrates must be secreted through the inner membrane, the mycomembrane, and the outer layer of the bacteria.  10  Figure1.4 - Schematic Representation of the Cell Wall of M. tuberculosis. The cell wall is mainly composed of mycolic acids, peptidoglycan and arabinogalactan. These three components are all covalently linked. This linkage provides a rigid hydrophobic layer just outside the inner membrane known as the mycomembrane. The outer part of the mycomembrane contains various free lipids such as trehalose dimycolate and a variety of phospholipids and glycolipids. The free lipids intercalate with the mycolic acids. Most of these lipids are specific for M. tuberculosis. Finally, the outer layer, otherwise known as the capsule or S-layer, mainly contains polysaccharides such as glucan and arabinomannan [12].  11  1.5 ESAT-6/CFP-10 ESAT-6 and CFP-10 both belong to the family of WXG100 proteins, which are widely distributed among actinobacteria and Gram-positive bacteria. These 100-residue proteins contain a conserved Tryptophan-X-Glycine (WXG) motif [15]. Initially, ESAT-6 and CFP-10 were purified separately [5, 7]. Using transposon mutagenesis, it was realized they are co-transcribed on the genome [11]. Yeast two hybrid assays provided evidence of an interaction [10], and the two proteins were expressed and purified together [16]. If expressed and purified alone, CFP-10 forms a random coil identified by circular dichroism. However, in the presence of purified ESAT6, CFP-10 changes its configuration from a random coil to a helical conformation [16]. In E. coli, the proteins co-purified in a 1:1 ratio, providing evidence for a 1:1 complex. The resolved solution structure of both proteins through nuclear magnetic resonance (NMR) confirmed complex formation [17]. The structure of the ESAT-6/CFP-10 complex provided insights on its secretion. The two proteins form a 1:1 heterodimeric complex, with a tight helical conformation (Figure 1.5). In spite of this, the C-terminal region of CFP-10 remained unstructured, evidence of a possible signal sequence [17].  12  Figure 1.5 - The Solution Nuclear Magnetic Resonance Structure of CFP-10 and ESAT-6. The 1:1 heterodimeric complex is shown along with the unstructured C-terminal tail of CFP-10. The last seven amino acids of CFP-10 code for its signal sequence, targeting it for secretion [17].  13  In light of the solved structure, recent studies have tried to elucidate a role of ESAT-6 and CFP-10 within the host cell. ESAT-6 binds specifically to liposomes. Due to its increase in helical content upon interaction with the lipids contained in the liposomes, it was suggested that ESAT6 inserts itself into the lipid bilayer [18]. Conversely, CFP-10 does not specifically interact with liposomes, and if ESAT-6 is in a complex with CFP-10, neither protein shows an interaction. The same study reported a dissociation of the complex under acidic conditions similar to the phagosome compartment (approximately pH 4.0-5.0) [18]. This suggests that CFP-10 is a chaperone for ESAT-6, and in acidic conditions, the complex dissociates and ESAT-6 interacts with a lipid bilayer. Supporting evidence has shown that purified ESAT-6 forms pores within host cell membranes [19]. This is important for the bacterium Mycobacterium marinum. This bacterium survives within the host by escaping the phagosome of the engulfing macrophage. Therefore, ESAT-6 may facilitate escape of the bacterium from the host cell by creating pores with the membrane of the phagosome [19]. However, no supporting evidence has been shown within M. tuberculosis.  1.6 Conserved Components of the ESX-1 Secretion System Mutagenesis of RD1 has indicated up to 14 different genes required for a functioning ESX-1 secretion system [10, 11, 20, 21]. Also, genetic screens using transposon mutagenesis identified a gene cluster outside of the extended RD1 that is essential for ESAT-6 and CFP-10 secretion [22]. In silico analysis of the ESAT-6 clusters have focused on six conserved genes (Figure 1.6). ESAT-6 and CFP-10 are two conserved components and are substrates of the system, as they are found in the culture filtrate [9, 12]. The proteins encoded by Rv3870 and Rv3871 are  14  part of the FtsK-SpoIIIE family of AAA ATPases (ATPases Associated with various cellular Activities) and represent another conserved component of the secretion system [9, 12]. FtsK is involved in cell division, while SpoIIIE is involved in sporulation. Both FtsK/SpoIIIE family proteins and Rv3870/Rv3871 contain P-loop ATP binding sites along with a transmembrane component [23]. FtsK/SpoIIIE family proteins encode in one gene both the membrane component and cytosolic ATP binding regions [23]. In this case, the membrane component and ATP binding regions are encoded by two genes, Rv3870 and Rv3871. Rv3870 contains two transmembrane segments and resembles the N-terminal region of FtsK, while Rv3871 is cytosolic and contains ATP binding regions, resembling the C-terminal region of FtsK. A fourth conserved component is the substilisin-like protease termed mycosin 1 or MycP1 [12]. This is a calcium dependent serine-endoprotease and may act as a signal peptidase, although its substrate is unknown. The fifth conserved component is a membrane protein encoded by the gene Rv3877 [12]. The primary sequence of this protein contains 11 transmembrane segments and may represent a protein translocation channel for this secretion system. The sixth conserved component is another membrane protein encoded by Rv3869 [12]. Although it is not considered a conserved component of the ESX-1 secretion system, Rv3868 is a putative ESX-1 secretion system chaperone and contains an AAA-ATPase domain [24]. Recently, this protein has been shown to form hexamers and has ATPase activity [24]. However, its role in the ESX-1 secretion system is unclear. Also, small proteins part of PE and PPE families are possibly part of the ESX-1 secretion system. These families are similar to ESAT-6 and CFP-10 because they also form gene pairs and 1:1 tight complexes [12]. They seem to have a functional link to ESX-1; however more evidence is needed to fully understand their role in secretion. 15  Figure 1.6 - Localization of Conserved ESX-1 Secretion System Components. Rv3870 and Rv3871 are ATPases part of the FtsK-SpoIIIE protein family. Rv3870 contains two transmembrane helix domains and is localized to the membrane. Rv3871 is a soluble protein localized to the cytosol. Rv3869 and Rv3877 are both localized to the membrane. Rv3869 contains one transmembrane region while Rv3877 contains 11 transmembrane regions. MycP1 is a substilisin-like protease and contains one transmembrane region, localizing it to the membrane. CFP-10 and ESAT-6 are soluble and are potential substrates of the ESX-1 secretion system. Although not a conserved component, Rv3868 is a putative cytosolic chaperone for M. tuberculosis with an AAA ATPase domain [12].  16  1.7 Protein-Protein Interactions between ESX-1 Secretion System Components Interactions between the conserved ESX-1 secretion system components were investigated to elucidate a mechanism of secretion [25, 26, 27]. Yeast two hybrid experiments were completed by Stanley et. al (2003) on Rv3870, Rv3871, ESAT-6 and CFP-10 [10]. The esat-6 and rv3871 genes were cloned into a vector containing the binding domain of GAL4 and cfp-10 was cloned into a vector containing the activating domain of GAL4. If the vector containing ESAT-6 or Rv3871 was transformed with the vector containing CFP-10, blue yeast colonies appeared indicating ESAT-6 and Rv3871 interact with CFP-10 [10]. Conversely, when cfp-10 was cloned with the binding domain of GAL4 and esat-6 and rv3871 were cloned with the activating domains of GAL4, the same result occurred, confirming the interaction. To investigate this interaction further, the same study cloned the C-terminal region (residues 252-747) of rv3871 into a plasmid. This truncated Rv3871 mutant also interacted with CFP-10, suggesting the Nterminal region of Rv3871 is not necessary for the interaction. Using the same technique, only full length Rv3871 was shown to interact with Rv3870 (Figure 1.7) [10].  17  Figure 1.7 - Yeast Two Hybrid Analysis of ESX-1 Secretion System Components. This figure represents the yeast two hybrid assay completed between ESAT-6, CFP-10, Rv3870 and Rv3871. The bait vectors contain a ESX-1 secretion system component fused with the GAL4 binding domain. The prey vectors contain a ESX-1 secretion system component fused with the GAL4 activating domain. If the two domains interact in the nucleus of the yeast cell, the reporter gene lacZ is transcribed, breaking down 5-bromo-4-chloro-3-indolyl-β-D-galactoside and producing a blue color. This is evident for interactions between: ESAT-6 and CFP-10, Rv3870 and Rv3871 (full length), CFP-10 and Rv3871 (full length) and CFP-10 and Rv3871 (C-terminal region) [10].  18  This study provided the first experimental evidence of an interaction between secretion system components. A mechanism of secretion was hypothesized based on the results obtained. First ESAT-6 and CFP-10 interact with each other, possibly with CFP-10 acting as a secreted chaperone. Rv3871 binds CFP-10, and brings ESAT-6 and CFP-10 to the membrane by binding the transmembrane component Rv3870. Together Rv3870 and Rv3871 may produce a completed ATPase and propel ESAT-6 and CFP-10 outside the inner membrane through Rv3877, the membrane protein (Figure 1.8) [10]. To prove this mechanism true, the protein-protein interactions between these separate components need to be investigated biochemically.  19  Figure 1.8 - Proven Protein-Protein Interactions of the ESX-1 Secretion System Components. Depicted are the interactions known to occur through yeast two hybrid assays. Through these interactions a model of secretion has been formed. CFP-10 and ESAT-6 form a heterodimeric complex. The C-terminal region of CFP-10 interacts with Rv3871. Rv3871 interacts with Rv3870, bringing CFP-10 and ESAT-6 to the membrane. Using its ATPase activity, Rv3871 and Rv3870 propel CFP-10 and ESAT-6 through the membrane using the integral membrane protein Rv3877, allowing the substrates to enter the extracellular milleu [12].  20  1.8 C-terminal Signal Sequence of CFP-10 To further examine the mechanism of secretion, Champion et. al (2006) investigated CFP10. In this study, the C-terminal region of CFP-10 was deleted. This CFP-10 mutant was used to test interaction through yeast two hybrid assays with Rv3871 and ESAT-6 [28]. No proteinprotein interactions were evident, suggesting the C-terminal region of CFP-10 is necessary for protein interaction and secretion. The N-terminal region of CFP-10 was also deleted, and this abolished an interaction with ESAT-6, but still maintained an interaction with Rv3871 (Figure 1.9). Because the C-terminal region of CFP-10 was required for interaction with both ESAT-6 and Rv3871, Champion et. al (2006) mutated specific residues within this region and tested the interaction with Rv3871 using yeast two hybrid assays [28]. In this case, mutants of CFP-10 were fused to the GAL4 binding domain and Rv3871 was fused to the GAL4 activating domain. Mutation of specific residues abolished the interaction between CFP-10 and Rv3871. Further experimentation revealed the last seven amino acids of CFP-10 were sufficient for Rv3871 interaction (Figure 1.9) [28].  21  Figure 1.9 - Rv3871 Interaction with CFP-10. Deletion mutants of CFP-10 were created and tested for interaction with Rv3871 using yeast two hybrid assays (A). An interaction is shown with Rv3871 and the C-terminal region of CFP-10. The last seven amino acids of CFP-10 were also tested using the same technique (B). Results show the last seven amino acids of CFP-10 interact with Rv3871, providing evidence of a signal sequence contained within this region [28].  22  To individually test this interaction, the study completed an in vitro pull down assay between CFP-10 and Rv3871 [28]. As shown in previous yeast two hybrid assays, the C-terminal region of Rv3871 was sufficient to interact with CFP-10. Therefore, residues 248-591 of Rv3871 were HA-affinity tagged and fused to agarose beads. These beads were incubated with E. coli lysate expressing either CFP-10 or a seven amino acid C-terminal deletion of CFP-10. To stabilize any weak interactions, a homo/bi-functional crosslinker Dithiobis succinimidyl propionate was added. After incubation, the agarose beads were isolated and any bound protein was identified through western blot. The western showed CFP-10 pulled down by the Rv3871 bound beads in the presence of a crosslinker. However, the C-terminal deletion of CFP-10 was not pulled down by the same beads (Figure 1.10) [28]. This provided evidence that the C-terminal region of CFP10 contains a signal sequence necessary for its targeting to the ATPase Rv3871 [28]. To determine whether the last seven amino acids of CFP-10 code for a signal sequence, mutant forms of CFP-10 were created, with either deletions or point mutations within the Cterminal region [28]. These mutant clones were transformed into a cfp-10 deletion strain of M. tuberculosis. The expression of ESAT-6 and CFP-10 was not affected by the mutations within CFP-10, but secretion of both virulence factors was eliminated. Therefore, the mutations directly affected secretion, suggesting the last seven amino acids are essential for export and code for a signal sequence [28].  23  Figure 1.10 - In vitro Analysis of C-terminal Signal Sequence of CFP-10. To further prove the last seven amino acids of CFP-10 interact with Rv3871, a pull down assay was performed. Rv3871 was fused to agarose beads. The beads were incubated with E. coli lysate expressing CFP-10 or the Cterminal deletion of CFP-10. The interaction was stabilized by the presence of a crosslinker. In comparison to background levels, Rv3871 specifically bound CFP-10, but did not bind the seven amino acid C-terminal deletion of CFP-10(CFP-10∆7CT) *28+.  24  Ubiquitin is highly conserved among many species, and is usually present within the cell. Therefore, Champion et. al (2006) fused yeast ubiquitin with the last seven amino acids of CFP10. The ubiquitin fusion protein was expressed in the CFP-10 knock out strain of M. tuberculosis. The fused ubiquitin was secreted from the bacterium, while the wild-type ubiquitin was not (Figure 1.11) [28]. This provided substantial evidence for the presence of a signal sequence within the C-terminal region of CFP-10, and provided insight on mechanism of secretion of the virulence factors. The information gathered within the last few years has precisely identified which genes within RD1 are necessary for secretion, as well as explained how CFP-10 and ESAT-6 are identified by components of the secretion system. However, the amount of biochemical evidence proving this mechanism is lacking, as many of the protein interactions have been studied in a cellular environment. Therefore, it is necessary to separate these components and study their interactions in vitro, to better understand the mechanism of secretion.  25  Figure 1.11 - C-terminal Signal Sequence of CFP-10 Fused to Yeast Ubiquitin. The last seven amino acids of CFP-10 was fused to yeast ubiquitin and transformed into a CFP-10 knock out M. tuberculosis strain. The cell wall pellet (P) and the culture filtrate (S) were compared. When the seven amino acid sequence was present, ubiquitin was secreted out of M. tuberculosis and into the culture filtrate. This provided concrete evidence for the C-terminal region of CFP-10 containing a signal sequence used for secretion in M. Tuberculosis [28].  26  1.9 Thesis Investigation The ESX-1 secretion system of M. tuberculosis has thus far been characterized primarily by means of genetic analysis. I propose to study the ESX-1 secretion system by analyzing the membrane localization, oligomeric state, and enzymatic activity of the core components Rv3870 (Snm1), Rv3871 (Snm2) and Rv3877 (Snm4). I will analyze the protein interactions between these three core components and the substrates ESAT-6 and CFP-10. This will shed light on the organization and specific function of the ESX-1 translocation components.  27  CHAPTER TWO  MATERIALS AND METHODS:  2.1 Cloning and Strains M. smegmatis genomic DNA available in the lab was used as a template for primer oligos (Invitrogen) targeted to homologues of Rv3870 (msmeg_0061, snm1), Rv3871 (msmeg_0062, snm2), Rv3877 (msmeg_0068, snm4), CFP-10 (msmeg_0065, cfp-10) and ESAT-6 (msmeg_0066, esat-6) in polymerase chain reactions. Restriction sites were utilized at an end point of each primer. The genetic constructs including restriction sites are listed (Table 2.1). Various tags and vectors are indicated, as well as the transformed expression strains for each construct.  2.2 Polymerase Chain Reaction Conditions A 50 microlitre (µL) polymerase chain reaction was completed using 0.5µg of M. smegmatis genomic DNA as a template, 2µL of 10mM deoxyribonucleotide triphosphates (Invitrogen), 2µL of 5µM primer (5’), 2µL of 5µM primer (3’), 10 µL of 10X GC Buffer (Finnzymes), 15µL of 4M betaine (Sigma) and 20% dimethyl sulfoxide (Fisher), 16.5µL of distilled H2O and 0.5µL of 2U/µL PhusionTM DNA Polymerase (Finnzymes). The reaction was performed in an Eppendorf Mastercycler 5332 thermomulticycler (Table 2.2).  28  Table 2.1 – Genetic constructs, primers, vectors and restriction sites. The melting temperature (Tm) is indicated, as well as the restriction enzyme site in each primer (shown in lower case). 29  Step  Temperature (°C)  Time (min)  1  95  4  2  95  1  3  60  0.75  4  72  3  5  GO TO STEP 2-4  REPEAT 25X  6  72  5  7  8  HOLD  Table 2.2 – Polymerase Chain Reaction Program.  2.3 Purification of Amplified DNA from Polymerase Chain Reaction PCR products were purified on 1.0% agarose gels in 89mM Tris, 89mM Borate and 2mM EDTA (TBE) buffer system (Bioshop). Bands were cut under a 302nm UV transilluminator. The agarose was melted in 600µL Qiagen buffer and incubated at 50°C for 10 minutes. DNA was isolated employing Qiagen DNA MiniPrep columns and procedure.  2.4 Plasmid and DNA fragment Restriction Digests All PCR products and vectors were digested in Invitrogen React Buffer 2 (50mM Tris-HCl, pH 8.0, 10mM MgCl2, and 50mM NaCl) using the respective restriction enzymes as indicated by Table 2.1.Fifteen microlitres of digested DNA, 10µL of distilled H2O, 3µL of Invitrogen React Buffer 2 and 1µL of each restriction enzyme (Invitrogen; New England Biolabs) were incubated for 2 hours at 37°C. Digestion products were purified using 1% TBE agarose gels and isolated using Qiagen DNA MiniPrep columns and procedure.  30  2.5 Ligation Reaction and Sequencing Digested PCR fragments were incubated with digested vectors at ratios varying from 1:1, 1:2, 1:3, 2:2, 2:1 and 3:1 in a volume of 8µL. One microlitre of 10X T4 DNA ligase buffer (New England Biolabs) was added to each reaction, as well as 20U T4 DNA ligase (New England Biolabs). Reactions were completed for 8-16 hours at 20°C before transformation into competent E. coli DH5α. Cells were plated with the addition of either 80µg/mL ampicillin (for the vectors pET23, pBad22 and pTip-QC1), or 50µg/mL kanamycin (for the vector pET28), on Luria Broth (Miller) agar plates overnight at 37°C. Chosen colonies were incubated with 5mL of Luria Broth (LB) at 37°C overnight. DNA was isolated using the Qiagen MiniPrep columns and procedure. DNA constructs were concentrated to 100ng/µL. DNA (10µL) was sent to Macrogen Sequencing Service, along with 15µL of 5µM 5’ annealing and 5µM 3’ annealing primers.  2.6 E. coli and Rhodococcus jostii RHA1 Competent Cells E. coli BL21λDE3 and BL21 pLysS Rare competent cells were prepared by growing 2mL of E. coli cells in LB at 37°C overnight. BL21 pLysS Rare was grown with the addition of 50µg/mL chloramphenicol in all media. One hundred millilitres of LB was inoculated with the 2mL overnight culture and grown at 37°C until the optical density at 600nm (O.D.600nm) reached 0.5. At this point, cells were centrifuged at 3000g for 10 minutes and resuspended in 1/10 th volume of ice cold transformation storage buffer (TSB), composed of LB broth containing 10% PEG (MW = 3350), 5% DMSO, 10mM MgCl2 and 10mM MgSO4 (Fisher). R. jostii RHA1 was prepared by growing 2mL of RHA1 cells in LB at 30°C overnight. Fifty millilitres of LB was inoculated with the 2mL overnight culture of RHA1 and grown at 30°C until  31  O.D.600nm reached 0.5. At this point, cells were centrifuged at 3000g. LB supernatant was removed, and the cells were resuspended in 5mL of sterile distilled H 2O. The resuspended cells were spun at 3000g and washed with 5mL of sterile distilled H2O repeatedly. The cell pellet was resuspended in 100µL of sterile distilled H2O.  2.7 Transformation of E. coli and RHA1 One hundred microlitres of E. coli competent cells (BL21λDE3, BL21 pLysS Rare) were incubated with approximately 50ng of DNA at 4°C for 10 minutes. The competent cell mixture was then incubated for 1 minute at 42°C, and immediately after placed at 4°C for 2 minutes. Three hundred microlitres of LB media was added, after which the mixture was shaken at 37°C for 80 minutes. Two hundred microlitres of the cell mixture was plated on LB agar plates containing either 80µg/mL ampicillin for pET23 and pBad22 vectors, or 50µg/mL kanamycin for pET28 vectors. Fifty µg/mL of chloramphenicol was added to all media for BL21 pLysS Rare. The plates were incubated at 37°C overnight. Fifty microlitres of RHA1 competent cells were incubated with approximately 50ng of DNA at 4°C for 10 minutes. The cell mixture was placed in a 1-mm-gapped electrocuvette (Bio-Rad Laboratories) and subjected to a 2.5-kV electric pulse from a Gene Pulser (Bio-Rad Laboratories). The cuvette was placed at 4°C and 200µL of LB media was added. The cells were thoroughly mixed and shook at 37°C for 5 hours. Two hundred microlitres of cells were plated onto LB agar containing 50µg/mL chloramphenicol. The plates were incubated at 30°C for at least 72 hours.  32  2.8 Expression of Gene Fragments in E. coli and RHA1 Twenty millilitres of overnight culture of transformed strains of E. coli (BL21λDE3 , BL21 pLysS Rare) were shaken overnight at 37°C and inoculated into 1.5L of LB media with the addition of the respective antibiotic for the transformed vector and strain. The culture was induced with a final concentration of 0.2% arabinose (Bioshop) for pBad22 vectors or 0.5mM Isopropyl β-D-1-thiogalactopyranoside (Invitrogen) for pET23 and pET28 vectors at log phase, O.D.600nm 0.6, and allowed to grow for up to three hours at 37°C. Twenty millilitres of overnight culture of transformed RHA1 was shaken for 30 hours at 30°C and inoculated into 1.5L of LB media with the addition of 50µg/mL chloramphenicol. The culture was induced with the final concentration of 50µg/mL thiostreptone (Sigma) at O.D.600nm 0.6 and allowed to grow overnight at 30°C. For expression tests, 100µL of whole cell samples were taken before induction, one hour after induction, and two hours after induction. The samples were spun at 10,000 rpm (9.3 rcf) in the Eppendorf 5415R table top centrifuge. The supernatant was removed, and the cell pellet was resuspended in 20µL of 2% sodium dodecyl sulphate (SDS; Bioshop). The samples were analyzed by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE).  2.9 Cell Lysis Three litres of LB containing induced E. coli or RHA1 cells were centrifuged at 5000g and resuspended in 20mL of the following lysis buffers. Snm2 expressed in E. coli BL21λDE3 was resuspended in Buffer A (50mM Tris-HCl (pH 7.9), 300mM NaCl, 5mM MgCl2 (Fisher), 2mM βmercatopethanol (Sigma) and 10% glycerol). Snm4 expressed in E. coli BL21 pLysS Rare was 33  resuspended in Buffer B (50mM Tris-HCl (pH 7.9), 300mM NaCl and 10% glycerol). CFP-10 expressed in E. coli BL21λDE3 was resuspended in Buffer C (20mM Tris-HCl (pH 7.9), 100mM NaCl, and 1mM EDTA). CFP10:ESAT6 co-expressed in E. coli BL21λDE3 was resuspended in Buffer D (25mM NaH2PO4 (Fisher), 100mM NaCl, and 1mM EDTA). Snm2 expressed in RHA1 was resuspended in Buffer E (50mM Tris-HCl (pH 7.9), 300mM NaCl, 1mM adenosine triphosphate (ATP; Sigma), and 10% glycerol). The cells were passed through a French cell pressure press (Sim-Aminco) at 14,000 psi (pounds per square inch) three times to complete cell lysis. Lysates were spun for 10 minutes at 7000g to remove cell debris, unbroken cells, and inclusion bodies. The supernatant was ultracentrifuged at 55,000 rpm for 45 minutes at 4°C in a Type 60 Ti rotor (Beckman). After centrifugation the pellet contained the cell membranes, while the supernatant represented the cytosolic fraction of the cell.  2.10 Purification of Snm2 expressed in E. coli Sepharose beads linked to iminodiacetic groups (Amersham) stored in 20% ethanol were washed extensively with distilled H2O. One millilitre of beads was placed in a gravity filtration column (Amersham). Three hundred millimolar NiSO4 (Fisher) was incubated with the beads for 5 minutes. The beads were washed again with distilled H2O. Buffer A was incubated with the beads for 5 minutes, and removed by gravity. The cytosolic fraction of E. coli BL21λDE3 expressing Snm2 was incubated with the nickel-charged beads at 4°C for 45 minutes with agitation. The beads were washed with 50mL of Buffer A and the flow through was collected by gravity. To further remove contaminants, the beads were washed with 20mL of Buffer A including 30mM imidazole. The wash fraction was collected by gravity and the step was  34  repeated twice. Approximately 10mL of Buffer A including 500mM imidazole was used for elution. One millilitre elution fractions were collected by gravitational force.  2.11 Purification of Snm4 Expressed in E. coli The membrane pellet of Snm4 expressed in E. coli BL21 pLysS Rare was collected after cell lysis. The pellet was resuspended in 10mL of Buffer B with 1% Triton X-100. The resuspended membranes were incubated at 4°C overnight. After incubation, the solubilised membranes were ultracentrifuged at 55,000 rpm for 45 minutes at 4°C in a Type 60 Ti rotor (Beckman) to remove any detergent or protein aggregates. The supernatant was collected for purification. A 5mL fast flow His-tag affinity column (His Trap FF; Amersham) was washed with 50mL of distilled H2O, 20mL of 0.3M NiSO4 and 50mL of distilled H2O at a flow rate of 1mL/min. At a flow rate of 0.5mL/min, Buffer B with 0.03% TritonX-100 was flown through the column for equilibration. Solubilised E. coli membranes expressing Snm4 were loaded onto the column and flown through at a rate of 0.5mL/min. The column was washed with Buffer B including 0.03% TritonX-100 and 50mM imidazole. Snm4 was eluted with 50mM Tris (pH 7.9), 50mM NaCl, 10% glycerol, 0.03% TritonX-100 and 500mM imidazole. The elution fractions were collected, and pooled for further purification. A fast flow MonoQ 5/50 GL column (Amersham) was equilibrated with 50mM Tris (pH 7.9), 50mM NaCl, 10% glycerol, and 0.03% TritonX-100. The elution fractions collected from his-tag affinity purification were loaded onto the column. A linear ion gradient from 50-400mM NaCl was performed over 30 minutes at 1mL/min from the equilibration buffer to 50mM Tris (pH  35  7.9), 400mM NaCl, 10% glycerol, 0.03% TritonX-100. One millilitre elution fractions were collected.  2.12 Expression of Snm4 in Inner Membrane Vesicles The membrane fraction of Snm4 expressed in E. coli BL21 pLysS Rare was obtained as described. The membrane pellet was resuspended in 5mL of Buffer B including 20% sucrose. To separate inner membrane vesicles from outer membrane vesicles of E. coli, a two-step sucrose gradient centrifugation technique was utilized. Five hundred microlitres of 70% sucrose, 800µL of 50% sucrose and 500µL of 20% sucrose were layered in a 2.2mL TLS55 centrifugation tube (Beckman). Four hundred microlitres of resuspended membranes were layered at the top. The samples were spun at 40,000 rpm in the TLS-55 rotor (Beckman) for 4 hours at 4°C. Inner membrane vesicles were collected at the 20-50% sucrose interface.  2.13 Purification of CFP-10 and CFP10:ESAT6 Expressed in E. coli CFP-10 was expressed and induced in E. coli strain BL21λDE3. The cells were lysed in Buffer C. A 5mL fast flow His-tag affinity column (His Trap FF; Amersham) was washed with 50mL of distilled H2O, 20mL of 0.3M NiSO4 and 50mL of distilled H2O at a flow rate of 1mL/min. Buffer C was flown through the column for equilibration. The cytosolic fraction containing CFP-10 was loaded onto the column at 0.5mL/min. The flow through material was collected. The column was washed with Buffer C including 20mM imidazole. One millilitre fractions of CFP-10 were eluted with Buffer C containing 500mM imidazole. CFP10:ESAT6 was expressed and induced in E. coli strain BL21λDE3. The cells were lysed in Buffer D. The cytosolic fraction was obtained, containing both CFP-10 and ESAT-6. A 5mL fast 36  flow His-tag affinity column (His Trap FF; Amersham) was washed with 50mL of distilled H2O, 20mL of 0.3M NiSO4 and 50mL of distilled H2O at a flow rate of 1mL/min. Buffer D was flown through the column for equilibration. The cytosolic fraction was loaded onto the column at 0.5mL/min, and the flow through material was collected in two fractions. The column was washed with Buffer D including 20mM imidazole, and eluted with Buffer D including 500mM imidazole. Fractions from the flow through, wash, and elution were collected and stored.  2.14 Purification of Snm2 expressed in RHA1 Snm2 was expressed and induced in RHA1. The cells were lysed in Buffer E. The cytosolic fraction was obtained. A 5mL fast flow His-tag affinity column (His Trap FF; Amersham) was washed with 50mL of distilled H2O, 20mL of 0.3M NiSO4 and 50mL of distilled H2O at a flow rate of 1mL/min. Buffer E was flown through the column for equilibration. The cytosolic fraction was loaded onto the column at 0.5mL/min, and the flow through material was collected. The column was washed with Buffer E including 20mM imidazole. One millilitre fractions containing Snm2 were eluted with Buffer E including 500mM imidazole.  2.15 Sodium Dodecyl Sulphate Polyacrylamide Gel Electrophoresis (SDS-PAGE) Twelve percent or 15% SDS polyacrylamide gels were utilized. Five microlitres of sample buffer was added to each sample and contained 150mM Tris-HCl (pH 6.8), 10% glycerol, 200mM β-mercaptoethanol, 2% SDS and 1.25% bromophenol blue. Sample buffer used for crosslinking experiments did not contain β-mercaptoethanol. An SDS-PAGE gel was run at 40mA (milli amperes) for 45 minutes. Gels were stained with a solution containing 10% methanol, 7%  37  acetic acid and 0.25% Coomassie Brilliant Blue R-250. Destaining was completed with 20% ethanol and 10% acetic acid.  2.16 Western Blot SDS polyacrylamide gels were placed in transfer buffer for 5 minutes containing 25mM Tris, 0.19M glycine, 0.05% SDS and 10% methanol. A pure 0.45µm nitrocellulose membrane (Bio Rad) was also incubated with transfer buffer for 5 minutes. A semi-dry apparatus was assembled, and transfer occurred at 400mA for 60 minutes. The nitrocellulose membrane was incubated with 5% milk diluted in phosphate-buffered saline (PBS) and 0.5% Tween-20. Successive washes were completed with PBS and 0.5% Tween-20. One in three thousand dilutions of the mouse anti-his monoclonal primary antibody and 1/2000 dilutions of the mouse monoclonal secondary antibody fluorescent at 680nm were incubated with the membrane in PBS and 0.5% Tween-20.  2.17 Gel Filtration Chromatography Gel filtration chromatography experiments were conducted using Amersham Superdex 200 10/300 GL and Superdex 75 HR 10/30 columns. For analysis of the CFP10, CFP10:ESAT6 complex, and Snm2, approximately 200µg of purified protein was loaded in their respective purification buffers. Injection was completed with a 500µL Hamilton syringe into an AKTA system. Gel filtration was continued at a rate of 0.4mL/min, and 500µL fractions were collected. For analysis of interactions between Snm2 and CFP-10, or Snm2 and CFP10:ESAT6, 200µg of Snm2 was incubated with 100µg of CFP10 or CFP10:ESAT6 for 10 minutes at 4°C. The mixture  38  was injected with a 500µL Hamilton syringe into an AKTA system. Gel filtration was continued at a rate of 0.4mL/min, and 500µL fractions were collected.  2.18 Clear Native and Blue Native PAGE Analysis Native gels were made with a linear 4-12% polyacrylamide gradient in 0.19M glycine and 25mM Tris (pH 8.8). Clear-Native gels were run at 20mA for 60 minutes in 0.19M glycine and 25mM Tris (pH 8.8). Two microlitres of 50% glycerol was added to each sample. Blue-Native gels were run at 20mA for 60 minutes with 0.01% Serva BW Blue Dye, 0.19M glycine and 25mM Tris in the cathode, and 0.19M glycine and 25mM Tris in the anode. Two microlitres of 1% Serva BW Blue Dye in 50mM Tris-HCl (pH 8.8) was added to each sample. Gels were stained with a solution containing 10% methanol, 7% acetic acid and 0.25% Coomassie Brilliant Blue R-250. Destaining was completed with 20% ethanol and 10% acetic acid. Clear-Native gels containing radiolabeled protein were created as described. The gels were fixed with 50% methanol and dried. The dried gels were exposed using an iodine isotope imager overnight at 20°C.  2.19 Radiolabeling of CFP-10 and CFP10:ESAT6 Sixty microlitres of 1.0mg/mL CFP-10 was placed into a tube coated with 50µg of 1,3,4,6tetrachloro-3α,6α-diphenylglycouril (Iodogen). The iodine radiolabeling of tyrosine residues occurred with the addition of 25µCi of I125 (Perkin Elmer Life Sciences) for 10 minutes at 4°C. A 20 fold dilution was made and used for subsequent experiments. The same procedure was used for radiolabeling CFP10:ESAT6.  39  2.20 Crosslinking of Snm2 with CFP-10 or CFP10:ESAT6 Crosslinking experiments were completed with the use of a homo/bi-functional crosslinker Dithiobis succinimidyl propionate (Sigma).  40  CHAPTER THREE  RESULTS 3.1 Expression and Purification of ESX-1 Secretion System Components in a Gram-negative System 3.1.1 Expression of Snm1 A homologue to rv3870 (snm1, 2.2 kilobase fragment) was amplified from Mycobacterium smegmatis genomic DNA using a polymerase chain reaction (PCR). The amplified gene fragment was cloned into a pET23 vector containing an in-frame C-terminal six histidine tag (his-tag), and a pBad22 vector containing an in-frame N-terminal his-tag (Figure 3.1). Snm1-pBad22 was transformed into two strains of E. coli, BL21 λDE3 and BL21 pLysS Rare. BL21 plysS Rare contains rare tRNA genes essential for expression of rare codons in E. coli. Snm1-pet23 was transformed into BL21 pLysS Rare. Whole cell samples of each transformed strain were taken before induction, one hour after induction, and two hours after induction. The samples were lysed and solubilised with SDS. The solubilised whole cells were analyzed by 12% SDS-PAGE. Due to numerous solubilised proteins within the cell, a more specific method was needed to confirm the presence of Snm1. A replica 12% SDS-PAGE gel was analyzed by western blot. A monoclonal his-tag antibody was used in conjunction with a fluorescent secondary antibody at a wavelength of 680nm. If the transformed vector containing the gene fragment snm1 was transcribed and translated into protein, a strong band corresponding to the theoretical molecular weight of Snm1 (81.8 kDa) 41  would have been visualized by SDS-PAGE as well as western blot. No expression of Snm1 was seen for any of the transformed strains (Figure 3.2). To further investigate the expression of Snm1, the amplified gene fragment used previously was cloned into the pET41b vector system containing an in-frame glutathione-S-transferase Nterminal tag (Figure 3.1). To promote expression, the snm1gene fragment was truncated by PCR, deleting the first 100 amino acids containing two possible transmembrane segments. The truncated fragment was cloned into the pET41b vector. The truncated Snm1 (Trnc-Snm1) and the full length Snm1 were introduced into the E. coli strain BL21 λDE3. Whole cell samples were taken before induction, one hour after induction and two hours after induction. The samples were solubilised with SDS and analyzed by western blot (Figure 3.2). Expression was observed for both Trnc-Snm1-pet41b and Snm1-pet41b within whole cells. However, the expressed protein formed aggregates during synthesis and appeared in the form of inclusion bodies (data shown). Purification was not completed due to the difficulty of obtaining soluble protein.  42  Figure 3.1 - Genetic Constructs of Snm1. Snm1 (745 amino acids) was amplified from M. smegmatis genomic DNA and inserted into pBad22, pET23 and pET41b. A truncation mutant (Trnc Snm1, 645 amino acids) was incorporated into pET41b, deleting the first 300 base pairs of the gene fragment, or 100 amino acids of the resulting protein. An in-frame six histidine tag (H6) was engineered at the Nand C-terminal ends of Snm1. An in-frame glutathione-S-transferase (GST) tag was engineered at the Nterminal end of Snm1 and the truncated mutant (Trnc Snm1). The numbers above the constructs represent the amino acid number.  43  Figure 3.2 - Expression of Snm1 in E. coli. Snm1-pBad22 was transformed into E. coli strains BL21 and BL21 pLysS Rare. Snm1-pet23 was transformed into BL21 pLysS Rare. One hundred microlitre samples of whole cells were taken during E. coli growth at 37°C before induction (BF), one hour after induction (1h) and two hours after induction (2h). Each sample was prepared as described in the Materials and Methods. Twenty microlitres of solubilised whole cells were run on 12% SDS-PAGE (top left). An arrow indicates where Snm1 should be located. As a control, partially purified Snm2 indicated by an arrow was loaded on the 12% SDS-PAGE. To confirm the presence of Snm1, a replica western blot (bottom left) was completed, using a primary monoclonal his-tag antibody and a secondary fluorescent antibody (680nm). The whole cell samples from the constructs Trnc-Snm1-pet41b and Snm1-pet41b were treated similarly, and a western blot was completed (right). Arrows indicate the appropriate migration of Snm1 and the truncated Snm1 (Trnc Snm1).  44  3.1.2 Expression and Purification of Snm2 A homologue to rv3871 (snm2, 1.9 kilobase fragment) was amplified from M. smegmatis genomic DNA using PCR. The gene fragment was cloned into a pET23 vector containing an inframe C-terminal his-tag, and a pET28 vector containing an in-frame N-terminal his-tag. Snm2pet23 and Snm2-pet28 were transformed into the E. coli strains BL21 λDE3 and BL21 pLysS Rare. Transformed strains grown at 37°C were tested for expression of Snm2. All transformed strains expressed his-tagged Snm2 at the correct molecular weight (64.5 kDa). E. coli cells were lysed as described in the Materials and Methods. The lysed cells were spun at a low speed, to remove any cell debris or intact whole cells. After the low speed spin, the supernatant was separated from the pellet, collected, and spun at a high speed to remove any membranes or membrane components. The supernatant remaining represented the cytosolic fraction of the bacterial cell. High expression of proteins due to the T7 promoter present in pET system vectors may cause protein aggregation. The aggregated protein may form dense inclusion bodies that spin down during the low speed spin after lysis, along with cell debris and whole cells. The best conditions of expression represent transformed strains that produce the most amount of soluble Snm2 when compared to aggregate Snm2. Therefore, to distinguish the quantity of soluble protein obtained compared to aggregated protein, fractions of both the low spin pellet and the cytosol were compared using 12% SDS-PAGE (Figure 3.3).  45  Figure 3.3 - Expression of Snm2 in E. coli. Snm2-pet23 and Snm2-pet28 were transformed into BL21 and BL21 pLysS Rare. All transformed strains were grown at 37°C and induced at O.D.600nm 0.6 with 0.5mM Isopropyl β-D-1-thiogalactopyranoside (IPTG). The cells were collected and lysed in Buffer B. A 5mL cytosolic fraction was obtained. The low spin pellet was resuspended in 5mL of Buffer B. One microlitre, 2µL and 4µL aliquots were taken from the low spin pellet and the cytosolic fraction. The samples were run on 12% SDS-PAGE. Each gel is labeled with its construct and strain. The black arrows indicate the location of Snm2.  46  Comparison of the expression tests revealed pet28-Snm2 expressed in BL21 at 37°C contained the most amount of Snm2 in the cytosol with the least amount of contamination from other proteins. Also, it contained minimal amounts of Snm2 in the low spin pellet. Therefore, purification of Snm2 was continued using this construct and expression condition. Immobilized metal affinity chromatography (IMAC) using the batch purification method was performed as described in the Materials and Methods. Iminodiacetic acid groups are coupled to Sepharose 6 Fast Flow beads by stable ether linkages via a 7-atom spacer. The resin is charged with nickel ions (Ni+2) that are immobilized within the iminodiacetic groups. The nickel ions strongly bind to any exposed histidine groups. Pet28-Snm2 contains an N-terminal his-tag. Due to the strong binding of histidines, Snm2 was extracted from the cytosolic fraction of E. coli at 90% purity and was analyzed by 12% SDS-PAGE (Figure 3.4). One litre of E. coli expressing Snm2 produced up to 5mg of purified protein. Purified Snm2 underwent high speed centrifugation to remove any aggregates or precipitated protein. Samples taken before and after centrifugation were compared on SDSPAGE. Analysis revealed a decrease in Snm2 concentration. Approximately half the protein formed a pellet during centrifugation, indicating a large amount of protein aggregation and precipitation (Figure 3.4). The remaining soluble protein was loaded onto a 10mL Superdex 200 gel filtration column. Gel filtration chromatography separates proteins based on their molecular weight. A large amount of Snm2 eluted in the void volume of the gel filtration column, at approximately 3mL (Figure 3.5). This result indicated the majority of Snm2 formed extremely high molecular weight complexes; leading to the conclusion the protein was highly aggregated.  47  Figure 3.4 – IMAC Batch Purification of N-terminal His-tagged Snm2 from E. coliI BL21. (A) Pet28-Snm2 was expressed in the E. Coli strain BL21. The cells were lysed in Buffer A. Nickel-charged sepharose beads were incubated with the cytosol at 4°C for 45 minutes. The flow through was collected, and the beads were washed with Buffer A including 30mM imidazole. Buffer A with 500mM imidazole was used for elution. Twenty microlitres of each elution fraction (numbered on the gel), 20µL of the flow through (FT) and 20µL of the wash (W) fractions were loaded onto 12% SDS-PAGE. (B) The elution fractions collected were spun at a high speed to remove protein aggregates. Before centrifugation, 10µL sample of a 1.1mg/mL fraction of Snm2 was loaded onto 12% SDS-PAGE (Total). After centrifugation, 10µL of the supernatant was loaded onto the same gel (Soluble). Snm2 protein concentration decreased by approximately half.  48  Figure 3.5 – Gel Filtration of Snm2 Expressed in E. coli. Snm2 in Buffer A was loaded onto a 10mL Superdex-200 gel filtration column. The protein absorbance was measured in arbitrary milli-absorbance units at A280nm (mAU). Aggregates of Snm2 eluted at the void volume of 3mL and soluble Snm2 eluted as a broad peak at 8mL (left). Five hundred microlitre fractions were collected. Twenty microlitres of each elution fraction was loaded on to a 12% SDS-PAGE (right). The numbers above the gel correspond to the fraction number.  49  3.1.3 Expression and Purification of Snm4 A homologue to rv3877 (snm4, 1.7 kilobase fragment) was amplified from M. smegmatis genomic DNA using PCR. The amplified gene fragment was cloned into a pET23 vector containing an in-frame C-terminal his-tag and a pBad22 vector containing an in-frame Nterminal his-tag. Snm4-pet23 and Snm4-pBad22 were transformed into the E. coli strains BL21 λDE3 and BL21 pLysS Rare. Transformed strains grown at 37°C were tested for expression of the transmembrane protein Snm4. Whole cell samples were taken before induction, one hour after induction and two hours after induction. It is difficult to visualize membrane proteins using Commassie staining of SDS-PAGE. Therefore, 20µL of solubilised whole cell samples were analyzed by western blot. Only strains transformed with Snm4-pBad22 expressed his-tagged Snm4 at the correct molecular weight (53.3 kDa) (Figure 3.6). The best condition of expression was Snm4-pBad22 expressed at 37°C in BL21 pLysS Rare. The membrane fraction was extracted and solubilised as described in the Materials and Methods. The solubilised solution was centrifuged to remove any detergent aggregates and non-solubilised protein. The supernatant was loaded onto an IMAC column, charged with Ni+2. His-tagged Snm4 was partially purified from other solubilised membrane components (Figure 3.7). To further purify the protein, the collected fractions from IMAC were loaded onto a Fast Flow Q-Sepharose column. The calculated pI of Snm4 is 6.06 including the engineered his-tag. Purification was completed in Tris buffer of pH7.9, allowing the protein to acquire a negative charge and bind to the positively charged beads of Q-Sepharose. The protein was eluted with increasing salt concentration. However, this method did not help separate Snm4 from other bacterial components, and the protein was approximately 60% purified (Figure 3.7). 50  Figure 3.6 – Western Blot Expression Test of Snm4 in E. coli. Snm4-pBad22 and Snm4-pet23 were transformed into E. coli strains BL21 and BL21 pLysS Rare. 100µL samples of whole cells were taken during E. coli growth at 37°C before induction (BF), one hour after induction (1h) and two hours after induction (2h). Samples were prepared as described in the Materials and Methods. Twenty microlitres of solubilised whole cells were run on 12% SDS-PAGE. To confirm the presence of Snm4, a western blot was completed, using a primary monoclonal his-tag antibody and a secondary fluorescent antibody (680nm). The arrow on the left indicates the position of Snm4.  51  Figure 3.7 - IMAC and Ion Exchange Chromatography of N-terminal His-tagged Snm4. (A) The solubilised membrane fraction was loaded onto a His-tag affinity resin in Buffer B with 0.03% TritonX100. The wash contained Buffer B with 0.03% TritonX-100 and 50mM imidazole, and the elution contained 50mM Tris (pH 7.9), 50mM NaCl, 10% glycerol, 0.03% TritonX-100 and 0.5M imidazole. Twenty microlitres of each elution fraction, flow through material (FT), wash material (W), and solubilised Snm4 (Sol.) was loaded onto a 12% SDS-PAGE. The numbers above the gel correspond to the elution fractions. The arrow indicates the position of Snm4. (B) The ion exchange profile of Snm4 is shown. The sample buffer contained 50mM Tris (pH 7.9), 50mM NaCl, 10% glycerol, 0.03% TritonX-100 and was loaded onto a MonoQ column. The elution buffer contained 50mM Tris (pH 7.9), 400mM NaCl, 10% glycerol, and 0.03% TritonX-100. A linear ion gradient from 50-400mM NaCl was performed, as indicated by the sloped line through the elution profile, measured by the left axis (mM NaCl). The numbers below the elution profile correspond to the elution volume in mL. (C) Twenty microlitres of the elution fractions were collected from ion exchange chromatography and loaded onto 12% SDS-PAGE. The numbers above the gel represent the elution volume in mL. The arrow indicates the position of Snm4.  52  3.1.4 Snm4 Expressed in Inner Membrane Vesicles Snm4 protein was over-expressed in the E. coli BL21 pLysS Rare strain. The bacterial cells were lysed, and the membrane fraction obtained as described earlier. E. coli bacterial cells contain two membrane barriers, the inner membrane and the outer membrane. Both are present in the membrane fraction. Snm4 is expressed within the inner membrane of the bacteria. To obtain only the inner membrane, the entire membrane fraction was subjected to sucrose density centrifugation. Inner membrane vesicles are less dense than outer membrane vesicles, and form a tight band within lower sucrose concentrations. The inner membrane vesicles were extracted and analyzed by western blot (Figure 3.8).  53  Figure 3.8 - Western Blot of Snm4 Enriched in Inner Membrane Vesicles of E. coli. Inner membrane vesicles were prepared from E. coli strain BL21 pLysS Rare. Samples were taken before induction and at 1 hour, 2 hour and 3 hours after induction. In each case 1mL of cells were centrifuged and resuspended in 0.2mL of 50mM Tris (pH 7.9). Twenty microlitres of this sample was loaded in each lane on a 12% SDS-PAGE gel. After induction with 0.2% arabinose, cell samples at 1 hour, 2 hour and 3 hour time points were lysed. The membrane fraction was separated and subjected to sucrose density centrifugation as described in the Materials and Methods. The inner membrane vesicles were resuspended to a concentration of 0.33mg/mL in 50mM Tris (pH 7.9), 50mM NaCl, and 10% glycerol. In each case, 10μL of inner membrane vesicles was loaded onto a 12% SDS-PAGE gel. To detect Snm4, a western blot was completed, using a primary monoclonal his-tag antibody with a secondary fluorescent antibody (680nm).  54  3.1.5 Expression and Purification of ESAT-6 and CFP-10 A homologue of esat-6 and cfp-10 were separately amplified from the M. smegmatis genome by PCR. The cfp-10 gene was cloned into the vectors pET28 and pET23. The esat-6 gene was cloned into the pET23 vector. As mentioned earlier, esat-6 and cfp-10 are situated next to each other on the M. smegmatis genome. Therefore, another gene fragment was amplified containing both cfp-10 and esat-6. This gene fragment was cloned into the pET28 vector, with an N-terminal his-tag fused to cfp-10 (Figure 3.9). The constructs were transformed into the E. coli strain BL21λDE3 and grown at 37°C. Protein expression of CFP-10 and ESAT-6 was tested and analyzed by SDS-PAGE. The theoretical molecular weight of his-tagged CFP-10 is 11.1 kDa, and the theoretical weight of un-tagged ESAT-6 is 9.7 kDa. It was determined CFP-10-pet23 and CFP-10-pet28 expressed a high yield of CFP-10 due to the substantial expression of a protein at the correct molecular weight, confirmed by western blot. The expression of ESAT-6 is not shown. The CFP10:ESAT6-pet28 construct expressed both CFP-10 and ESAT-6, based on the theoretical molecular weight of both proteins (Figure 3.10). Based on yeast two hybrid assays [10, 28], CFP-10 has been shown to interact with Rv3871 (Snm2). Therefore, the purification of CFP-10 was explored. The E. coli strain BL21 λDE3 expressing CFP-10-pet28 was lysed, and the cytosolic fraction obtained. The purification using IMAC was continued as previously published (16). CFP-10 was obtained at >90% purity (Figure 3.11).  55  Figure 3.9 - Genetic Constructs of CFP-10 and ESAT-6. ESAT-6 (100 amino acids) was amplified from M. smegmatis genomic DNA and inserted into pET23 (ESAT6-pet23) with an in-frame Cterminal six histidine tag (H6). CFP-10 (100 amino acids) was amplified from M. smegmatis genomic DNA and inserted into pET23 (CFP10-pet23) with an in-frame C-terminal his-tag and pET28 (CFP10pet28) with an in-frame N-terminal his-tag. CFP-10 and ESAT-6 were amplified together from the M. smegmatis genome and inserted together into pET28, with an in-frame N-terminal his-tag attached to CFP-10 (CFP10:ESAT6-pet28).  Figure 3.10 - Expression of CFP-10 and ESAT-6 in E. coli BL21. CFP10-pet23, CFP10-pet28 and CFP10:ESAT6-pet28 were transformed into E. coli strain BL21. One hundred microlitre samples of whole cells were taken during E. coli growth at 37°C before induction (BF), one hour after induction (1h) and two hours after induction (2h). The samples were prepared as described in the Materials and Methods. Twenty microlitres of solubilised whole cells were loaded on a 15% SDS-PAGE. Arrows indicate the migration of CFP-10 (11.1kDa ) and ESAT-6 (9.7kDa). To confirm molecular weight, protein standards were loaded on the right of the gel, the numbers correspond to the kilo-Daltons of each standard. Also, purified Snm2 was loaded for comparison of molecular weight.  56  ESAT-6 and CFP-10 form a 1:1 heterodimer with one another. Previous studies have shown if both components are purified separately, CFP-10 forms a random coil configuration, while ESAT-6 forms inclusion bodies and must be denatured and refolded into its natural helix conformation. To prevent this from occurring, ESAT-6 and CFP-10 were co-expressed. If expressed together, both proteins remain soluble and tightly bind one another. Therefore, CFP10 would be in a folded helix conformation with ESAT-6. CFP10:ESAT6-pet28 was expressed in E. coli BL21λDE3. The cytosolic fraction was obtained. CFP-10 contained an N-terminal his-tag; therefore IMAC purification was implemented as described in the Materials and Methods. The column was washed and any protein bound onto the beads was eluted. A large amount CFP-10 was found in the wash and elution fractions, with a very small amount of ESAT-6. CFP-10 binds ESAT-6 at a 1:1 ratio. The wash and elution fractions collected did not have an equal amount of both proteins, and therefore would not contain the CFP10:ESAT6 complex. However, analysis of the late fractions of the flow through material (FT2) revealed CFP-10 and ESAT-6 eluted together in a 1:1 ratio, at >90% purity. These fractions were pooled together and concentrated for further analysis (Figure 3.11).  57  Figure 3.11 – IMAC of CFP-10 and CFP10:ESAT6. (A) CFP-10 was expressed and induced in E. coli strain BL21. The cells were lysed in Buffer C. The cytosolic fraction was obtained, and loaded onto a nickel chelating sepharose column. The flow through material was collected. The column was washed with Buffer C including 20mM imidazole. The fractions were eluted with Buffer C including 500mM imidazole. Twenty microlitres of flow through material (FT) and elution fractions (numbered above the gel) were loaded on 15% SDS-PAGE. To confirm the molecular weight, protein standards were run on the right of the gel, the numbers correspond to the kiloDaltons of each standard. (B) CFP-10 and ESAT-6 were co-expressed in the E. coli strain BL21. The cells were lysed in Buffer D. The cytosolic fraction was separated, containing both CFP-10 and ESAT-6, and was loaded onto a nickel chelating sepharose column. The flow through material was collected in two fractions. Initially, many of the contaminant proteins eluted flowed through together (FT1); however a substantial amount of protein was retained and flowed through late (FT2). The column was washed with Buffer D including 20mM imidazole, and eluted with Buffer D including 500mM imidazole. 20µL of each elution fraction, FT1, FT2, and ESAT-6 inclusion bodies were loaded onto 15% SDS-PAGE and are labeled above the gel.  58  3.1.6 Formation of the CFP10:ESAT6 Heterodimeric Complex To further analyze purified CFP-10 and purified CFP10:ESAT6 fractions, gel filtration chromatography was completed using a Superdex 75 gel filtration column. Purified CFP-10 eluted as one peak at the elution volume of approximately 11mL, corresponding to a molecular weight of approximately 14 kDa. Purified CFP10:ESAT6 fractions eluted as one peak at the elution volume of 10.6mL, corresponding to a molecular weight of approximately 23 kDa (calculations in Appendix). This indicated CFP-10 with ESAT-6 may form a complex, as they eluted together during gel filtration chromatography. However, due to the small size difference between CFP-10 and the CFP10:ESAT6 complex (8 kDa), the proteins may have eluted together because they cannot be separated by gel filtration chromatography (Figure 3.12). To further prove both proteins form a complex, Native-PAGE analysis was employed. CN-PAGE was used to visualize the CFP10:ESAT6 complex. CN-PAGE is created by a polyacrylamide gel gradient, and provides an environment devoid of any compound that may produce ionic interactions with the protein of interest. It is a completely non-denaturing technique, where protein complexes are separated due to their ionic charge and molecular weight. Purified CFP-10 and purified CFP10:ESAT6 were analyzed by CN-PAGE to determine if a molecular weight shift was presented due to the formation of the CFP10:ESAT6 complex. Results showed a band corresponding to purified CFP-10 running higher than a single band shown for CFP10:ESAT6. This suggested CFP10:ESAT6 formed a complex, changing the surface ionic properties of CFP-10, causing the complex to migrate faster than purified CFP-10 alone. These results confirm a conformational change, and provide evidence for complex formation between CFP-10 and ESAT-6 during co-purification (Figure 3.12). 59  Figure 3.12 – Gel Filtration Chromatography and CN-PAGE Analysis of CFP-10 and CFP10:ESAT6. (A) Purified CFP-10 was loaded onto Superdex 75 gel filtration column in Buffer C. The protein eluted as a single peak at approximately 11mL. The protein absorbance was measured in arbitrary milliabsorbance units at A280nm (mAU) as seen on the left of the profile. Five hundred microlitre fractions were collected. Twenty microlitres of the relevant elution fractions were loaded onto 15% SDS-PAGE, with the corresponding elution volume above the gel. (B) Co-purified CFP10:ESAT6 collected from IMAC was loaded onto the Superdex 75 gel filtration column in Buffer D. Both proteins eluted as a single peak at 10.6mL. The protein absorbance was measured in arbitrary milli-absorbance units at A280nm (mAU) as seen on the left of the profile. Five hundred microlitre fractions were collected. Twenty microlitres of the relevant elution fractions were loaded onto 15% SDS-PAGE, with the corresponding elution volume above the gel. (C) CN-PAGE containing a polyacrylamide gradient of 4-12% (pH 8.8) was created. Fractions collected from purification of CFP-10 and co-purification of CFP10:ESAT6 were loaded on CN-PAGE. Refolded ESAT-6 was also loaded onto CN-PAGE. Arrows indicate the position of CFP-10, ESAT-6 and the CFP10:ESAT6 complex. 60  3.2 Expression and Purification of ESX-1 Secretion System Components in a Gram-positive System 3.2.1 Expression and Purification of Snm2 Although Snm2 was expressed in E. coli, the protein did not remain soluble, and the majority of the expressed protein formed aggregates. A Gram-positive system may mimic the M. tuberculosis protein synthesis machinery better than a Gram-negative system, and may allow the protein to be synthesized in a completely soluble form. To accomplish this task, RHA1 was utilized. The gene fragment for Snm2 was cloned into a specialized expression vector, termed pTip-QC1, containing an in-frame N-terminal his-tag. The vector was transformed using electroporation, and the transformed strain was grown as described in the Materials and Methods. Whole cell samples were taken before induction, 2 hours after induction and after overnight induction. The cells were solubilised with SDS, and the solubilised protein was analyzed on 12% SDS-PAGE (Figure 3.13). A large amount of Snm2 was expressed in RHA1. To determine whether Snm2 was expressed in the cytosol, the cells were lysed in a similar manner as E. coli. The lysate was spun at a low speed to remove cell debris and whole cells, and a high speed to remove membrane components. The remaining solution represented the cytosolic fraction of RHA1. To compare all three fractions, both the low spin pellet and high spin pellet were resuspended in Buffer B to the same volume as the cytosolic fraction. A small portion of each fraction was loaded onto SDS-PAGE. Little protein was found in the low spin pellet, indicating Snm2 did not form a large amount of aggregates or inclusion bodies. The results showed the most amount of Snm2 was collected in the cytosolic fraction, indicating the protein was in a soluble form (Figure 3.13). 61  Purification tests were completed to determine the best buffer and condition of purification using IMAC.  62  Figure 3.13 – Expression of Snm2 in RHA1. Snm2-pTip was transformed in RHA1. The strains were induced at O.D.600nm 0.6 with thiostreptone and were grown for three additional hours. (A) One hundred microlitre samples of whole cells were taken before induction (BF), one hour after induction (1h) and two hours after induction (2h). The samples were prepared as described in the Materials and Methods. Twenty microlitres of solubilised whole cells were run on 12% SDS-PAGE. The migration position of Snm2 is indicated to the left of the gel. (B) The remaining cells were spun down at 10,000rpm for 10 minutes in a table top centrifuge and resuspended in 5mL of Buffer B. The cells were lysed, and spun at low speed to remove cell debris and aggregates creating a low spin pellet. They were also spun at a high speed to remove any membranes, creating a membrane pellet. The 5mL supernatant remaining represented the cytosolic fraction (cytosol). The low spin pellet and membrane pellet were resuspended in 5mL of Buffer B. One microlitre, 2µL and 4µL aliquots were taken from the low spin pellet, the membrane pellet and the cytosolic fraction. The samples were run on 12% SDS-PAGE. To confirm the molecular weight, protein standards were run on the right of the gel, the numbers correspond to the kilo-Daltons of each standard. The migration position of Snm2 is indicated to the left.  63  Figure 3.14 – IMAC Purification of Snm2. Snm2 was expressed and induced in RHA1. The cells were lysed in Buffer E. The cytosolic fraction was obtained, and loaded onto a nickel chelating sepharose column. The flow through material was collected. The column was washed with Buffer E including 20mM imidazole. The fractions were eluted with Buffer E including 500mM imidazole. Twenty microlitres of flow through material (FT), the wash material (W), and elution fractions (numbered above the gel) were loaded on 12% SDS-PAGE. Also, 10µL of the cytosolic fraction (Cyto.) was loaded for comparison. An arrow indicating the position of migration of Snm2 is shown.  64  Due to the difficulty of obtaining soluble Snm2, the best condition of purification included ATP in all cell lysis and purification buffers. With ATP, Snm2 was separated from all other cytosolic proteins and was purified using IMAC at greater than 90% purity (Figure 3.14). As Snm2 contains ATP binding regions, the ATP present in the buffer may force the protein to stay in a soluble conformation. However, this conformation may not only be induced by ATP, but may be induced by other proteins, such as Snm1.  3.2.2 Gel Filtration and Native PAGE Analysis of Soluble Snm2 To determine the oligomeric state of Snm2, gel filtration was employed. Purified Snm2 obtained from IMAC was loaded onto a 20mL Superdex 200 gel filtration column. The elution profile showed three peaks eluting (Figure 3.15). The first peak represented any protein aggregates present in the solution, as it eluted at the void volume of the column (7.6mL). The second peak eluted at the apparent molecular weight of 189 kDa (11.0mL). The final peak eluted at the apparent molecular weight of 85 kDa (13.0mL; calculations not shown). Although the theoretical molecular weight of the monomeric form of Snm2 is 64.5 kDa, the inclusion of ATP may have shifted the apparent molecular weight to 85 kDa, corresponding to the third and final peak of the elution profile. Therefore, the second peak corresponding to 189 kDa may represent the dimer of Snm2. These results were complemented with SDS-PAGE analysis of the gel filtration fractions (Figure 3.15). They were also confirmed using Dynamic Light Scattering completed by the Strynadka laboratory (data not shown).  65  Figure 3.15 - Gel Filtration Chromatography of Snm2. (A) Two hundred micrograms of purified Snm2 was loaded onto a Superdex 200 gel filtration column in Buffer E. The protein eluted as three separate peaks. The first peak eluted at the void volume 7.6mL, representing protein aggregates. The second peak eluted at 11.0mL representing dimers of Snm2, and the third peak eluted at 13.0mL representing monomers of Snm2. Five hundred microlitre elution fractions were collected. The protein absorbance was measured in arbitrary milli-absorbance units at A280nm (mAU) as seen on the left of the profile. (B) Twenty microlitres of the relevant elution fractions were loaded onto 12% SDS-PAGE, with the corresponding elution volume above the gel. An arrow indicates the position of Snm2. 66  To investigate the importance of ATP, subsequent gel filtrations of purified Snm2 were completed with various concentrations of ATP. At 5µM ATP only one peak was visualized on the gel filtration profile, corresponding to the molecular weight of the monomeric Snm2. However, if the concentration of ATP was increased, or if 5mM MgCl2 was added to the sample buffer, the dimer and monomer conformation of Snm2 were visualized. Although these observations could not be confirmed by another technique, it does suggest the dimer formation depends on the concentration of ATP in solution as well as the presence of MgCl2 (Figure 3.16). To observe the presence of monomer and dimer Snm2 using a different technique, CNPAGE was utilized. Two bands were expected representing the monomer and dimer of Snm2. However, only one high molecular weight band was visualized (Figure 3.17). CN-PAGE may compact Snm2 into a single band due to its ionic properties (pI 6.4), or due to the absence or fast migration of ATP. Snm2 will not dissociate into its monomeric form under these conditions. To confirm this result, fractions were collected containing only dimeric Snm2 or only monomeric Snm2. If the migration of Snm2 was affected by its ionic properties or the presence of ATP, all conformational states of Snm2 would migrate to the same position. Both dimer and monomer of Snm2 were analyzed by CN-PAGE and migrated at the same position (Figure 3.17). This experiment was completed with and without the presence of ATP in the cathode buffer. In each case, the same result was obtained. For comparison, Blue-Native Polyacrylamide Gel Electrophoresis (BN-PAGE) was employed. In this case, two bands were visualized, one representing Snm2 in monomeric form, and the other representing Snm2 in dimeric form. Therefore, the blue dye must participate in ionic  67  interactions with Snm2 to break apart protein complexes and produce both conformational states (Figure 3.17). Both dimer and monomer forms of Snm2 were tested for their ability to hydrolyzed ATP. Since Snm2 has ATP binding regions and contains similar domains to AAA-ATPases, it was hypothesized the protein would hydrolyze ATP. However, using the Malachite Green Solution to visualize phosphate release, no ATP hydrolysis was recorded. Therefore, the role of ATP in the oligomerization and function of Snm2 is still unclear.  68  Figure 3.16 – Gel Filtration Chromatography of Snm2 with Various ATP Concentrations. (A) Two hundred micrograms of purified Snm2 was loaded onto a Superdex 200 gel filtration column in 50mM Tris (pH 7.9), 300mM NaCl, 10% glycerol and 5µM ATP. The protein eluted as a single peak at 13.0mL representing monomers of Snm2. The protein absorbance was measured in arbitrary milli-absorbance units at A280nm (mAU) as seen on the left of the profile. (B) Two hundred micrograms of purified Snm2 was loaded onto a Superdex 200 gel filtration column in 50mM Tris (pH 7.9), 300mM NaCl, 10% glycerol, 5µM ATP and 5mM MgCl2. The protein eluted as a three separate peaks. The first peak eluted at the void volume 8.0mL, representing protein aggregates. The second peak eluted at 11.0mL representing dimers of Snm2, and the third peak eluted at 13.0mL representing monomers of Snm2. The protein absorbance was measured in arbitrary milliabsorbance units at A280nm (mAU) as seen on the left of the profile.  69  Figure 3.17 – Native-PAGE Analysis of Snm2. One microlitre, 2µL and 4µL of 2.7mg/mL purified Snm2 containing dimer and monomer mix (Snm2 mix) was loaded onto CN-PAGE containing a polyacrylamide 4-12% gradient (pH 8.8). A single migrating band appeared indicated by the left-hand arrow. Snm2 dimers and monomers were purified by gel filtration and loaded onto CN-PAGE (Dimer, Mono.). Both conformations migrated at the same position as the mixture of Snm2. One microlitre, 2µL 4µL, and 8µL samples were taken from 2.7mg/mL solution of purified Snm2. The samples were mixed with 1% 2µL BW Blue dye and loaded onto BN-PAGE containing a polyacrylamide 412% gradient (pH 8.8). Arrows indicating the migratory position of Snm2 dimer and monomer are on the right.  70  3.3 Binding Interactions Between Snm2 and CFP10:ESAT6 3.3.1 Gel Filtration Chromatography of Snm2 with CFP-10 and CFP10:ESAT6 Yeast two hybrid assays have indicated a protein interaction between Snm2 and CFP-10 [10, 28]. Furthermore, an interaction was shown by an in vitro pull down assay between CFP-10 and the immobilized C-terminal region of Snm2 [28]. To characterize the interaction using purified Snm2 and CFP-10, gel filtration chromatography was utilized. However, evidence obtained from circular dichroism suggests CFP-10 obtains its helical secondary structure after interaction with ESAT-6. If CFP-10 is not in its helical structure, it is unknown if the last seven residues sufficient for an interaction with Snm2 would be readily accessible. Therefore, fractions containing the CFP10:ESAT6 complex were also tested for an interaction. Purified Snm2 was incubated with purified CFP-10. Snm2 must be in the presence of ATP to remain soluble. To prevent aggregation of Snm2, CFP-10 was dialyzed in the same purification buffer as Snm2 prior to incubation. Samples were loaded onto a Superdex 75 gel filtration column. For comparison Snm2 and CFP-10 were separately loaded on the same column. The gel filtration profile of Snm2 showed three peaks as previously described. The first peak eluted in the void volume of the column, approximately 8.3mL, and represented protein aggregates. The second peak eluted at approximately 8.8mL, and represented the dimer form of Snm2. The last peak eluted at approximately 9.6mL and represented the monomeric form of Snm2. Purified CFP-10 was analyzed on the same column, and eluted at 11.6 mL. When both CFP-10 and Snm2 were incubated together and loaded on the gel filtration column, three peaks were visualized. The first two peaks represented the aggregates and the 71  dimer (8.5mL), and monomer form of Snm2 (9.6mL). The last peak represented the elution of CFP-10 (11.6mL). This was confirmed by SDS-PAGE analysis of the fractions collected, and by comparison of the three peaks to the elution profiles of the individual proteins. A molecular weight shift was not observed for either protein using gel filtration chromatography (Figure 3.18). Therefore, a complex between Snm2 and CFP-10 was not detected. As mentioned, purified CFP-10 was not in its native folded state due to the absence of ESAT6. To provide evidence of a protein interaction, CFP-10 and ESAT-6 must be in the correct conformational state. Therefore, the same experiment was completed with the CFP10:ESAT6 complex. The complex was dialyzed in the same purification buffer as Snm2, and eluted at 10.6mL on the Superdex 75 gel filtration column. When Snm2 and CFP10:ESAT6 were incubated together, four peaks were visualized. The first three peaks eluted at 8.3mL, 8.8mL and 9.6mL. These peaks corresponded to the same elution volumes as the aggregates, the dimer, and the monomer of Snm2 respectively. The last peak eluted at 10.6mL, corresponding to the elution volume of CFP10:ESAT6. Analysis of the fractions on SDS-PAGE confirmed this result. Therefore, no molecular weight shift was visualized, indicating a complex was not detected between Snm2 and CFP10:ESAT6 (Figure 3.19).  72  Figure 3.18 – Gel Filtration Chromatography of Binding Interactions Between Snm2 and CFP-10. (A) Two hundred micrograms of purified Snm2 in Buffer E was loaded onto the 20mL Superdex 75 gel filtration column. The protein absorbance was measured in arbitrary milli-absorbance units at A280nm (mAU) as seen on the left of the profile. Three peaks eluted from the column. Snm2 aggregates eluted at the void volume of the column (8.3mL), the dimer of Snm2 eluted at 8.8mL, and the monomer of Snm2 eluted at 9.6mL. (B) Purified Snm2 and CFP-10 were incubated together for 10 minutes at 4°C in Buffer E. The protein absorbance was measured in arbitrary milli-absorbance units at A280nm (mAU) as seen on the left of the profile. The mixture was loaded onto the 20mL Superdex 75 gel filtration column. Three peaks were visualized. The first broad peak at 8.5mL contained aggregates and the dimer form of Snm2. The second peak eluted at 9.6mL and contained the monomeric form of Snm2. The last peak eluted at 11mL and contained purified CFP-10. Five hundred microlitre fractions were collected. (C) Twenty microlitres of the elution fractions from the gel filtration of Snm2 with CFP-10 were loaded on 15% SDS-PAGE. The corresponding elution volumes are represented in mL above the gel. 73  Figure 3.19 - Gel Filtration Chromatography of Binding Interactions Between Snm2 and CFP10:ESAT6. (A) Two hundred micrograms of purified Snm2 in Buffer E was loaded on the 20mL Superdex 75 gel filtration column. The protein absorbance was measured in arbitrary milliabsorbance units at A280nm (mAU) as seen on the left of the profile. Three peaks eluted from the column. Snm2 aggregates eluted at the void volume of the column (8.3mL), the dimer of Snm2 eluted at 8.8mL, and the monomer of Snm2 eluted at 9.6mL. (B) Purified Snm2 and CFP10:ESAT6 were incubated together for 10 minutes at 4°C in Buffer E. The mixture was loaded on the 20mL Superdex 75 gel filtration column. The protein absorbance was measured in arbitrary milli-absorbance units at A280nm (mAU) as seen on the left of the profile. Four peaks were visualized. The first peak contained Snm2 aggregates (8.3mL), the second peak contained Snm2 dimers (8.8mL), the third peak contained Snm2 monomers (9.6mL), and the last peak contained purified CFP10:ESAT6 complex (10.6mL). Twenty microlitres of the elution fractions from the gel filtration of Snm2 with CFP10:ESAT6 were loaded on 15% SDS-PAGE, shown below the elution profile. The corresponding elution volumes are represented in mL above the gel. Arrows indicate the position of Snm2, CFP-10, and ESAT-6.  74  3.3.2 Native-PAGE Analysis of Snm2 and CFP10:ESAT6 To further investigate the interaction between Snm2 and CFP10:ESAT6, Native-PAGE analysis was utilized. Various amounts of Snm2 were incubated with CFP10:ESAT6. Snm2, CFP10:ESAT6, and the mixture of Snm2 and CFP10:ESAT6 were analyzed by CN-PAGE. Snm2 and CFP10:ESAT6 were separately loaded on CN-PAGE in their respective purification buffers. They were also incubated and loaded together in either respective purification buffer. This was to ensure the buffer would not have an effect on protein migration. The procedure was completed with the use of 0.5mM ATP present in the electrophoretic running buffer (cathode). A high molecular weight band appeared during the migration of the mixture containing Snm2 and CFP10:ESAT6. This was not seen during the separate migrations of Snm2 or CFP10:ESAT6. It may represent a protein complex of Snm2 with CFP10:ESAT6 or aggregates that migrated through a small portion of the CN-PAGE (Figure 3.20). CFP-10 alone was not tested using this technique because it did not migrate as a sharp band.  75  Figure 3.20 – CN-PAGE Analysis of Binding Interactions between Snm2 and CFP10:ESAT6. CNPAGE contained a linear 4-12% polyacrylamide gradient (pH 8.8). Two microlitres and 4µL of 2.7mg/mL purified Snm2 containing dimer and monomer mix (Snm2) in Buffer E was loaded in the first two lanes. Two microlitre and 4µL of 2.7mg/mL purified Snm2 mix in Buffer D was loaded in the next two lanes. Two microlitre and 4µL of 6.0 mg/mL purified CFP10:ESAT6 in Buffer E was loaded in the first two lanes under the label CFP10:ESAT6. The same amount of CFP10:ESAT6 in Buffer D was loaded in the next two lanes under the label CFP10:ESAT6. Two microlitre of 2.7mg/mL Snm2 mix was incubated with 4µL of 6.0mg/mL CFP10:ESAT6 and 4µL of 2.7mg/mL Snm2 mix was incubated with 2µL of 6.0mg/mL CFP10:ESAT6 in Buffer D. These samples were loaded in the first two lanes under the label Snm2 with CFP10:ESAT6. The same amount of protein was incubated in Buffer E and loaded in the next two lanes under the label Snm2 with CFP10:ESAT6. Arrows indicate the position of Snm2, CFP10:ESAT6 and the possible complex (Snm2 with CFP10:ESAT6?).  76  Figure 3.21 – Binding Interaction of Radiolabeled CFP-10 with Snm2. CFP-10 was radiolabeled using the radioisotope I125. (A) Sixty nanograms of radiolabeled CFP-10 and radiolabeled CFP10:ESAT6 were run on a 15% SDS-PAGE gel. The gel was dried and analyzed using an iodine isotope imager. In the CFP10:ESAT6 complex, only CFP-10 reacted with the iodine and was radiolabeled, as ESAT-6 was not visualized. The arrow to the left indicates the location of radiolabeled CFP-10. (B) A CN-PAGE with a 4-12% polyacrylamide gradient (pH 8.8) was created. Sixty nanograms, 120ng and 180ng of radiolabeled CFP10:ESAT6 and CFP-10 were run in its original purification buffer. One hundred and twenty nangorams and 180ng of CFP10:ESAT6 were incubated with 10µg of Snm2 containing both dimer and monomer in Buffer E. These samples were loaded in the first two lanes under CFP10:ESAT6 + Snm2. One hundred and twenty nanograms and 180ng were incubated with 10µg of Snm2 containing both dimer and monomer in Buffer D, loaded in the next two lanes under CFP10:ESAT6 + Snm2. One hundred and twenty nangorams and 180ng of CFP-10 was incubated with 10 µg of Snm2 containing both dimer and monomer in Buffer E, loaded onto the last two lanes of the gel under CFP10 + Snm2.  77  3.3.3 Binding Interaction of Snm2 and CFP-10 using the Radioisotope I125 In many cases, protein interactions are very weak, and cannot always be visualized by standard techniques. The isotope I125 emits low-energy gamma particles and provides a sensitive detection method of proteins. This would allow a protein to be labeled with a radioactive isotope, and if the radiolabeled protein formed a complex, small amounts of the complex could be detected. To incorporate this isotope into proteins, the iodine is oxidized from I- to I3- which covalently modifies the phenyl ring of tyrosine residues, generating monoand di-iodo derivatives. CFP-10 and ESAT-6 both contain one tyrosine residue. However, if the CFP10:ESAT6 complex was radiolabeled, the tyrosine residue must be accessible by the iodine, and not buried within the helical structure. Analysis of the solution NMR structure of CFP10:ESAT6 revealed the tyrosine residue of CFP-10 was located at the C-terminal region and remained accessible to the outside environment. Therefore, only CFP10 in the CFP10:ESAT6 complex would be radiolabeled. Purified CFP-10 and the CFP10:ESAT6 complex were labeled using the iodine isotope I125. To ensure the proteins were radiolabeled, a small amount of each protein was analyzed by SDSPAGE and visualized using an iodine isotope imager (Figure 3.21). The absence of radiolabeled ESAT-6 confirmed CFP-10 was the only protein with an accessible tyrosine residue in CFP10:ESAT6. Nanomolar amounts of radiolabeled CFP-10 and CFP10:ESAT6 were mixed with microgram quantities of Snm2. CFP-10, CFP10:ESAT6, and the mixture of Snm2 with either CFP10 or CFP10:ESAT6 was analyzed by CN-PAGE using an iodine isotope imager. The addition of Snm2 with CFP-10 or CFP10:ESAT6 did not produce a change in migration of the radiolabeled  78  protein, suggesting a complex was not formed between Snm2 and CFP-10, or could not be visualized using this technique (Figure 3.21).  3.3.4 Interactions between Snm2 and CFP10:ESAT6 using Crosslinking To confirm previously published results of an interaction between Snm2 and CFP-10 in the presence of ESAT-6 shown by an in vitro pull down assay [28], similar conditions were applied to the purified protein. Purified (150pmoles) of Snm2 and purified (500pmoles) of CFP10:ESAT6 were mixed in the presence of the same crosslinker previously used, Dithiobis succinimidyl propionate (DSP), in various amounts and analyzed on SDS-PAGE. To eliminate any non-specific crosslinking protein bands, Snm2 and CFP10:ESAT6 were separately incubated with the crosslinker and analyzed by SDS-PAGE. The results show a single band specific to the mixture of Snm2 and CFP10:ESAT6 migrating at a high molecular weight, suggesting a complex between Snm2 and CFP10:ESAT6. However, to prove this band represents a complex, the crosslinking experiment was repeated using a mixture of radiolabeled and non-labeled CFP10:ESAT6. A high molecular weight band corresponding to a complex of Snm2 with CFP10:ESAT6 was also visualized using this technique, indicating a crosslink between CFP-10 and Snm2 (Figure 3.22). The band is relatively weak, possibly due to competitive binding of the non-labeled CFP10:ESAT6 onto Snm2. Therefore, a crosslink experiment was completed using both radiolabeled Snm2 and CFP10:ESAT6 incubated with non-labeled species of CFP10:ESAT6 and Snm2 respectively. The same molecular weight band appeared as earlier, confirming a crosslink between the two species under radiolabeling conditions. Finally, the crosslink between the two species was tested again using SDS-PAGE. In this case, a firm decrease of CFP-10 and Snm2 is detected corresponding to the increase in high molecular weight species created by the 79  crosslink. No decrease was visualized for ESAT-6, providing further evidence the crosslink is between Snm2 and CFP-10.The high molecular weight band was approximated to be 120 kDa. This may indicate two Snm2 molecules binding onto a CFP10, or possibly Snm2 binding onto more than one CFP-10 molecule (Figure 3.23).  80  Figure 3.22 – Crosslinking Between Snm2 and CFP-10. (A) Ten micrograms of Snm2 was incubated with 0.1mM, 0.2mM and 0.5mM DSP for 5 minutes at room temperature. The samples were loaded onto the first three lanes of 15% SDS-PAGE. Ten micrograms of CFP10:ESAT6 was incubated with 0.05mM, 0.25mM, 0.5mM or 1mM DSP for 5 minutes at room temperature and the samples were loaded under the label CFP10:ESAT6. Finally, to test the interaction between the two proteins 10µg of Snm2 was incubated with 10µg of CFP10:ESAT6 complex and 0.05mM, 0.25mM, 0.5mM or 1mM DSP was added for 5 minutes at room temperature. The samples were loaded under the label Snm2 with CFP10:ESAT6. A control sample included 10µg of Snm2 incubated with 10µg of CFP10:ESAT6 without the addition of DSP. The migration of each protein as well as the possible crosslink between Snm2 and CFP10 is indicated to the right of the gel. (B) One hundred and eighty nanograms of radiolabeled with 5µg of non-labeled CFP10:ESAT6 was incubated with 0.05mM, 0.25mM, 0.5mM or 1mM DSP for 5 minutes at room temperature. Samples were loaded onto a 12% SDS-PAGE gel. One hundred and eighty nanograms of radiolabeled with 5µg of non-labeled CFP10:ESAT6 was incubated with 10µg of Snm2 and 0.05mM, 0.25mM, 0.5mM or 1mM DSP was added for 5 minutes at room temperature. Samples were loaded between lanes 5-9 of the gel. 360ng of radiolabeled with 5µg of non-labeled CFP10:ESAT6 was incubated with 10µg of Snm2 and 0.05mM, 0.25mM, 0.5mM or 1mM DSP was added for 5 minutes at room temperature. Samples were loaded onto the last four lanes of the gel. The possible crosslink between Snm2 and CFP-10 is indicated on the right of the gel, as well as the migration of the radiolabeled CFP-10. 81  Figure 3.23 – Crosslinking Between Radiolabeled and Non-Radiolabeled Snm2 and CFP-10. (A) Ten micrograms of Snm2 was incubated with 0.1mM, 0.2mM and 0.5mM DSP for 5 minutes at room temperature. The samples were loaded onto the first three lanes of 15% SDS-PAGE. To test the binding interaction, 10µg of Snm2 was incubated with 10µg of CFP10:ESAT6 complex and 0.05mM, 0.25mM, 0.5mM or 1mM DSP was added for 5 minutes at room temperature. The samples were loaded under the label Snm2 with CFP10:ESAT6. Five micrograms of Snm2 was incubated with 10 µg of CFP10:ESAT6 and 0.05mM, 0.25mM, 0.5mM or 1mM DSP was added for 5 minutes at room temperature. The samples were loaded in the last four lanes of the gel. The migration of each protein as well as the possible crosslink between Snm2 and CFP-10 is indicated to the left of the gel. To confirm the molecular weight, protein standards were run on the right of the gel, the numbers correspond to the kilo-Daltons of each standard. (B) One hundred and eighty nanograms of radiolabeled Snm2 was loaded onto a 12% SDS-PAGE without the addition of DSP (first lane). One hundred and eighty nanograms of radiolabeled CFP10:ESAT6 was incubated with 5µg of non-labeled Snm2 and 0mM, 0.05mM, 0.25mM, 0.5mM or 1mM DSP was added for 5 minutes at room temperature. Samples were loaded between lanes 2-6 of the gel. One hundred and eighty nanograms of radiolabeled Snm2 was incubated with 5µg of nonlabeled CFP10:ESAT6 and 0.05mM, 0.25mM, 0.5mM or 1mM DSP was added for 5 minutes at room temperature. Samples were loaded onto the last four lanes of the gel. The possible crosslink between Snm2 and CFP-10 is indicated on the left of the gel, as well as the migration of the radiolabeled CFP-10 and Snm2. 82  CHAPTER FOUR 4.1 Discussion and Conclusion Three essential ESX-1 secretion system components investigated in vivo were cloned from M. smegmatis and expressed, in the Gram-negative organism E. coli. Of the three, the transmembrane protein Snm4 and the ATPase Snm2 were purified. However, in the case of Snm1, expression was not possible in E. coli without the inclusion of a 220 amino acid Nterminal GST-tag. Snm1 contains two transmembrane segments within the N-terminal region. The GST-tagged Snm1 expressed the membrane portion of Snm1 as an internal fragment. This could have promoted expression. However, the GST-tag possibly hindered the formation of the natural conformational state of Snm1, causing it to aggregate. Expression of Snm1 was tested in RHA1, but no protein was detected. Therefore, Snm1 may require another protein component for co-expression, or may be expressed only in Mycobacteria species. The transmembrane protein Snm4 was expressed at 37°C in E. coli. Different solubilisation and purification conditions were investigated. However, the same result was obtained; Snm4 could not be purified from various contaminant proteins. The use of anion exchange did not help to separate the contaminant proteins eluted from IMAC. Therefore, the purification of Snm4 was not continued. Different detergents or possibly different expression constructs may enable less contaminant to be solubilised with Snm4 and provide better purification results. Although purification was incomplete, Snm4 was expressed in inner membrane vesicles of E. coli. These vesicles could be used to test for binding interactions with other essential ESX-1  83  secretion system components, in the presence of the CFP10:ESAT6 complex. Also, it may provide a tool to investigate in vitro secretion using components of the ESX-1 secretion system. The expression and purification of Snm2 was first completed in E. coli. However, using this organism and expression conditions, the protein was fairly unstable and insoluble. To produce soluble protein, different expression conditions were investigated. Surprisingly, by using a Gram-positive expression system RHA1, soluble Snm2 protein was produced. This implies the Gram-positive protein synthesis machinery of RHA1 was better equipped to express Snm2 than E. coli. Also, the proteins present in RHA1 may have helped the folding and stability of Snm2. Once expressed and purified from the cytosolic fraction of RHA1, Snm2 only remained soluble in the presence of ATP. The reason for this is unknown, but possibly ATP binding alters the protein conformation to a more soluble state, burying exposed hydrophobic regions. The oligomeric state of Snm2 was further investigated. Gel filtration chromatography indicated two conformational states of Snm2 when 1mM of ATP was present. The molecular weight of each conformational state was calculated by plotting the elution volumes against the known molecular weight and elution volumes of standard proteins. Although monomeric Snm2 was calculated to be 85 kDa rather than the predicted 64.5 kDa, it may have been due to the presence of ATP in the buffer. ATP bound Snm2 may have a higher molecular weight than Snm2 without ATP. To determine the extent of Snm2 solubility, various ATP concentrations were investigated. If gel filtration was conducted with Snm2 in a buffer containing 5µM ATP, only one peak was  84  observed on the elution profile, corresponding to the molecular weight of the monomer. However, if MgCl2 was included in the gel filtration buffer or if the ATP concentration was raised to 100µM ATP, both dimer and monomer were visualized. Although this could not be confirmed by another method, it suggests the Snm2 dimer only forms if enough ATP is present, or if the ATP is stably bound in the active site with the assistance of the Mg+2 ion. The homologues of both CFP-10 and ESAT-6 in M. smegmatis were cloned and co-purified. Various techniques were utilized to ensure a complex formed between CFP-10 and ESAT-6, including gel filtration chromatography and Native-PAGE analysis. These proteins were essential, as Snm2 has previously been shown to interact with CFP-10 in vivo and in vitro. To confirm this interaction, CFP-10 was purified with and without ESAT-6, and both forms were used in subsequent binding interactions with Snm2. Although a binding interaction was expected with CFP-10, none was observed when both purified CFP-10 and Snm2 were incubated together using gel filtration chromatography or radioisotopes on Native-PAGE. This could be due to the conformational state of Snm2. Snm2 is an ATPase, and no hydrolysis of ATP was observed after its purification. Therefore, ATP may be keeping the protein soluble, but at the same time may be promoting a conformational state that would inhibit the formation of protein complexes. Also, Snm2 might need a specific protein interaction or environment to become active and hydrolyze ATP, such as the presence of Snm1. Since these two genes together encode FtsK-similar AAA-ATPase, it might be possible the two proteins require one another to become active. Co-expression of these two proteins in a Gram-  85  positive system may promote expression of Snm1, and help determine the importance of each protein in the function of the ATPase. To show any interaction between the two proteins CFP-10 and Snm2, a crosslinking assay was completed. Previous studies reported a specific interaction between the C-terminal region of CFP-10 and Snm2 in vitro. This interaction was stabilized by the use of a crosslinker [28]. To establish a connection between the two purified proteins, the same crosslinker was utilized and a possible complex was visualized on SDS-PAGE between the two proteins. However, further investigation into the domains Snm2 involved in this crosslink complex is required. Although Snm2 was purified and obtained in a soluble form, many questions have been left unanswered. With regards to Snm2, binding partners are still unknown, as well proper functioning of the protein for ATP hydrolysis. Furthermore, expressed and purified Snm1 may reveal information about the function and location of Snm2 during secretion, whether ATP hydrolysis occurs in the cytosol or at the membrane interface. However, purified Snm2 as well as inner membrane vesicles of Snm4 can provide a tool for further investigations into the mechanism of ESX-1 secretion.  86  4.2 Future Directions An important area of investigation is the role of Snm1 in the ESX-1 secretion system. Although the protein was unable to be expressed in a soluble form within a Gram-negative expression system, co-expression of Snm1 with Snm2 in other expression systems may help create an active ATPase complex. This may also provide information on binding interactions between Snm1 and Snm2. This is the first biochemical evidence provided for purified Snm2. Using this purified protein, the structure can be determined, and the role of ATP in the solubility of Snm2 can be explored. The conditions of oligomerization of Snm2 can be further characterized and investigated. Additional binding partners of Snm2 can be identified using protein-protein interaction techniques. Snm4 expressed within inner membrane vesicles of E. coli can be used for further experimentation on the ESX-1 secretion system. This provides a valuable tool to study secretion in vitro. Purification of Snm4 can be continued using different conditions, and hopefully with purified protein more knowledge can be obtained about secretion through the membrane layer of the bacterium. A complex between Snm2 and CFP-10 was not detected without the use of a crosslink to help stabilize the interaction. Although binding between secretion system components may be transient and weak, further characterization of this interaction may provide additional information on the ESX-1 secretion mechanism.  87  REFERENCES 1. Mahairas GG, Sabo PJ, Hickey MJ, Singh DC, Stover CK. Molecular analysis of genetic differences between Mycobacterium bovis BCG and virulent M. bovis. J Bacteriol. 1996 Mar; 178(5):127482. 2. Pym AS, Brodin P, Majlessi L, Brosch R, Demangel C, Williams A, Griffiths KE, Marchal G, Leclerc C, Cole ST. Recombinant BCG exporting ESAT-6 confers enhanced protection against tuberculosis. Nat Med. 2003 May; 9(5):533-9. 3. Pym AS, Brodin P, Brosch R, Huerre M, Cole ST. Loss of RD1 contributed to the attenuation of the live tuberculosis vaccines Mycobacterium bovis BCG and Mycobacterium microti. Mol Microbiol. 2002 Nov; 46(3):709-17. 4. Lewis KN, Liao R, Guinn KM, Hickey MJ, Smith S, Behr MA, Sherman DR. Deletion of RD1 from Mycobacterium tuberculosis mimics bacille Calmette-Guérin attenuation. J Infect Dis. 2003 Jan 1; 187(1):117-23. 5. Sørensen AL, Nagai S, Houen G, Andersen P, Andersen AB. Purification and characterization of a low-molecular-mass T-cell antigen secreted by Mycobacterium tuberculosis. Infect Immun. 1995 May; 63(5):1710-7. 6. Harboe M, Oettinger T, Wiker HG, Rosenkrands I, Andersen P. Evidence for occurrence of the ESAT-6 protein in Mycobacterium tuberculosis and virulent Mycobacterium bovis and for its absence in Mycobacterium bovis BCG. Infect Immun. 1996 Jan; 64(1):16-22. 7. Berthet FX, Rasmussen PB, Rosenkrands I, Andersen P, Gicquel B. A Mycobacterium tuberculosis operon encoding ESAT-6 and a novel low-molecular-mass culture filtrate protein (CFP-10). Microbiology. 1998 Nov; 144 ( Pt 11):3195-203. 8. van Pinxteren LA, Ravn P, Agger EM, Pollock J, Andersen P. Diagnosis of tuberculosis based on the two specific antigens ESAT-6 and CFP10. Clin Diagn Lab Immunol. 2000 Mar; 7(2):155-60. 88  9. Gey Van Pittius NC, Gamieldien J, Hide W, Brown GD, Siezen RJ, Beyers AD. The ESAT-6 gene cluster of Mycobacterium tuberculosis and other high G+C Gram-positive bacteria. Genome Biol. 2001; 2(10):RESEARCH0044. 10. Stanley SA, Raghavan S, Hwang WW, Cox JS. Acute infection and macrophage subversion by Mycobacterium tuberculosis require a specialized secretion system. Proc Natl Acad Sci U S A. 2003 Oct 28; 100(22):13001-6. 11. Brodin P, Majlessi L, Marsollier L, de Jonge MI, Bottai D, Demangel C, Hinds J, Neyrolles O, Butcher PD, Leclerc C, Cole ST, Brosch R. Dissection of ESAT-6 system 1 of Mycobacterium tuberculosis and impact on immunogenicity and virulence. Infect Immun. 2006 Jan; 74(1):88-98. 12. Abdallah AM, Gey van Pittius NC, Champion PA, Cox J, Luirink J, Vandenbroucke-Grauls CM, Appelmelk BJ, Bitter W. Type VII secretion--mycobacteria show the way. Nat Rev Microbiol. 2007 Nov; 5(11):883-91. 13. Minnikin DE, Kremer L, Dover LG, Besra GS. The methyl-branched fortifications of Mycobacterium tuberculosis. Chem Biol. 2002 May; 9(5):545-53. 14. Bayan N, Houssin C, Chami M, Leblon G. Mycomembrane and S-layer: two important structures of Corynebacterium glutamicum cell envelope with promising biotechnology applications. J Biotechnol. 2003 Sep 4; 104(1-3):55-67. 15. Pallen MJ. The ESAT-6/WXG100 superfamily -- and a new Gram-positive secretion system? Trends Microbiol. 2002 May; 10(5):209-12. 16. Renshaw PS, Panagiotidou P, Whelan A, Gordon SV, Hewinson RG, Williamson RA, Carr MD. Conclusive evidence that the major T-cell antigens of the Mycobacterium tuberculosis complex ESAT-6 and CFP-10 form a tight, 1:1 complex and characterization of the structural properties of ESAT-6, CFP-10, and the ESAT-6*CFP-10 complex. Implications for pathogenesis and virulence. J Biol Chem. 2002 Jun 14; 277(24):21598-603.  89  17. Renshaw PS, Lightbody KL, Veverka V, Muskett FW, Kelly G, Frenkiel TA, Gordon SV, Hewinson RG, Burke B, Norman J, Williamson RA, Carr MD. Structure and function of the complex formed by the tuberculosis virulence factors CFP-10 and ESAT-6. EMBO J. 2005 Jul 20; 24(14):2491-8. 18. de Jonge MI, Pehau-Arnaudet G, Fretz MM, Romain F, Bottai D, Brodin P, Honoré N, Marchal G, Jiskoot W, England P, Cole ST, Brosch R. ESAT-6 from Mycobacterium tuberculosis dissociates from its putative chaperone CFP-10 under acidic conditions and exhibits membrane-lysing activity. J Bacteriol. 2007 Aug; 189(16):6028-34. 19. Smith J, Manoranjan J, Pan M, Bohsali A, Xu J, Liu J, McDonald KL, Szyk A, LaRonde-LeBlanc N, Gao LY. Evidence for pore formation in host cell membranes by ESX-1-secreted ESAT-6 and its role in Mycobacterium marinum escape from the vacuole. Infect Immun. 2008 Dec; 76(12):547887. 20. Gao LY, Guo S, McLaughlin B, Morisaki H, Engel JN, Brown EJ. A mycobacterial virulence gene cluster extending RD1 is required for cytolysis, bacterial spreading and ESAT-6 secretion. Mol Microbiol. 2004 Sep; 53(6):1677-93. 21. Guinn KM, Hickey MJ, Mathur SK, Zakel KL, Grotzke JE, Lewinsohn DM, Smith S, Sherman DR. Individual RD1-region genes are required for export of ESAT-6/CFP-10 and for virulence of Mycobacterium tuberculosis. Mol Microbiol. 2004 Jan; 51(2):359-70. 22. MacGurn JA, Raghavan S, Stanley SA, Cox JS. A non-RD1 gene cluster is required for Snm secretion in Mycobacterium tuberculosis. Mol Microbiol. 2005 Sep; 57(6):1653-63. 23. Massey TH, Mercogliano CP, Yates J, Sherratt DJ, Löwe J. Double-stranded DNA translocation: structure and mechanism of hexameric FtsK. Mol Cell. 2006 Aug; 23(4):457-69. 24. Luthra A, Mahmood A, Arora A, Ramachandran R. Characterization of Rv3868, an Essential Hypothetical Protein of the ESX-1 Secretion System in Mycobacterium tuberculosis. J Biol Chem. 2008 Dec 26; 283(52):36532-41.  90  25. Okkels LM, Andersen P. Protein-protein interactions of proteins from the ESAT-6 family of Mycobacterium tuberculosis. J Bacteriol. 2004 Apr; 186(8):2487-91. 26. Singh A, Mai D, Kumar A, Steyn AJ. Dissecting virulence pathways of Mycobacterium tuberculosis through protein-protein association. Proc Natl Acad Sci U S A. 2006 Jul 25; 103(30):11346-51. 27. Teutschbein J, Schumann G, Möllmann U, Grabley S, Cole ST, Munder T. A protein linkage map of the ESAT-6 secretion system 1 (ESX-1) of Mycobacterium tuberculosis. Microbiol Res 2006; (2007), doi:10.1016/j.micres.2006.11.016. 28. Champion PA, Stanley SA, Champion MM, Brown EJ, Cox JS. C-terminal signal sequence promotes virulence factor secretion in Mycobacterium tuberculosis. Science. 2006 Sep 15; 313(5793):1632-6.  91  APPENDIX  Standard Curve S75 Gel Filtration 6 Void Volume  5.8 5.6  Snm2 (dimer)  log (MW)  5.4  Snm2 and CFP10  5.2 Snm2 (monomer)  5  Standard Curve  4.8 4.6 CFP10:ESAT6 4.4 4.2  CFP-10  4 8  8.5  9  9.5 Elution Volume (mL)  10  10.5  11 y = -0.6319x + 11.083  92  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0067093/manifest

Comment

Related Items