Figures
Abstract
Background
H37Rv and H37Ra are well-described laboratory strains of Mycobacterium tuberculosis derived from the same parental strain, H37, that show dramatically different pathogenic phenotypes.
Methodology/Principal Findings
In this study, the transcriptomes of the two strains during axenic growth in broth and during intracellular growth within murine bone-marrow macrophages were compared by whole genome expression profiling. We identified and compared adaptations of either strain upon encountering an intracellular environment, and also contrasted the transcriptomes of the two strains while inside macrophages. In the former comparison, both strains induced genes that would facilitate intracellular survival including those involved in mycobactin synthesis and fatty acid metabolism. However, this response was stronger and more extensive for H37Rv than for H37Ra. This was manifested as the differential expression of a greater number of genes and an increased magnitude of expression for these genes in H37Rv. In comparing intracellular transcriptional signatures, fifty genes were found to be differentially expressed between the strains. Of these fifty, twelve were under control of the PhoPR regulon. Further differences between strains included genes whose products were members of the ESAT-6 family of proteins, or were associated with their secretion.
Conclusions/Significance
Along with the recent identification of single nucleotide polymorphisms in H37Ra when compared to H37Rv, our demonstration of differential expression of PhoP-regulated and ESX-1 region-related genes during macrophage infection further highlights the significance of these genes in the attenuation of H37Ra.
Citation: Li AH, Waddell SJ, Hinds J, Malloff CA, Bains M, Hancock RE, et al. (2010) Contrasting Transcriptional Responses of a Virulent and an Attenuated Strain of Mycobacterium tuberculosis Infecting Macrophages. PLoS ONE 5(6): e11066. https://doi.org/10.1371/journal.pone.0011066
Editor: Anil Kumar Tyagi, University of Delhi, India
Received: February 16, 2010; Accepted: April 29, 2010; Published: June 10, 2010
Copyright: © 2010 Li et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was funded by the Canadian Bacterial Diseases Network, Networks of Centres of Excellence, the British Columbia Lung Association, and the British Columbia TB Vets Foundation. R.W.S. is the recipient of a Child and Family Research Institute Investigatorship Award. A.L. was supported by a Hardwick Fellowship from the Department of Pathology, University of British Columbia and a graduate fellowship from the Child and Family Research Institute. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Mycobacterium tuberculosis (Mtb) is a bacterium that can be spread via aerosols generated by infected individuals when they sneeze, cough, or even speak. Inhaled organisms are deposited into alveoli where one outcome is that the bacterium encounters and is ingested by alveolar macrophages [1]. Mtb has strategies that enable its survival and replication within the host macrophage, though variations in the pathogenicity of strains have been found [2]. Thus, a comparison of the genomic makeup of a virulent (H37Rv) and an attenuated (H37Ra) strain of Mtb should identify virulence genes. Furthermore, comparing gene expression profiles of these strains during their interaction with macrophages should shed light on the genes that enable intracellular survival and the microenvironments encountered during infection.
Here, we have compared the transcriptomes of broth grown and intracellular H37Rv and H37Ra Mtb. These two strains were derived from the parental strain H37 Mtb that was isolated from the sputum of a patient suffering from chronic pulmonary tuberculosis in 1906 [3]. For most of the next two decades H37 became a widely used laboratory strain of Mtb and was shown to be highly virulent for guinea pigs and only moderately so for rabbits – a characteristic described as marking the bacterium as human in type [3]. Subsequently, two daughter strains were identified of which H37Rv maintained its virulence whereas H37Ra did not. However, it should be emphasized that H37Ra is not completely avirulent, only attenuated in its virulence compared to H37Rv. In common animal models for TB such as mice and guinea pigs, H37Ra does establish an infection, and does multiply within these hosts [4], [5], [6].
On the assumption that the attenuated variant must necessarily exhibit alterations to either the genome or the expression of virulence genes compared to the virulent variant, many studies have been conducted to assess differences between Mtb H37Ra and H37Rv in order to find mycobacterial virulence factors [7], [8], [9], [10], [11], [12], [13]. Studies comparing the gene expression differences between broth-grown H37Ra and H37Rv or intracellular H37Rv versus broth-grown bacteria have identified differentially expressed genes, some of which have been shown to have roles in mycobacterial virulence [14], [15], [16]. Using microarray technology, Gao et al, examined genes involved in the cording and non-cording phenotypes of the respective H37Rv and H37Ra and identified twenty-two genes differentially regulated between them [14]. Using differential display, Kinger and Tyagi examined genes differentially expressed between aerobic cultures of Mtb H37Ra and H37Rv and identified the two component system devR/devS (also known as dosR/dosS) [15]. This regulatory system has subsequently been observed to aid Mtb survival and growth under anoxic conditions [17], [18], [19]. Thus, these studies highlight the importance of the H37Rv/H37Ra model in the search for genes with significant roles in mycobacterial pathogenesis.
Microarray technologies are widely accepted tools used to ask genome-wide questions and numerous studies investigating mycobacterial pathogenesis have successfully utilized these platforms. [19], [20], [21], [22]. However, while there have been microarray studies examining intracellular bacteria [23], [24], [25], none have compared the intracellular transcriptomes of H37Ra and H37Rv, strains that are well documented for their disparities in virulence. Additionally, we sought to examine not just the differences between the two strains inside a host cell, but also the adaptations made by Mtb upon encountering a macrophage. For our specific purposes, we chose the bone-marrow-derived macrophage for their reproducibility and similarities to the alveolar macrophage [26], a presumed natural niche for Mtb upon infecting the host lung. Previously, our laboratory had used an alternative genome analytical method, bacterial artificial chromosome fingerprint arrays (BACFA), to examine intracellular transcriptomes of Mtb [27]. Using BACFA, we were able to identify an operon of genes (frdABCD) encoding fumarate reductase, which appeared to be important for intracellular survival of the tubercle bacterium. Thus, we sought to extend our previous findings to compare the gene expression patterns of intracellular H37Rv and H37Ra using whole genome transcriptional profiling to further identify genes that may influence the outcome of mycobacterium-macrophage interactions.
Materials and Methods
Ethics Statement
All use of animals was approved by the University of British Columbia Committee on Animal Care in accordance with the Canadian Council on Animal Care recommendations.
Bacterial Strains
Broth cultures of Mtb for RNA extraction were grown in 7H9 liquid medium supplemented with 10% oleic acid, albumin, dextrose complex (OADC) and 0.05% Tween-80. Cultures were grown in 300 mL roller bottles at 3 rpm until mid-logarithmic phase (day 4). For bacteria used to infect macrophages, Mtb cultures were grown in Proskauer and Beck media supplemented with 0.05% Tween-80 in 100 mL roller bottle cultures until mid-logarithmic phase. Aliquots were frozen down and thawed as needed for infections.
Isolation and infection of bone marrow-derived macrophages (BM-MΦ)
BM-MΦ were obtained from the femora, tibiae, and humeri of CD-1 mice as previously described [28], [29]. Monolayers were incubated with a multiplicity of infection (MOI) of 10 bacteria to 1 MФ in binding medium (138 mM NaCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4, 2.7 mM KCl, 0.6 mM CaCl2, 1.0 mM MgCl2 and 5.5 mM D-glucose) for four hours resulting in an average uptake of approximately 0.1 bacteria per MΦ. Following this incubation, the monolayers were washed three times with pre-warmed media to rinse off unbound bacteria. One group of coverslips were immediately processed and the others were submerged again in cRPMI (RPMI 1640 with 10% FCS, 2 mM L-glutamate, and 1 mM sodium pyruvate) and incubated at 37°C. The remaining groups were processed at days 4 and 7 post-infection; for these groups, both cover slips and supernatants were processed in order to include all viable bacteria within the closed system [30]. Samples to be processed were briefly sonicated (10 s using a VC50T 50 W+3 mm probe (Sonics & Materials, Danbury, CT) tuned as per manufacturer's recommendations) to release and disperse intracellular bacteria, and plated on 7H10 agar plates supplemented with 10% OADC to determine colony forming units (CFU) per well. For all macrophage experiments, three replicate cover slips were assessed at each time point for each of three independent experiments. Statistical significance of comparisons between H37Ra and H37Rv or between two time points was determined using a two-tailed, unpaired Student's t test. P values <0.05 were marked as significant.
Infection of MФ for intracellular bacteria RNA extraction
3×107 macrophages were seeded into a 150 cm2 tissue culture flask and infected at a MOI of 10 bacteria to 1 MФ in binding medium. After a four hour incubation, monolayers were washed three times with pre-warmed media and then submerged in 30 mLs of cRPMI. At day 5, 25 mLs of cRPMI were added to feed the monolayer. At day 7, 50 mLs of 5 M GTC lysis buffer (per litre: 5 M guanidium isothiocyanate, 7 mL beta-mercaptoethanol [14.3 M, MP Biomedicals], 3.5 mL Tween-80, 0.25% sodium lauryl-sulfate, and 25 mM tri-sodium citrate) were added to the contents of each flask (monolayer and medium). Bacteria were recovered by centrifugation, before Trizol extraction. DNase treatment and purification of extracted RNA was performed as previously described [27].
Comparative Genomic Hybridization
Genomic DNA was extracted from the bacteria using previously published protocols by Belisle et al [31]. Two micrograms each of Mtb H37Ra and H37Rv genomic DNA were hybridized to TBv2.1.1 M. tuberculosis complex whole genome microarray, generated by the Bacterial Microarray Group at St. George's, as previously described [32] DNA extracted from two biological replicates were hybridized in duplicate. The array design is available in BμG@Sbase (accession number: A-BUGS-23; http://bugs.sgul.ac.uk/A-BUGS-23) and also ArrayExpress (accession number: A-BUGS-23).
Transcriptome Comparison
RNA was extracted and purified from the bacteria using a differential lysis method developed by Mangan et al. [33], from both axenic broth cultures of bacteria as well as intracellular bacteria at day 7 post-infection. Microarray hybridizations were conducted as previously described [32]with 4 µg Cy5-labelled cDNA derived from either H37Rv or H37Ra RNA hybridized against 2 µg Cy3-labelled M. tuberculosis H37Rv genomic DNA. RNA extracted from three biological replicates of each strain were hybridized in duplicate.
Microarray data analysis
The hybridized slides were scanned sequentially at 532 nm and 635 nm corresponding to Cy3 and Cy5 excitation maxima using the Affymetrix 428™ Array Scanner (MWG). Comparative spot intensities from the images were calculated using Imagene 5.5 (BioDiscovery), and imported into GeneSpring GX™ 7.2 (Agilent Technologies) for further analysis. The array data were normalized to the 50th percentile of all genes detected to be present on the array and further normalized to the mean of specific samples depending on the comparison e.g. H37Rv in broth versus H37Ra in broth. The dataset was filtered to include only genes called present or marginal (>2 fold signal/background in either Cy3 or Cy5 channels) in at least 60% of hybridizations. Significantly differentially expressed genes were identified using a t-test (P<0.05) with Benjamani and Hochberg multiple testing correction and further filtered to include genes with >1.5 fold change. For genomic comparisons, H37Rv and H37Ra hybridizations were compared using Arraypipe (http://koch.pathogenomics.ca/cgi-bin/pub/arraypipe.pl). Genes were filtered for 1.5-fold difference and then analyzed via ANOVA (P<0.05). The hypergeometric distribution was used to determine if previously published functional categories of genes were significantly enriched in the comparisons. MIAME-compliant raw and fully annotated microarray data has been deposited in BμG@Sbase (accession number: E-BUGS-86; http://bugs.sgul.ac.uk/E-BUGS-86) and also ArrayExpress (accession number: E-BUGS-86).
Quantitative RT-PCR analysis of selected candidate genes identified by microarray analysis
One to four micrograms of total RNA were reverse transcribed by priming with random hexamers, with the following added to each tube of total RNA: 5 µL 5X SSII buffer, 2.5 µL 0.1 M DTT, 2.3 µL dNTPs (5 mM dA/G/C/TTP), 1.7 µL water, and 2.5 µL Superscript II. Gene specific primers (Table S5) were used with SYBR Green reagent (Finnzymes) to assess expression of candidate genes. qRT-PCR conditions were as follows: 95°C for 10 minutes followed by 35 cycles of 94°C for 30 seconds, 57°C for 20 seconds, and 72°C for 30 seconds. qRT-PCR data is presented as fold difference of expression in H37Ra relative to H37Rv using the 2−ΔΔCt method as described previously [34] with a 71 bp product (henceforth called rrnAP1) from the promoter region of rrnS (Rvnr01) in Mtb [35] and dnaK used as normalizing genes. Two normalizing genes were used for each time point as it has been shown that normalizing genes do not necessarily stay constant over time, and that utilizing more than one control gene is required for reliable expression analysis [36]. Neither rrnAP1 nor dnaK were significantly differentially expressed, as determined by microarray analysis, in intracellular or H37Rv/Ra comparisons in this study.
Results
The strains H37Rv and H37Ra of Mtb are highly related, and yet, their growth rates inside macrophages and ability to cause disease are significantly different. We, therefore, sought to identify genes involved in pathogenesis by comparing the gene expression patterns of the two strains. Previously published transcriptomic profiles of H37Rv and H37Ra have focused on axenic broth comparisons, or on differences in gene expression between broth grown and intracellular H37Rv. Our study confirms and extends these reports by comparing directly the respective transcriptomes of intracellular H37Rv and H37Ra.
1. Contrasting broth grown H37Rv and H37Ra
Comparing genomic differences between the strains using CGH microarray analysis confirmed the recently reported sequence similarity [37], [38] along with the RvD2 region of difference present only in H37Ra (Table 1). The growth rate of H37Rv and H37Ra in broth did not differ significantly over the course of seven days (Fig. 1A). In addition, the doubling times calculated at mid-logarithmic phase (4 days) were very similar between strains (H37Ra 22.7±0.7 h, H37Rv 22.2±0.5 h; P>0.05). Therefore, gene expression profiling of axenic broth cultures of H37Rv and H37Ra was employed to identify genes that may be constitutively differentially expressed between H37 variants in mid-logarithmic phase (Table S1). Our results correlate well with previous studies that examined cording phenotypes of H37Rv and H37Ra as well as genes under control of the phoP regulon. Specifically, we observed significant overlap of genes under-expressed in H37Ra relative to H37Rv with those found by Gao et al. [14] (hypergeometric p-value 1.77×10−3), as well as genes under the regulation of phoP found by Walters et al. [39], hypergeometric p-value 1.27×10−8). Of particular interest among the 436 genes identified as differentially expressed between the strains were pks3 and fadD21, a pks3,4-associated fatty acid activating enzyme involved in polyacyltrehalose (PAT) and 2,3-di-O-acyltrehalose (DAT) synthesis [40] that were repressed in H37Ra compared to H37Rv. pks3 and pks4 are regulated by phoP, and phoP mutants in Mtb have a deficiency in cord factor synthesis, and an absence of PATs, DATs, and sulpholipids [38], [41].
A) Mtb H37Ra and H37Rv were grown in 7H9 enriched media in aerated roller bottle cultures over 7 days. The graph shows the means ± SEM calculated from three independent experiments performed in duplicate. RNA for microarray experiments was extracted from both cultures at day 4. B) BM-MΦ were infected with Mtb H37Ra (grey bars) or H37Rv (black bars). Four hours after infection, unbound bacteria were rinsed off and intracellular bacteria were enumerated. Both strains had been internalized equally (ca. 104 bacteria per 106 macrophages). Bacterial CFU at days 4 and 7 post-infection were assessed and are plotted as Log10 change from day 0. The means ± SEM calculated from three independent experiments (each performed in triplicate) are shown. Statistical significance was determined using the student's t-test: *P = 0.007, **P = 0.0041. Mtb RNA was extracted from this macrophage infection model at day 7.
2. Intracellular Mtb gene expression signature
To understand the response of H37Rv and H37Ra to an intracellular environment, the transcriptional profile of axenic broth grown bacilli was compared to the respective gene expression pattern derived from bacilli within murine macrophages. During macrophage infection, 81 genes were identified to be significantly induced by both H37Rv and H37Ra, along with 70 that were repressed relative to axenic broth grown bacilli (Table S2). Mtb genes induced within the intracellular environment were involved in lipid metabolism (e.g. fadD33, fadE5, Rv3229c, Rv1344), intermediary metabolism and respiration (e.g. Rv1463, icl), transcription regulation (e.g. Rv1460, Rv139, sigB), and response to oxidative stress (e.g. ahpC, ahpD), These data correlated well with previous reports of Mtb gene expression in murine macrophages (Schnappinger et al. [23]) (hypergeometric function p = 1.98×10−17), or human monocyte-derived macrophage and dendritic cells (Tallieux et al. [24]) (hypergeometric function p = 1.97×10−17). These commonalities reveal a set of genes, irrespective of macrophage phenotype or bacterial strain, which, presumably enable the bacterium to survive within a host cell.
Genes involved in fatty acid metabolism were induced in intracellular bacteria, particularly by the virulent H37Rv. Whereas a subset of these genes were significantly induced by both intracellular H37Ra and H37Rv (fadD2, 13, 19, 26, 33; fadE5, 14, 21, 23, Table S2), the virulent strain induced an additional seven fatty acid-CoA synthases (fadD15, 21, 25, 28, 29, 30, 31), an acyl-CoA dehydrogenases (fadE5), an enoyl CoA hydratase (echA6), and three acetyl CoA transferases (fadA2,3,4). The upregulation of these genes reflects the utilization of β-oxidation in fatty-acid metabolism and, as suggested previously [23], the induction of multiple genes within these large gene families may point to the requirement of different isoenzymes to catabolize a variety of different fatty acids. However, five of the fadD genes (fadD21, 26, 28, 29, and 30) are those previously found to encode for fatty-acyl CoA ligases involved in complex lipid synthesis rather than lipid oxidation, which may reflect the need to maintain or modify the cell wall architecture in a challenging environment [42].
Members of the mbt gene cluster (mbtA-J) which encode for components necessary for mycobactin biogenesis [43] were also differentially represented between intracellular and broth-grown Mtb; mbtB,D,E,F,H and mbtI were expressed at higher levels by intracellular bacteria. Mycobactins are mycobacterial siderophores that act as iron chelators [44]. Iron is often limited during infection and microorganisms have developed strategies, such as the production of siderophores, to maximize iron acquisition. The induction of mycobactin synthesis corresponds with the presumably decreased availability of iron inside the phagosome as compared to broth cultures.
3. Comparison of intracellular H37Rv and H37Ra profiles reveals differential expression of genes involved in transcriptional regulation and pathogenesis
Bacterial growth (Fig. 1B) and the doubling times of intracellular bacteria differed significantly between the two strains at both 4 days (H37Ra 51.5 h±1.7 h, H37Rv 32.8 h±0.9 h; P = 0.0007) and 7 days post infection (H37Ra 42.1 h±1.6 h, H37Rv 31.3 h±0.8 h; P = 0.0041). Direct comparison of the transcriptional profiles of Mtb H37Ra and H37Rv at 7 days post-infection revealed 50 genes that were significantly different in their expression between the two strains (Tables 2 and 3, Figure 2). Of these, 15 genes were over-expressed in H37Ra compared to intracellular H37Rv (Table 2, Figure 2), and 35 genes were under-expressed in the attenuated strain (Table 3, Figure 2). Additionally, the differences in transcriptional response of H37Rv and H37Ra to the intracellular environment were plotted by chromosome position using a moving average to identify clusters of genes differentially expressed between the two strains inside the macrophage (Figure 3, Table S3). This divergence in gene expression not only provides insight into genes that may influence outcomes of infection, but also highlights the polymorphisms uncovered in recent sequence comparisons of H37Rv and H37Ra that probably affect mycobacterial pathogenesis.
Fifty genes differentially expressed between intracellular H37Rv and H37Ra at significant levels are clustered relative to the gene expression pattern derived from each respective broth culture. H37 variants (in 6 replicate hybridizations) are marked as columns, genes as rows. Genes colored in red are induced after macrophage infection compared to respective axenic broth growth; genes colored in blue are repressed. Filled bars mark clusters of genes expressed to a greater level by H37Ra inside macrophages. Open bars mark genes induced intracellularly by H37Rv that are expressed at a lower level by H37Ra inside the macrophage.
The difference between intracellular transcriptional profiles of H37Ra and H37Rv were plotted onto the chromosome using a moving average. Seven clusters of genes differentially expressed between the two strains inside murine macrophages were identified (marked with asterisks 1–7). Y-axis details differential expression as fold change (H37Ra relative to H37Rv); x-axis describes chromosome location as Rv number. Clusters 1 and 4 were expressed at higher levels intracellularly by H37Ra relative to H37Rv, and clusters 2,3,5,6,7 were expressed at lower levels intracellularly by H37Ra relative to H37Rv. Clusters 2, 4, 5 and 7 contain genes associated with the PhoP regulon, and cluster 6 contains genes (Rv3614c-3616c) involved in ESAT-6 secretion.
Genes over-expressed in H37Ra indicative of an enhanced stress response.
Of the genes induced in the attenuated Mtb H37Ra, two were hypothetical proteins (Rv1991c, Rv0637), one encoded for a component of the isoniazid inducible protein (iniB,), and one for a probable ferredoxin (fdxC). Both desA1 and desA2, encoding a protein desaturase that is involved in fatty acid biosynthesis and mycolic acid biosynthesis, were also expressed at a higher level in H37Ra compared to H37Rv, suggesting a possible compensatory mechanism to produce desaturated fatty acids to replace lipid moieties lacking in H37Ra [45], [46].
Examining chromosomal gene clusters over-represented in intracellular H37Ra (clusters #1 and #4 – Figure 3, Table S3), we note genes involved in transcriptional regulation (Rv1990c, Rv1994c) as well as genes induced under antibiotic stress (iniB, iniA, iniC). This suggests that the intracellular environment encountered is more stressful to H37Ra bacilli than it is to H37Rv bacilli 7 days after infection of murine macrophages.
Genes involved in ESX-1 secretion are under-expressed in intracellular H37Ra.
One of the gene clusters (cluster #6 Fig. 3), containing adjacent genes Rv3616c to Rv3612c, was expressed at a lower level in intracellular H37Ra compared to intracellular H37Rv. Two members of which (Rv3614c and Rv3615c) have been designated a part of the ESX-1 secretion system [47], [48]. Additionally, we also observed a decrease in expression of Rv3616c (EspA) in intracellular H37Ra compared to H37Rv by qRT-PCR analysis (Figure 4A). While not in this gene cluster, our microarray data also identified Rv3849 (espR), a recently characterized secreted regulator of ESX-1 [49], as under-expressed in H37Ra. Other genes related to the ESAT-6 family that were expressed at lower levels in intracellular H37Ra compared to H37Rv included PE13 (Rv1195), PPE18 (Rv1196) and Rv1198. It is interesting that PE13 and PPE18 were under-represented along with Rv1198 as it has been previously reported that genes of the ESAT-6 family are often clustered with PE, PPE genes [43], [50]. These ESAT-6 chromosomal arrangements are often flanked by conserved hypothetical proteins, and it has been proposed that such genomic arrangement may encode a secretory apparatus for the ESAT-6-like proteins [43], [50].
Quantitative RT-PCR was used to confirm the expression of genes over or under-expressed in intracellular H37Ra compared to intracellular H37Rv. A) Highlighting the pks3/4 (Rv1180-82) and ESAT-6-like (Rv3613-16c) gene clusters under-expressed in H37Ra compared to H37Rv, and iniB, phoP and desA over-expressed in H37Ra vs H37Rv; and B) genes that were induced by intracellular bacteria, but whose expression did not significantly differ between strains. Data are means ± SEM calculated from three independent biological samples analyzed in triplicate.
Members of the phoP regulon are differentially expressed between intracellular Mtb H37Ra and H37Rv.
Twelve members of the phoP regulon [39] (pks3, pks4, mmpL8, mmpL10, papA3, fadD21, PPE18, PPE19, Rv1639c, Rv2376c, PE31, and PPE60) were under-expressed in intracellular H37Ra compared to H37Rv. Contrasting these data to a study comparing the transcriptomes of broth grown H37Rv and a phoP knockout mutant [39], identified 12 genes regulated by phoP that were under-expressed in intracellular H37Ra versus H37Rv (hypergeometric pvalue = 3.78×10−31). Several of these genes are involved in the synthesis of cell envelope constituents that may explain the different lipid profiles between the two strains [51]. In addition, comparing our data to that of Gao et al. [14], we found 7 genes implicated in the cording phenotype to be under-expressed in intracellular H37Ra (hypergeometric pvalue = 6.55×10−10).
Taken together, the above results suggest a connection between the ESX-1 secretion system and members of the phoP regulon [52]. Combining the fifty differences between the intracellular bacteria with changes in gene expression resulting from the move to an intracellular environment within the macrophage phagosome, i.e. examining how the fifty genes differentially expressed intracellularly between H37Ra and H37Rv are regulated inside the macrophage versus axenic broth culture, we are able to delineate which genes virulent strains of Mtb, like H37Rv, induce within an intracellular environment. We suggest that this is representative of a successful infection (Figure 2, open bars). The observation that not only does H37Ra over-express only a small subset of these fifty genes (Figure 2, filled bars), but also represses the expression of others, highlights the importance of both ESX-1 and phoP-regulated genes during a successful infection by virulent bacilli.
4. qRT-PCR validation of microarray data
Expression changes determined by microarray analysis were confirmed by qRT-PCR of three groups of genes: over-expressed (iniB, phoP, desA), or under-expressed (Rv3822, papA3, Rv3615c), in H37Ra when compared to H37Rv (Fig. 4A); and genes induced in intracellular bacteria but unchanged between strains (Rv3839, Rv0282, icl, Fig. 4B). Additional genes were also included in the qPCR analysis to establish correlations between expression of genes that appear to be located in operons on the Mtb chromosome. These included members of the PhoP regulon (pks3, pks4, mmpL10), and the cluster of genes (Rv3613c-3616c) previously characterized to be important for intracellular survival and for interactions with products from the RD1 region [48], [53]. qRT-PCR confirmed the differential regulation of these genes in intracellular H37Ra versus H37Rv (Fig. 4A).
5. Transcriptomics and single nucleotide polymorphisms
Recent publication of the genomic sequence of H37Ra identified 57 genes that may be affected by SNPs in H37Ra (compared to H37Rv) [37]. Of these, we found that three were over-expressed in H37Ra compared to H37Rv in an intracellular environment (phoP, Rv0637, and PPE38), and one under-expressed: (PPE18). The differential expression of these genes highlights the importance of these 4 genes over the other 53 genes in mediating the attenuated phenotype of H37Ra inside the host. Protein-protein network associations were mapped for these four genes using STRING (string-db.org [54]) in an attempt to identify processes important for intracellular survival that may be affected by SNPs in H37Ra (Figure S1).
PPE18 (Rv1196), under-expressed in H37Ra compared to H37Rv and a member of the PhoPR regulon, has been found to induce interferon-gamma production in infected or previously immunized calves [55]. Examining known protein-protein associations, Rv3616c (MT3718 or EspA) and EspB (Rv3881c) both appear to associate with PPE18. Both EspA and EspB are required for the secretion of ESAT-6 [48], [56]. The expression of espA was lower in H37Ra relative to H37Rv in both broth and inside the macrophage, whereas the expression of espB differed only between broth-grown bacteria. Interestingly, when modeling PPE38 (induced in intracellular H37Ra over H37Rv), we find associations with several members of the RD1 region (Rv3875-3878), with Rv3875 induced in broth-grown H37Ra. PPE38 is not a known member of the PhoPR regulon, and thus, these associations, if proven correct, may indicate a PhoP-independent pathway of ESX-1 secretion.
Rv0637, induced in H37Ra compared to H37Rv intracellularly, encodes a subunit of hydroxyacyl-ACP dehydratase with a role in fatty acid synthesis [43]. Examining its protein-protein associations, we find both InhA and Rv0289. InhA is an NADH-dependent enoyl (acyl-carrier protein) reductase whose role in mycolic acid synthesis is well characterized [57]. Rv0289, on the other hand, encodes EspG3, an ESX-1 secretion associated protein [58]. However, neither of these genes shows differences in gene expression. Genes that are both associated with Rv0637 and are induced in broth-grown H37Ra include Rv0636 (fatty acid synthesis), rpmG2 (ribosomal protein Rv0634B), and the conserved membrane protein Rv3587c.
Lastly, phoP, the DNA-binding domain of the PhoPR two-component regulatory system that appears to be involved in bacterial virulence in other organisms [59], [60], was found in this microarray study to be induced in both broth-grown and intracellular H37Ra versus H37Rv (Table 2 and Fig. 4). Protein-association mapping of PhoP leads to several two-component systems (MtrB, TrcS, PrrB, TcrY), one of which (TcrY) was induced in broth-grown H37Ra. These associations suggest that regulation is very finely coordinated between several regulatory systems, and minor changes to one may have profound effects on the cell. Recent genomic sequencing of Mtb H37Ra has shown that the sequences of phoP in H37Ra and H37Rv differ by one base-pair: cytosine to thymine at position 656. Additionally, it was suggested [38] that this SNP may actually affect the binding of, and regulation by, PhoPR, due to steric hindrance resulting from the change in primary amino acid sequence from serine to leucine at position 219. Sequencing a 200-bp fragment of the phoP gene in the H37Ra and H37Rv strains utilized in our laboratory, confirmed this SNP to be present in the H37Ra used to generate the expression data presented here (data not shown). The atomic level mapping of PhoP has been previously published by Wang et al. [61] and Gonzalo-Asensio et al [62]. We further corroborate these models by using an alternative approach by modeling the differences between H37Ra and H37Rv at the molecular level. Using DeepView (http://www.expasy.ch/spdbv/) to generate three-dimensional models of the PhoP molecules found in the respective strains (Figure 5), the substitution of leucine for serine does appear to change the structure of the protein, and the extra side group may indeed affect binding of PhoP.
Using DeepView (Swiss-PDB viewer: http://www.expasy.ch/spdbv/), a 3D depiction of the PhoP molecules from H37Rv (top panel) and H37Ra (bottom panel) were generated including the SNP that results in an amino acid change (from serine to leucine) in the H37Ra PhoP molecule. This substitution is boxed and may result in steric hindrance of binding events with the PhoP molecule.
Discussion
This study was initiated to investigate the reasons why an attenuated strain of Mtb, H37Ra, grows significantly less well in macrophages than does a related virulent strain, H37Rv. Initially, genomic differences between H37Ra and H37Rv were assessed using microarrays to complement our previous attempts at isolating genomic differences between these strains via two-dimensional gel electrophoresis [63]. Our experiments were in agreement with previous studies that described few major genomic insertion or deletion events between the two strains, other than the RvD2 region of difference. Other differences previously described include restriction site differences, and more recently, base-pair differences [11], [37], [38]. We did not detect point mutations affecting restriction digestion patterns or those identified by genome sequencing [37]. As a single base-pair alteration may not affect annealing of sequences during microarray analysis, these subtle genomic differences were not identified.
The next element of the study surveyed gene expression patterns of the two strains of Mtb grown in enriched broth as a means to identify any inherent transcriptional differences between strains. We identified differential expression of phoP and genes relating to the cording phenotype [14], which are themselves under the regulation of phoP [39]. PhoP is the transcriptional regulator of the two-component system, PhoPR, which is an important regulator of genes with potential roles in mycobacterial virulence.
The transcriptomes of intracellular bacteria, after uptake by murine macrophages, were then compared to their broth-grown equivalents to elucidate differences that may arise due to adaptation to the intracellular compartment. Genes involved in fatty acid metabolism were upregulated by both H37 variants after macrophage infection, along with genes that may protect the bacterium against host defenses. Additionally, genes whose products are involved in aerobic respiration were repressed by intracellular bacteria, probably due to a decreased reliance on aerobic respiratory pathways inside the macrophage. Correspondingly, genes that encode products providing alternative means of biosynthesis or respiration were induced in intracellular bacteria compared to broth-grown bacteria. We compared our data to that of Schnappinger et al. [23] that had examined the expression profiles of intracellular versus broth-grown Mtb strain 1254. Despite differences between the two studies in the duration of the infection, the bacterial strains used and the growth characteristics of the intracellular bacteria, there was a level of correlation between datasets suggesting that a common set of genes determine a successful infection of a host cell (Figure 6, Table S4). We also compared our data with results obtained using transposon site hybridization, or TraSH, to identify genes important for survival inside murine bone-marrow-derived macrophages [64]. Only one gene, ahpD, was common to all 3 studies. ahpD encodes an alkylhydroperoxidase which acts in concert with another alkylhydroperioxidase, AhpC to provide antioxidant protection for Mtb [65]. Although ahpC was in both the Schnappinger et al. and our datasets, it was not an absolute requirement for optimal growth inside a macrophage as defined by the TraSH screen [64]. AhpC/AhpD may be important in isoniazid-resistant strains where the catalase-peroxidase KatG has undergone mutation to provide a resistant phenotype [65]. KatG has been ascribed an important role in virulence as it has been found to be required for the catabolism of exogenous peroxides generated by the oxidative burst or peroxinitrites generated by the reaction of superoxide and nitric oxide [66]. Interestingly, katG was not found to be induced intracellularly, in either our dataset or that of Schnappinger et al. nor was katG found to be an essential gene for survival in the macrophage [23], [64]. One possible explanation would be the constitutive expression and subsequent post-translational regulation of katG, which would not be detected in a comparative microarray analysis.
The overlap of genes identified to be of functional importance in studies of the intracellular environment is graphically represented as a Venn diagram using GeneVenn. “UBC-SGM” represents genes induced in M. tuberculosis H37Rv identified in this study, “Schnappinger 2003” represents genes induced in intracellular M. tuberculosis strain 1254 [23], and “TraSH MΦ” represents genes identified as essential for survival inside BM-MΦ [64].
With regards to genes commonly upregulated by both intracellular M. tuberculosis H37Rv and strain 1254 used in the Schnappinger study, genes involved in fatty acid metabolism (fadB2, fadD19, fadD26, fadD33, and fadE5), mycobactin synthesis (mbtB, mbtD-J), and [Fe-S] cluster biosynthesis (Rv1461, Rv1462, csd, Rv1465) were present in both lists. Additionally, both virulent strains induced a multidrug transporter (Rv1348/Rv1349) which a recent study suggested is involved in both siderophore synthesis and mycobacterial virulence [67]. Finally, pckA which encodes phosphoenolpyruvate carboxykinase (PEPCK) was also seen to be commonly upregulated in these two virulent strains when inside a macrophage. PEPCK catalyses the interconversion of oxaloacetate (OAA) to phosphoenolpyruvate (PEP) and has been thought to be involved in mycobacterial virulence either through its contributions to gluconeogenesis for carbohydrate formation or to the maintenance of the TCA cycle through the conversion of PEP to OAA [68]. Knocking-out pckA in M. bovis BCG resulted in an attenuated phenotype of the mutant versus wild-type BCG in both aerosol-infected C57BL/6 mice as well as C57BL/6 macrophages [69].
Comparing our list of genes up-regulated in intracellular H37Rv with the list of 126 genes required for optimal growth of H37Rv inside macrophages as defined by TraSH, 18 genes were common between the two datasets. Of this, half were conserved hypothetical genes that have not yet been fully characterized, and others, such as secA2 [70] and phoT [64] have been well studied and are required for growth in both mice and macrophages. fabG1, involved in the elongation of fatty acids, has recently been described as an essential gene even under axenic culture conditions [71], [72]. Also present in both the transcriptional and TraSH screens was clpC, which encodes for a chaperone protein presumably involved in a varied number of roles including secretion, gene regulation, protein folding and degradation [73], [74], [75]. A recent study characterizing protein-protein interactions in mycobacteria found that ClpC specifically associates with Cfp-10, a component of the ESX-1 secretion system [76]. Such associations will likely further the dissection of secretion mechanisms via the ESX-1 secretion system, and in combination with the requirement of clpC for optimal growth, highlight the importance of secretory mechanisms in mycobacterial virulence.
The main focus of this study however, was to compare intracellular expression profiles of H37Ra and H37Rv to identify factors that may explain the differences between the two strains in their pathogenesis and their ability to grow in macrophages. Recent comparisons between H37Ra and H37Rv have identified a base-pair polymorphism in phoP [37], [38]. Examining Mtb mutants of phoP, it has been observed that the PhoPR regulon affects a multitude of cellular adaptation pathways including hypoxic responses (via DevRS), respiration (nuo and ald operons that maintain NAD+/NADH flux), ESAT-6 secretion (impacting T-cell responses to Mtb), antibiotic sensitivity, cellular morphology, and lipid metabolism [39], [52], [77]. These studies not only defined genes that may be under control of the PhoPR regulon, but also demonstrated that one single nucleotide variation (SNV) had profound effects on Mtb virulence.
Our study in intracellular bacteria provides further confirmation of the importance of this SNV and its effects on expression of the PhoPR regulon inside the host cell. This single base-pair difference in H37Ra influences the regulation of phoP itself with an increase in phoP gene expression in H37Ra compared to H37Rv both in broth and inside macrophages. Interestingly, in both axenic culture and intracellularly, genes within the PhoP regulon were expressed at a lower level in H37Ra compared to H37Rv (e.g. pks3/4, fadD21, papA3, mmpL8, mmpL10). We speculate that the SNV in the phoP DNA binding domain of the attenuated strain results in ineffectual expression of the PhoP regulon in H37Ra, and that this in turn is reflected in the compensatory transcriptional induction of the PhoP regulator in H37Ra. Previous studies have described similarities between the lipid profiles (especially the components of PAT, DAT and sulpholipids) of the phoP mutants and H37Ra [39], [40], [41] – although it is unclear if the attenuation of the phoP mutant is comparable to that of H37Ra. The increased transcript levels of genes involved in PAT, DAT, and sulpholipid synthesis in H37Rv, as seen in our study comparing intracellular H37Ra and H37Rv, are predicted to be a requirement for remodeling of the bacterial cell envelope upon internalization [40]. Additionally, PhoPR involvement in metabolic processes such as fatty acid metabolism has led to the supposition that faulty regulation mediated by this two-component system could also affect bacterial adaptation to host environments in its ability to utilize alternate means of lipid degradation and synthesis. In Salmonella enterica Serovar Typhimurium, over-expression of phoPQ resulted in an attenuation of bacterial virulence as pathogenesis depended on fine control of transcriptional regulation [60]. Thus, we propose a similar situation where less efficient or unspecific binding of PhoP to regulatory motifs results in the loss of regulatory control of genes involved in cell-wall lipid synthesis resulting in the obvious phenotypic differences between H37Ra and H37Rv, and, additionally, as part of a compensatory mechanism, the over-expression of PhoP by H37Ra.
The difference in phoP regulon expression, however, may not be the sole effect of the phoP SNV or the sole reason for H37Ra attenuation. Modeling protein-protein interactions via STRING DB, we noted several two-component systems that have been observed to interact with PhoPR, including PrrB, the sensor kinase of the PrrA/B two-component system that may be required for environmental adaptations and early intracellular survival. These interactions suggest that several two-component systems, in concert with PhoPR, coordinate mycobacterial pathogenesis.
An additional contributor to the differing virulence phenotypes may be the ESX-1 secretion system. ESX-1 is encoded for by Region of Difference (RD) 1 which consists of the genes Rv3871 to Rv3878 and secretion of ESX-1 appears to be aided by genes upstream of RD1 (Rv3868-Rv3870) [78], [79]. RD1 was originally identified via subtractive hybridization to be present in wild-type M. bovis strains, but not in the vaccine strain BCG [79]. RD-1 is also present in Mtb H37, and interestingly, its expression is lower in broth-grown H37Ra compared to broth-grown H37Rv [80]. ESX-1 secretes ESAT-6 and its chaperone, CFP-10. ESAT-6 has been characterized as a secreted protein, and was originally isolated in short term broth culture filtrates of Mtb [81]. ESAT-6 is a virulence factor of Mtb and is an important target for T-cell responses in mice, humans, cattle and guinea pigs [82], [83], [84], [85]. Furthermore, ESAT-6 has been shown to have a role in cytolysis enabling virulent mycobacteria to spread intercellularly [86]. Mutants lacking the entire ESX-1 secretion system elicit significantly decreased tissue necrosis in the murine lung [87]. In addition to ESAT-6 and CFP-10, gene products encoded by Rv3614c-Rv3616c interact with and are also secreted via ESX-1 [8], [48]. Given that our studies show the expression of these associated products and an ESAT-6 family member were higher in intracellular H37Rv versus H37Ra, one could hypothesize that the virulent H37Rv exhibits greater secretory activity for an important set of virulence factors. If the secreted proteins do indeed act as effectors modulating host responses to Mtb, it could be reasoned that H37Ra, with its lower expression of these putative virulence factors, is unable to sufficiently alter the host activation state and finds itself in a more hostile intracellular compartment compared to H37Rv. Recently, a study examining the ESAT-6 secretion in H37Ra has postulated a model where ESX-1 function is linked to the PhoPR regulon [52]. It was shown that ESAT-6 secretion in H37Ra was deficient compared to H37Rv, and complementation of H37Ra with H37Rv phoP partially restored virulence of the attenuated strain in the mouse and macrophage models. The results of our study reveal a pattern of H37Rv and H37Ra gene expression in broth and inside a murine macrophage infection model that is consistent with this hypothesis. Additionally, modeling of protein-networks of other SNVs (Rv0637, PPE18, and PPE38) suggests genes unrelated to the PhoPR regulon that may also contribute to the regulation of ESX-1 secretion (Figure S1). While PPE18 is described as under regulation of PhoPR, Rv0637 and PPE38 have not been shown to be members of the regulon. However, products of both genes appear to interact with proteins encoded by genes found in the RD1 region (Rv3875-Rv3878) and even an ESX-1 secreted protein, EspG3. This suggests an additional regulatory step of ESX-1 secretion that is independent of PhoP.
Previously, our laboratory reported the use of a bacterial artificial chromosome (BAC) based approach, BAC fingerprint array (BACFA), to analyze expression differences between the two highly related Mtb strains, H37Ra and H37Rv [27]. The data presented here are complementary to those obtained using BACFA analysis in that both data sets suggest a scenario where H37Rv is better equipped to adapt to intracellular environments. Although these platforms were complementary there were expression differences exclusively identified using BACFA. This discrepancy may be due to the size and representation of the array elements used in the two techniques, however, this remains to be determined experimentally. Regardless, the microarray experiments did not confirm the differential expression of the frd operon, as detected by BACFA analysis and confirmed by qRT-PCR. The frd operon encodes for fumarate reductase, which in other bacteria, have been found to be involved in anaerobic respiration [88]. Further, our own studies treating Mtb-infected macrophages with a putative inhibitor of fumarate reductase resulted in significant reduction of intracellular growth of mycobacteria [7]. Differential expression of the frd operon highlights an important advantage of the BACFA technique: the ability to identify differences in operons. Had the genes of the frd operon been on different fragments in the fingerprint arrays, the individual signals would have been too weak to have warranted further analysis even though our additional studies indicate that fumarate reductase may have a role in mycobacterial pathogenesis. With microarrays, however, each gene has its own individual element, analyses designed to incorporate chromosomal location (such as that represented in Figure 3) are required to ensure that genes whose expression are biologically, if not numerically, significant are not overlooked.
By directly comparing intracellular transcriptomes of H37Ra and H37Rv, we have gained further insight into significant pathways required for pathogenesis. Previous microarray studies have examined either genomic or axenic broth differences between the two strains, but our study is the first to attempt a comprehensive examination of these strains in an intracellular environment. The attenuation of H37Ra involves a failure to adapt to both changes in the environment and metabolic constraints, evidenced by the lower expression of genes involved in mycobactin synthesis and fatty acid metabolism when the attenuated strain is moved from broth culture to the macrophage. We demonstrate for the first time differential regulation of the phoP regulon by both H37Ra and H37Rv inside a host cell. Additionally, we highlight not only the correlation between PhoP regulation and ESX-1 in intracellular mycobacteria, but also, we suggest additional investigation into PhoP-independent pathways of ESX-1 regulation.
Ultimately, the results presented here demonstrate that the virulent H37Rv responds to entering the intracellular environment by inducing a more extensive and vigorous transcriptomic response compared to H37Ra, that includes genes whose products allow it to adjust to metabolic challenges faced within the host. H37Ra on the other hand, initiates a moderate or, in the case of frd, delayed transcriptional program that may not be as effective for enabling adaptation to the novel host environment. This inadequate response to changes in the environment may be the result of dysfunctional control of key signaling cascades such as the PhoPR regulon. This attenuation of H37Ra may be further compounded by a failure to moderate the host intracellular milieu via ESX-1 mediated secretion of effectors.
Supporting Information
Figure S1.
Protein-protein network modeling of interactions that may be influenced by SNPs affecting H37Ra compared to H37Rv. We constructed network models for Rv0637, PPE18 (Rv1196) PPE38 (Rv2352c) and PhoP (Rv0757) using String DB (stringdb.org).
https://doi.org/10.1371/journal.pone.0011066.s001
(26.10 MB TIF)
Table S1.
H37Ra compared to H37Rv in axenic broth culture. Significantly differentially expressed genes were identified using a t-test (p<0.05) with Benjamani and Hochberg multiple testing correction and further filtered to include genes with >1.5 fold change in broth grown H37Ra compared to broth grown H37Rv.
https://doi.org/10.1371/journal.pone.0011066.s002
(0.09 MB XLS)
Table S2.
Common H37Ra and H37Rv response to the intracellular environment. Significantly differentially expressed genes were identified using a t-test (p<0.05) with Benjamani and Hochberg multiple testing correction and further filtered to include genes with >1.5 fold change in both intracellular H37Ra compared to broth grown H37Ra, and intracellular H37Rv compared to broth grown H37Rv.
https://doi.org/10.1371/journal.pone.0011066.s003
(0.05 MB XLS)
Table S3.
Gene clusters revealed by chromosome location mapping as differentially expressed intrcellularly in H37Ra compared to H37Rv. The difference between intracellular transcriptional profiles of H37Ra and H37Rv were plotted onto the chromosome using a moving average. Seven clusters of genes differentially expressed between the two strains inside murine macrophages were identified, and are listed below by chromosome position.
https://doi.org/10.1371/journal.pone.0011066.s004
(0.05 MB XLS)
Table S4.
Genes common to different intracellular expression studies and macrophage TraSH.
https://doi.org/10.1371/journal.pone.0011066.s005
(0.02 MB XLS)
Table S5.
Primers (5′-3′) used in qPCR validation of microarray results.
https://doi.org/10.1371/journal.pone.0011066.s006
(0.02 MB XLS)
Author Contributions
Conceived and designed the experiments: AHL SJW JH CAM MB PDB RWS. Performed the experiments: AHL SJW. Analyzed the data: AHL SJW JH CAM MB REWH WLL PDB RWS. Contributed reagents/materials/analysis tools: AHL SJW JH CAM MB REWH WLL PDB RWS. Wrote the paper: AHL SJW JH CAM MB REWH WLL PDB RWS.
References
- 1. Flynn JL, Chan J (2001) Immunology of tuberculosis. Annu Rev Immunol 19: 93–129.
- 2. Dunn PL, North RJ (1995) Virulence ranking of some Mycobacterium tuberculosis and Mycobacterium bovis strains according to their ability to multiply in the lungs, induce lung pathology, and cause mortality in mice. Infect Immun 63: 3428–3437.
- 3. Steenken W (1946) History of H37 strain of tubercle bacillus. American Review of Respiratory Disease 54: 62–66.
- 4. Collins FM, Smith MM (1969) A comparative study of the virulence of mycobacterium tuberculosis measured in mice and guinea pigs. Am Rev Respir Dis 100: 631–639.
- 5. Larson CL, Wicht WC (1964) Infection of Mice with Mycobacterium tuberculosis, Strain H37ra. Am Rev Respir Dis 90: 742–748.
- 6. Steenken W Jr (1961) The persistence of tubercle bacilli in caseous lesions in the experimental animal (guinea pig). The American review of respiratory disease 83: 550–554.
- 7. Li AH, Lam WL, Stokes RW (2008) Characterization of genes differentially expressed within macrophages by virulent and attenuated Mycobacterium tuberculosis identifies candidate genes involved in intracellular growth. Microbiology 154: 2291–2303.
- 8. MacGurn JA, Raghavan S, Stanley SA, Cox JS (2005) A non-RD1 gene cluster is required for Snm secretion in Mycobacterium tuberculosis. Mol Microbiol 57: 1653–1663.
- 9. Rivera-Marrero CA, Burroughs MA, Masse RA, Vannberg FO, Leimbach DL, et al. (1998) Identification of genes differentially expressed in Mycobacterium tuberculosis by differential display PCR. Microb Pathog 25: 307–316.
- 10. Sreevatsan S, Pan X, Stockbauer KE, Connell ND, Kreiswirth BN, et al. (1997) Restricted structural gene polymorphism in the Mycobacterium tuberculosis complex indicates evolutionarily recent global dissemination. Proc Natl Acad Sci U S A 94: 9869–9874.
- 11. Imaeda T (1985) Deoxyribonucleic acid relatedness among strains of Mycobacterium tuberculosis, Mycobacterium bovis BCG, Mycobacterium microti and Mycobacterium africanum. International Journal of Systemic Bacteriology 35: 147–150.
- 12. Lari N, Rindi L, Garzelli C (2001) Identification of one insertion site of IS6110 in Mycobacterium tuberculosis H37Ra and analysis of the RvD2 deletion in M. tuberculosis clinical isolates. J Med Microbiol 50: 805–811.
- 13. Brosch R, Philipp WJ, Stavropoulos E, Colston MJ, Cole ST, et al. (1999) Genomic analysis reveals variation between Mycobacterium tuberculosis H37Rv and the attenuated M. tuberculosis H37Ra strain. Infect Immun 67: 5768–5774.
- 14. Gao Q, Kripke K, Arinc Z, Voskuil M, Small P (2004) Comparative expression studies of a complex phenotype: cord formation in Mycobacterium tuberculosis. Tuberculosis (Edinb) 84: 188–196.
- 15. Kinger AK, Tyagi JS (1993) Identification and cloning of genes differentially expressed in the virulent strain of Mycobacterium tuberculosis. Gene 131: 113–117.
- 16. Wei J, Dahl JL, Moulder JW, Roberts EA, O'Gaora P, et al. (2000) Identification of a Mycobacterium tuberculosis gene that enhances mycobacterial survival in macrophages. J Bacteriol 182: 377–384.
- 17. Dasgupta N, Kapur V, Singh KK, Das TK, Sachdeva S, et al. (2000) Characterization of a two-component system, devR-devS, of Mycobacterium tuberculosis. Tuber Lung Dis 80: 141–159.
- 18. Park HD, Guinn KM, Harrell MI, Liao R, Voskuil MI, et al. (2003) Rv3133c/dosR is a transcription factor that mediates the hypoxic response of Mycobacterium tuberculosis. Mol Microbiol 48: 833–843.
- 19. Sherman DR, Voskuil M, Schnappinger D, Liao R, Harrell MI, et al. (2001) Regulation of the Mycobacterium tuberculosis hypoxic response gene encoding alpha -crystallin. Proc Natl Acad Sci U S A 98: 7534–7539.
- 20. Hampshire T, Soneji S, Bacon J, James BW, Hinds J, et al. (2004) Stationary phase gene expression of Mycobacterium tuberculosis following a progressive nutrient depletion: a model for persistent organisms? Tuberculosis (Edinb) 84: 228–238.
- 21. Talaat AM, Ward SK, Wu CW, Rondon E, Tavano C, et al. (2007) Mycobacterial bacilli are metabolically active during chronic tuberculosis in murine lungs: insights from genome-wide transcriptional profiling. J Bacteriol 189: 4265–4274.
- 22. Stokes RW, Waddell SJ (2009) Adjusting to a new home: Mycobacterium tuberculosis gene expression in response to an intracellular lifestyle. Future Microbiol 4: 1317–1335.
- 23. Schnappinger D, Ehrt S, Voskuil MI, Liu Y, Mangan JA, et al. (2003) Transcriptional adaptation of Mycobacterium tuberculosis within macrophages: Insights into the phagosomal environment. J Exp Med 198: 693–704.
- 24. Tailleux L, Waddell SJ, Pelizzola M, Mortellaro A, Withers M, et al. (2008) Probing host pathogen cross-talk by transcriptional profiling of both Mycobacterium tuberculosis and infected human dendritic cells and macrophages. PLoS One 3: e1403.
- 25. Rohde KH, Abramovitch RB, Russell DG (2007) Mycobacterium tuberculosis invasion of macrophages: linking bacterial gene expression to environmental cues. Cell Host Microbe 2: 352–364.
- 26. Stokes RW, Thorson LM, Speert DP (1998) Nonopsonic and opsonic association of Mycobacterium tuberculosis with resident alveolar macrophages is inefficient. J Immunol 160: 5514–5521.
- 27. Li AH, Lam WL, Stokes RW (2008) Bacterial artificial chromosome fingerprint arrays for the differentiation of transcriptomic differences in mycobacteria. J Microbiol Methods 75: 416–424.
- 28. Rooyakkers AW, Stokes RW (2005) Absence of complement receptor 3 results in reduced binding and ingestion of Mycobacterium tuberculosis but has no significant effect on the induction of reactive oxygen and nitrogen intermediates or on the survival of the bacteria in resident and interferon-gamma activated macrophages. Microb Pathog 39: 57–67.
- 29. Stokes RW, Haidl ID, Jefferies WA, Speert DP (1993) Mycobacteria-macrophage interactions. Macrophage phenotype determines the nonopsonic binding of Mycobacterium tuberculosis to murine macrophages. J Immunol 151: 7067–7076.
- 30. Stokes RW, Doxsee D (1999) The receptor-mediated uptake, survival, replication, and drug sensitivity of Mycobacterium tuberculosis within the macrophage-like cell line THP-1: a comparison with human monocyte-derived macrophages. Cell Immunol 197: 1–9.
- 31. Belisle JT, Sonnenberg MG (1998) Isolation of genomic DNA from mycobacteria. Methods Mol Biol 101: 31–44.
- 32.
Waddell SJ, Butcher , D P (2008) Whole genome analysis using microarrays. In: Parish T, editor. Mycobacteria Protocols Totowa. NJ: Humana Press. pp. 83–93.
- 33.
Mangan JA, Monahan IM, Butcher PD (2002) Gene expression during host-pathogen interactions: approaches to bacterial mRNA extraction and labeling for microarray analysis;. Wren BW, Dorrell N, editors. pp. 137–151.
- 34. Schmittgen TD, Zakrajsek BA, Mills AG, Gorn V, Singer MJ, et al. (2000) Quantitative reverse transcription-polymerase chain reaction to study mRNA decay: comparison of endpoint and real-time methods. Anal Biochem 285: 194–204.
- 35. Menendez Mdel C, Rebollo MJ, Nunez Mdel C, Cox RA, Garcia MJ (2005) Analysis of the precursor rRNA fractions of rapidly growing mycobacteria: quantification by methods that include the use of a promoter (rrnA P1) as a novel standard. J Bacteriol 187: 534–543.
- 36. Vandesompele J, De Paepe A, Speleman F (2002) Elimination of primer-dimer artifacts and genomic coamplification using a two-step SYBR green I real-time RT-PCR. Anal Biochem 303: 95–98.
- 37. Zheng H, Lu L, Wang B, Pu S, Zhang X, et al. (2008) Genetic basis of virulence attenuation revealed by comparative genomic analysis of Mycobacterium tuberculosis strain H37Ra versus H37Rv. PLoS ONE 3: e2375.
- 38. Lee JS, Krause R, Schreiber J, Mollenkopf HJ, Kowall J, et al. (2008) Mutation in the transcriptional regulator PhoP contributes to avirulence of Mycobacterium tuberculosis H37Ra strain. Cell Host Microbe 3: 97–103.
- 39. Walters SB, Dubnau E, Kolesnikova I, Laval F, Daffe M, et al. (2006) The Mycobacterium tuberculosis PhoPR two-component system regulates genes essential for virulence and complex lipid biosynthesis. Mol Microbiol 60: 312–330.
- 40. Gonzalo Asensio J, Maia C, Ferrer NL, Barilone N, Laval F, et al. (2006) The virulence-associated two-component PhoP-PhoR system controls the biosynthesis of polyketide-derived lipids in Mycobacterium tuberculosis. J Biol Chem 281: 1313–1316.
- 41. Perez E, Samper S, Bordas Y, Guilhot C, Gicquel B, et al. (2001) An essential role for phoP in Mycobacterium tuberculosis virulence. Mol Microbiol 41: 179–187.
- 42. Trivedi OA, Arora P, Sridharan V, Tickoo R, Mohanty D, et al. (2004) Enzymic activation and transfer of fatty acids as acyl-adenylates in mycobacteria. Nature 428: 441–445.
- 43. Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C, et al. (1998) Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393: 537–544.
- 44. Rodriguez GM, Smith I (2003) Mechanisms of iron regulation in mycobacteria: role in physiology and virulence. Mol Microbiol 47: 1485–1494.
- 45. Soto CY, Andreu N, Gibert I, Luquin M (2002) Simple and rapid differentiation of Mycobacterium tuberculosis H37Ra from M. tuberculosis clinical isolates through two cytochemical tests using neutral red and nile blue stains. J Clin Microbiol 40: 3021–3024.
- 46. Goren MB, Brokl O, Schaefer WB (1974) Lipids of putative relevance to virulence in Mycobacterium tuberculosis: correlation of virulence with elaboration of sulfatides and strongly acidic lipids. Infect Immun 9: 142–149.
- 47. DiGiuseppe Champion PA, Cox JS (2007) Protein secretion systems in Mycobacteria. Cell Microbiol 9: 1376–1384.
- 48. Fortune SM, Jaeger A, Sarracino DA, Chase MR, Sassetti CM, et al. (2005) Mutually dependent secretion of proteins required for mycobacterial virulence. Proc Natl Acad Sci U S A 102: 10676–10681.
- 49. Raghavan S, Manzanillo P, Chan K, Dovey C, Cox JS (2008) Secreted transcription factor controls Mycobacterium tuberculosis virulence. Nature 454: 717–721.
- 50. Tekaia F, Gordon SV, Garnier T, Brosch R, Barrell BG, et al. (1999) Analysis of the proteome of Mycobacterium tuberculosis in silico. Tuber Lung Dis 79: 329–342.
- 51. Chesne-Seck ML, Barilone N, Boudou F, Gonzalo Asensio J, Kolattukudy PE, et al. (2008) A point mutation in the two-component regulator PhoP-PhoR accounts for the absence of polyketide-derived acyltrehaloses but not that of phthiocerol dimycocerosates in Mycobacterium tuberculosis H37Ra. J Bacteriol 190: 1329–1334.
- 52. Frigui W, Bottai D, Majlessi L, Monot M, Josselin E, et al. (2008) Control of M. tuberculosis ESAT-6 secretion and specific T cell recognition by PhoP. PLoS Pathog 4: e33.
- 53. Sassetti CM, Rubin EJ (2003) Genetic requirements for mycobacterial survival during infection. Proc Natl Acad Sci U S A 100: 12989–12994.
- 54. Jensen LJ, Kuhn M, Stark M, Chaffron S, Creevey C, et al. (2009) STRING 8–a global view on proteins and their functional interactions in 630 organisms. Nucleic Acids Res 37: D412–416.
- 55. Mustafa AS, Skeiky YA, Al-Attiyah R, Alderson MR, Hewinson RG, et al. (2006) Immunogenicity of Mycobacterium tuberculosis antigens in Mycobacterium bovis BCG-vaccinated and M. bovis-infected cattle. Infect Immun 74: 4566–4572.
- 56. Xu J, Laine O, Masciocchi M, Manoranjan J, Smith J, et al. (2007) A unique Mycobacterium ESX-1 protein co-secretes with CFP-10/ESAT-6 and is necessary for inhibiting phagosome maturation. Mol Microbiol 66: 787–800.
- 57. Dessen A, Quemard A, Blanchard JS, Jacobs WR Jr, Sacchettini JC (1995) Crystal structure and function of the isoniazid target of Mycobacterium tuberculosis. Science 267: 1638–1641.
- 58. Bitter W, Houben EN, Bottai D, Brodin P, Brown EJ, et al. (2009) Systematic genetic nomenclature for type VII secretion systems. PLoS Pathog 5: e1000507.
- 59. Teng F, Wang L, Singh KV, Murray BE, Weinstock GM (2002) Involvement of PhoP-PhoS homologs in Enterococcus faecalis virulence. Infect Immun 70: 1991–1996.
- 60. Miller SI, Mekalanos JJ (1990) Constitutive expression of the phoP regulon attenuates Salmonella virulence and survival within macrophages. J Bacteriol 172: 2485–2490.
- 61. Wang S, Engohang-Ndong J, Smith I (2007) Structure of the DNA-binding domain of the response regulator PhoP from Mycobacterium tuberculosis. Biochemistry 46: 14751–14761.
- 62. Gonzalo-Asensio J, Soto CY, Arbues A, Sancho J, del Carmen Menendez M, et al. (2008) The Mycobacterium tuberculosis phoPR operon is positively autoregulated in the virulent strain H37Rv. J Bacteriol 190: 7068–7078.
- 63. Dullaghan EM, Malloff CA, Li AH, Lam WL, Stokes RW (2002) Two-dimensional bacterial genome display: a method for the genomic analysis of mycobacteria. Microbiology 148: 3111–3117.
- 64. Rengarajan J, Bloom BR, Rubin EJ (2005) Genome-wide requirements for Mycobacterium tuberculosis adaptation and survival in macrophages. Proc Natl Acad Sci U S A 102: 8327–8332.
- 65. Sherman DR, Mdluli K, Hickey MJ, Barry CE 3rd, Stover CK (1999) AhpC, oxidative stress and drug resistance in Mycobacterium tuberculosis. Biofactors 10: 211–217.
- 66. Ng VH, Cox JS, Sousa AO, MacMicking JD, McKinney JD (2004) Role of KatG catalase-peroxidase in mycobacterial pathogenesis: countering the phagocyte oxidative burst. Mol Microbiol 52: 1291–1302.
- 67. Rodriguez GM, Smith I (2006) Identification of an ABC transporter required for iron acquisition and virulence in Mycobacterium tuberculosis. J Bacteriol 188: 424–430.
- 68.
Madigan MT, Martinko JM, Parker J (1997) Brock biology of microorganisms - Chapter 13: Metabolic diversity among the microorganisms. pp. 473–531. Upper Saddle River, New Jersey.
- 69. Liu K, Yu J, Russell DG (2003) pckA-deficient Mycobacterium bovis BCG shows attenuated virulence in mice and in macrophages. Microbiology 149: 1829–1835.
- 70. Kurtz S, McKinnon KP, Runge MS, Ting JP, Braunstein M (2006) The SecA2 secretion factor of Mycobacterium tuberculosis promotes growth in macrophages and inhibits the host immune response. Infect Immun 74: 6855–6864.
- 71. Marrakchi H, Ducasse S, Labesse G, Montrozier H, Margeat E, et al. (2002) MabA (FabG1), a Mycobacterium tuberculosis protein involved in the long-chain fatty acid elongation system FAS-II. Microbiology 148: 951–960.
- 72. Parish T, Roberts G, Laval F, Schaeffer M, Daffe M, et al. (2007) Functional complementation of the essential gene fabG1 of Mycobacterium tuberculosis by Mycobacterium smegmatis fabG but not Escherichia coli fabG. J Bacteriol 189: 3721–3728.
- 73. Bolhuis A, Matzen A, Hyyrylainen HL, Kontinen VP, Meima R, et al. (1999) Signal peptide peptidase- and ClpP-like proteins of Bacillus subtilis required for efficient translocation and processing of secretory proteins. J Biol Chem 274: 24585–24592.
- 74. Nielsen E, Akita M, Davila-Aponte J, Keegstra K (1997) Stable association of chloroplastic precursors with protein translocation complexes that contain proteins from both envelope membranes and a stromal Hsp100 molecular chaperone. Embo J 16: 935–946.
- 75. Porankiewicz J, Wang J, Clarke AK (1999) New insights into the ATP-dependent Clp protease: Escherichia coli and beyond. Mol Microbiol 32: 449–458.
- 76. Singh CR, Moulton RA, Armitige LY, Bidani A, Snuggs M, et al. (2006) Processing and presentation of a mycobacterial antigen 85B epitope by murine macrophages is dependent on the phagosomal acquisition of vacuolar proton ATPase and in situ activation of cathepsin D. J Immunol 177: 3250–3259.
- 77. Gonzalo-Asensio J, Mostowy S, Harders-Westerveen J, Huygen K, Hernandez-Pando R, et al. (2008) PhoP: a missing piece in the intricate puzzle of Mycobacterium tuberculosis virulence. PLoS One 3: e3496.
- 78. Behr MA, Wilson MA, Gill WP, Salamon H, Schoolnik GK, et al. (1999) Comparative genomics of BCG vaccines by whole-genome DNA microarray. Science 284: 1520–1523.
- 79. Gordon SV, Brosch R, Billault A, Garnier T, Eiglmeier K, et al. (1999) Identification of variable regions in the genomes of tubercle bacilli using bacterial artificial chromosome arrays. Mol Microbiol 32: 643–655.
- 80. Mostowy S, Cleto C, Sherman DR, Behr MA (2004) The Mycobacterium tuberculosis complex transcriptome of attenuation. Tuberculosis (Edinb) 84: 197–204.
- 81. Andersen P, Askgaard D, Ljungqvist L, Bennedsen J, Heron I (1991) Proteins released from Mycobacterium tuberculosis during growth. Infect Immun 59: 1905–1910.
- 82. Elhay MJ, Oettinger T, Andersen P (1998) Delayed-type hypersensitivity responses to ESAT-6 and MPT64 from Mycobacterium tuberculosis in the guinea pig. Infect Immun 66: 3454–3456.
- 83. Pollock JM, Andersen P (1997) Predominant recognition of the ESAT-6 protein in the first phase of interferon with Mycobacterium bovis in cattle. Infect Immun 65: 2587–2592.
- 84. Ravn P, Demissie A, Eguale T, Wondwosson H, Lein D, et al. (1999) Human T cell responses to the ESAT-6 antigen from Mycobacterium tuberculosis. J Infect Dis 179: 637–645.
- 85. Brandt L, Oettinger T, Holm A, Andersen AB, Andersen P (1996) Key epitopes on the ESAT-6 antigen recognized in mice during the recall of protective immunity to Mycobacterium tuberculosis. J Immunol 157: 3527–3533.
- 86. Hsu T, Hingley-Wilson SM, Chen B, Chen M, Dai AZ, et al. (2003) The primary mechanism of attenuation of bacillus Calmette-Guerin is a loss of secreted lytic function required for invasion of lung interstitial tissue. Proc Natl Acad Sci U S A 100: 12420–12425.
- 87. Junqueira-Kipnis AP, Basaraba RJ, Gruppo V, Palanisamy G, Turner OC, et al. (2006) Mycobacteria lacking the RD1 region do not induce necrosis in the lungs of mice lacking interferon-gamma. Immunology 119: 224–231.
- 88.
Madigan MT, Martinko JM, Parker J (1997) Brock biology of microorganisms - Chapter 4: Nutrition and metabolism. Upper Saddle River, New Jersey: Prentice-Hall. pp. 136–137.