Skip to main content

Bioinformatic identification of Mycobacterium tuberculosis proteins likely to target host cell mitochondria: virulence factors?



M. tuberculosis infection either induces or inhibits host cell death, depending on the bacterial strain and the cell microenvironment. There is evidence suggesting a role for mitochondria in these processes.

On the other hand, it has been shown that several bacterial proteins are able to target mitochondria, playing a critical role in bacterial pathogenesis and modulation of cell death. However, mycobacteria–derived proteins able to target host cell mitochondria are less studied.


A bioinformaic analysis based on available genomic sequences of the common laboratory virulent reference strain Mycobacterium tuberculosis H37Rv, the avirulent strain H37Ra, the clinical isolate CDC1551, and M. bovis BCG Pasteur strain 1173P2, as well as of suitable bioinformatic tools (MitoProt II, PSORT II, and SignalP) for the in silico search for proteins likely to be secreted by mycobacteria that could target host cell mitochondria, showed that at least 19 M. tuberculosis proteins could possibly target host cell mitochondria. We experimentally tested this bioinformatic prediction on four M. tuberculosis recombinant proteins chosen from this list of 19 proteins (p27, PE_PGRS1, PE_PGRS33, and MT_1866). Confocal microscopy analyses showed that p27, and PE_PGRS33 proteins colocalize with mitochondria.


Based on the bioinformatic analysis of whole M. tuberculosis genome sequences, we propose that at least 19 out of 4,246 M. tuberculosis predicted proteins would be able to target host cell mitochondria and, in turn, control mitochondrial physiology. Interestingly, such a list of 19 proteins includes five members of a mycobacteria specific family of proteins (PE/PE_PGRS) thought to be virulence factors, and p27, a well known virulence factor. P27, and PE_PGRS33 proteins experimentally showed to target mitochondria in J774 cells. Our results suggest a link between mitochondrial targeting of M. tuberculosis proteins and virulence.


In spite of the huge efforts to overcome the burden of tuberculosis (TB), nearly 10 million incident cases of TB cases, the death of 1.1 million HIV–negative TB patients and an additional 0.35 million deaths from HIV–associated TB are reported each year. Unfortunately, the selection and spread of multidrug–resistant (MDR) Mycobacterium tuberculosis strains worsen the scenario, since an estimated 0.65 million cases of MDR–TB were documented for the year 2010 [1]. Clearly, in addition to the improvement of human population welfare, the development of new vaccines, early diagnosis tests, and pharmacological treatments, a precise knowledge on mycobacteria–host cell interactions is also a requirement for the successful control of TB.

In this regard, some bacterial pathogenicity factors have been shown to contain N–terminal mitochondrial targeting signals [2, 3] and a diverse array of bacterial proteins including some bacterial toxins from enteropathogenic E.coli. Salmonella spp, N. gonorrhoea, N. meningitides, A. baumanii, H. pylori, S. aureus, S. pneumoniae, C. sordelli, and C. difficile have been shown to target mitochondria [4].

Two proteins of particular interest are mitochondrial–associated protein (Map, former Orf19) and EspF from enteropathogenic E.coli, which enter the host cell via the type 3 secretion system and colocalize with host mitochondria [2]. It is worth noting that, Map causes alterations, in the form of mitochondrial membrane potential (Δψm) dissipation but apparently it is not responsible for enteropathogenic E.coli–induced host cell apoptosis [2, 4, 5].

Mycobacterial infection affects mitochondrial function [68] and it has been suggested that Δψm changes are related to the mycobacterial strain′s virulence [7, 8]. However, the identity of the molecules responsible for such a response has not been defined. Recently, the M. tuberculosis protein PE_PGRS33 was shown to localize within host cell mitochondria, and in doing so, induces host cell apoptosis [9].

Since mitochondria targeting proteins such as Map and EspF (from enteropathogenic E. coli) are secreted proteins, we thought that it is likely that at least some of the mycobacterial proteins that potentially target host cell mitochondria are also secreted proteins. Therefore, we set up a bioinformatic search of M. tuberculosis proteins likely to be secreted and to target host cell mitochondria.

The M. tuberculosis genome was first made available in 1998, opening the possibility for in–depth analysis of the possible pathogenic mechanisms involved in the course of mycobacterial infections [10]. Other mycobacterial genomes have been elucidated since, and new bio–informatic resources allow for in silico analysis of their genomes [11, 12].

Genome wide analysis predicts that M. tuberculosis H37Rv and CDC1551 contain 3,924 and 4,246 genes, respectively [10, 11].

By using freely available data bases and bioinformatic tools, we were able to single out 19 M. tuberculosis CDC1551 predicted proteins as likely candidates for targeting mitochondria.

Among these, five PE/PE_PGRS family proteins deserve particular attention. About 10% of the potential coding capacity of M. tuberculosis accounts for two large unrelated gene families encoding the PE and PPE proteins. The names PE and PPE are derived from the motifs Pro–Glu (PE) and Pro–Pro–Glu (PPE), and the largest class of the PE family in M. tuberculosis H37Rv is the PE_PGRS subfamily which consist of proteins with a PE domain followed by a C–terminal glycine–rich extension encoded by the PGRS motif (polymorphic GC–rich repetitive sequence) [13, 14]. As a family, the PE/PE_PGRS proteins are polymorphic [10] and account for many of the differences found between the avirulent (H37Ra) and virulent (H37Rv) strains of M. tuberculosis and, in consequence, have been proposed as possible virulence factors [15].


M. tuberculosisproteins predicted to be secreted and to target host cell mitochondria

The whole 4,246 predicted proteins from the M. tuberculosis CDC1551 genome as shown in the JCVI/CMR webpage were analyzed for the presence of mitochondrial targeting sequences by using the MitoProt II–v1.101 algorithm. Although previous analyses using this algorithm on other bacterial proteomes have set the probability limit at 0.700 to consider a protein as likely to target mitochondria [16], we decided for a more stringent analysis and therefore we set the probability limit at 0.8500 that would help to reduce false negatives. As shown in Figure 1, a total of 337 out of 4,246 M. tuberculosis CDC1531 proteins are predicted to have mitochondrial targeting potential. Next we analyzed those 337 proteins by using another mitochondrial targeting predictive algorithm (PSORT II), and found that 136 out of the 337 proteins scored as likely to target host cell mitochondria. Finally, the 136 predicted proteins that scored positive for both predictive algorithms were further analyzed for the probability of harboring secretory signal peptides by using the SignalP 3.0 software. Taken together, these three bioinformatic resources allowed us to identify 19 predicted proteins with the probability of harboring secretory peptide signals as well as mitochondrial targeting signals. Table 1, lists the identity of the 47 M. tuberculosis predicted proteins (Top–47) with the highest combined MitoProt II and PSORT II probability of targeting mitochondria. Table 2, list the 19 predicted proteins from M. tuberculosis CDC1551strain, likely to be secreted by the mycobacteria and to target host cell mitochondria.

Figure 1
figure 1

Distribution of predicted M. tuberculosis CDC1551 proteins according to their signal peptides. The 4, 246 predicted proteins of M. tuberculosis CDC1551 were analyzed for the presence of mitochondrial targeting sequences by the predictive algorithm MitoProt II, the 337 proteins that scored positive were analyzed by PSORT II, rendering 136 proteins likely to target mitochondria. Finally, these 136 proteins were analyzed for the presence of secretory signal peptides by the SignalP software.

Table 1 Top–47 scored M. tuberculosis CDC1551 proteins likely to target mitochondria
Table 2 M. tuberculosis CDC1551 proteins predicted to be secreted and targeted to host cell mitochondria

PE and PE_PGRS family proteins are among the M. tuberculosisCDC1551 proteins likely to be secreted and targeted to host cell mitochondria

Table 2 shows that the bioinformatic procedure used in this work selected 19 out of 4,246 predicted proteins from the M. tuberculosis CDC1551 strain genome that are predicted to be secreted and targeted to the host cell mitochondria. By looking at the annotated characteristics of those proteins, it is evident that members from the PE and PE_PGRS family proteins are overrepresented (5/19, 26.3%), since the total number of PE and PE_PGRS proteins is estimated to be 94, by manually counting them from the JCVI/CMR M. tuberculosis CDC1551 list (94/4,246, 2.21%). Interestingly, a recent comparative analysis among mycobacterial genomes, performed in an attempt to identify the mutations that lead to the loss of virulence in the M. tuberculosis H37Ra strain, showed that most mutations were confined to the PE/PPE/PE_PGRS family genes, highlighting the likely role of those proteins as virulent factors [15]. Accordingly, we performed a search for homologous proteins in M. tuberculosis H37Ra, M. tuberculosis H37Rv, and M. bovis BCG Pasteur 1173P2 strain and then analyzed those homologous proteins by using MitoProt II, PSORT II, and SignalP algorithms as previously described. Table 3 shows the protein codes for each mycobacterial strain, as well as the possible identity and the scores for the three aforementioned predictive algorithms. These analyses showed that the percentage of similarity among the identified M. tuberculosis CDC1551 strain PE or PE_PGRS proteins and their H37Ra, H37Rv, and BCG Pasteur strain homologous proteins ranged from 53.4% to 81.9%. The H37Ra (MRA_0115) and BCG (BCG_0142) homologous to CDC1551 MT_0118 protein are predicted as not mitochondrial targeted, whereas the H37Rv (Rv0109) scored positive just as well as MT_0118, and the same holds true for the three homologous proteins to MT_0349, but none of them seem to have secretory signal sequences. All six homologous proteins to CDC1551 MT_0854 and MT_1866 did not score positive for mitochondrial targeting sequences. On the other hand, the three homologous proteins for CDC1551 MT_2690 are predicted to harbor mitochondrial targeting sequences. With the exception of MRA_0344, Rv0335c, and BCG_0374, all other homologous proteins are predicted to harbor secretory signal sequences.

Table 3 Comparative analysis of some M. tuberculosis CDC1551 strain PE and PE_PGRS proteins with their H37Ra, H37Rv and BCG Pasteur strain orthologes

Rv1411c (p27), and Rv1818c (PE_PGRS33) M. tuberculosisproteins target host cell mitochondria

Based on the list of 19 M. tuberculosis CDC1551 proteins likely to target host cell mitochondria (Table 2) as well as on the H37Ra, H37Rv, and BCG Pasteur 1173P2 homologous proteins to M. tuberculosis CDC1551 (Table 3), we selected four proteins to be experimentally tested for targeting mitochondria: MT_1866, and its H37Rv ortholog Rv1818c (PE_PGRS33); the H37Rv ortholog of MT_0118 (Rv0109 or PE_PGRS1); and the H37Rv ortholog of MT_1455 (Rv1411c or p27). In addition, the M. tuberculosis Rv2031c (α–crystallin) protein that scored negative for mitochondria targeting sequence in a Mitoprot II and PSORT II query, was used as a negative control. Rv1818c has recently shown to target mitochondria [9] and therefore served the purpose of a positive control. All M. tuberculosis recombinant proteins harbour a histidine tail that allows detection by means of a FITC–labelled anti–His antibody, as described in methods.

Figure 2 shows that p27, and PE_PGRS33 colocalize with mitochondria in J774 cells cultured for 2 h in the presence of 1 μg/ml of the correspondent recombinant protein. MT_1866 and PE_PGRS1 do not seem to colocalize with mitochondria, whereas α–crystallin protein, as expected, do not colacalize with mitochondria.

Figure 2
figure 2

M. tuberculosis p27 and PE_PGRS33 target J774 cell mitochondria. J774 macrophages exposed for 2 h to 1 μg/ml of p27, MT_1866, PE_PGRS1, PE_PGRS33, or α–crystallin His–tagged recombinant proteins were labelled with Mito Tracker red CMX ROS (mitochondria, red), FITC–labelled anti His antibody (recombinant protein, green), and DAPI (nuclei, blue), and then analyzed by confocal microscopy, in order to assess targeting of recombinant proteins to host cell mitochondria. Results are representative of multiple microscope fields form three independent experiments.


Macrophages are the main host cells for M. tuberculosis, therefore considerable attention has been devoted to the analysis of the mycobacteria–containing phagosomes [17] and the way by which mycobacteria inhibit their association with lysosomes [1820], and more recently the host–pathogen cross–talk by transcriptional analyses has gained interest [21, 22].

On the other hand, it has been shown that M. tuberculosis infection controls host cells, either by inducing [23, 24] or inhibiting their death [25, 26], and some M. tuberculosis–derived molecules have been associated with those opposing biological activities [2729].

Mitochondria play a pivotal role in programmed cell death [30], and mitochondrial activity has been shown to be disrupted by mycobacterial infection in a process that appears to be dependent on the strain′s virulence [68].

It has been shown that several secreted microbial proteins are transferred to the cell during infection and may target mitochondria, playing an important role in bacterial pathogenicity [4, 31]; and more recent observations suggest that the mitochondrial role in the regulation of cellular responses to pathogens may go further than the sole regulation of host cell death [32].

Taking all this into account we wanted to analyze the M. tuberculosis genome and their predicted proteome in search of moieties likely to play a mitochondrial–mediated pathogenicity role, by using bioinformatic tools.

Gomez et al. [33] have already identified M. tuberculosis secreted proteins by performing a bioinformatic search on the 3,924 initially predicted proteins from the M. tuberculosis H37Rv genome [10].

Here, we used a different approach. The whole 4,246 predicted proteins from the M. tuberculosis CDC1551 genome [11] were first analyzed for the presence of mitochondrial targeting sequences by using MitoProt II–v1.101 and PSORT II algorithms. The CDC1551 strain was chosen on the basis of it being isolated from a relatively recent tuberculosis outbreak.

The MitoProt II analysis we performed was more stringent than similar searches on other bacterial proteomes were authors concluded that at least 5% of proteins from E.coli are predisposed for targeting mitochondria [16]. In this work, by using two predictive algorithms, it was found that, in M. tuberculosis CDC1551, 136 out of 4,246 (3.2%) proteins are likely to target mitochondria. Mitoprot II–V1.101 has a level of accuracy of 94.76%–97.54% for successfully predicting mitochondrial proteins and 76.79%–87.92% for successfully predicting non–mitochondrial proteins in a protein training group [34], and PSORT II has an accuracy of approximately 60%–86% [35].

The 136 predicted proteins likely to target mitochondria were further analyzed for the presence of secretory signal peptides by using the SignalP 3.0 hidden Markov model, which, incidentally, has recently been reported to outperform other similar tools in predicting M. tuberculosis secreted proteins with a level of accuracy of 100%, as it predicted no false positives or false negatives in a negative set (n=61), and in a positive set (n=57) of mycobacterial proteins, respectively. In addition, this algorithm showed 78.9% accuracy in predicting the actual cleavage site [36]. Their analysis predicted that between 7.8% and 10.5% of the proteins in the proteomes of different mycobacteria (including the CDC1551 strain) are secreted proteins. Our SignalP v3.00 hidden Markov model search predicted that 19 out of the 136 proteins likely to target mitochondria are also secreted proteins, i.e., 13.9%.

In their analysis, Gomez et al. [33] excluded 47 PE/PPE proteins out of the top M. tuberculosis 208 proteins predicted to be secretory, arguing that biased amino acid composition and a repetitive primary sequence could lead to unreliable results in sequence analyses. We did not exclude these proteins from our analysis and we found that some PE_PGRS proteins are not predicted to harbour mitochondrial targeting or secretory signals (data not shown); thus, suggesting that the amino acid composition of these proteins did not introduce any bias in our bioinformatic output. In addition, it has recently been shown that some PE_PGRS and PPE proteins are secreted by at least one mycobacterial species (M. marinum) and that those proteins are transported via the type VII secretion system ESX–5 [37].

Interestingly, among the 19 CDC1551 proteins likely to be secreted and to targeted mitochondria, 5 belong to the PE/PE_PGRS protein family (Table 2).

About 10% of the coding capacity of the genome is devoted to two large unrelated families of acidic, glycine–rich proteins, the PE (Pro–Glu) and PPE (Pro–Pro–Glu) families, often based on multiple copies of the polymorphic repetitive sequence referred to as PGRSs (polymorphic GC–rich repetitive sequences). It has been suggested that these protein families could be of immunological importance [10], and a possible source of antigenic variability that would help M. tuberculosis to evade the host immune system during infection [14]. In addition, by using a computational strategy based on phylogenetic profiling and comparative proteomic analysis, Meszaros et al. [38] have proposed that M. tuberculosis PE/PPE proteins can be considered as potential targets for drug design. Although the biological function of the PE proteins is still under study, the PE_PGRS member annotated as Rv1759c has been characterized as fibronectin–binding protein [39] and a genetic screen of M. bovis BCG Pasteur mutagenized with the transposon Tn5367, identified a gene, identical to the Rv1818c gene of M. tuberculosis, encoding a PE_PGRS protein that influences the interactions of mycobacteria with macrophages [40]. Rv1818c has also been implied in T lymphocytes apoptosis [41], and it has been shown that most genetic differences between M. tuberculosis H37Rv and H37Ra strains, likely to function as virulence factors are in the PE/PPE/PE_PGRS family genes [15].

The availability of suitable bioinformatic tools allowed us to single out 19 M. tuberculosis proteins as likely to be secreted and to target host cell mitochondria, among them five PE/PE_PGRS family protein members. Recently, Cadieux et al. [9] showed that the M. tuberculosis PE_PGRS33 protein (encoded by the Rv1818c gene) co–localize with mitochondria when expressed in human rhabdomyosarcoma (RD) cells, and mitochondrial localization of PE_PGRS33 protein was followed by induction of cell death [9]. This finding is somewhat in contrast with our in silico analysis, since PE_PGRS33 (RV1818c) is not predicted to harbour a mitochondrial targeting sequence by MitoProt or PSORT II (Table 3) Moreover, the M. tuberculosis CDC1551 protein MT_1866 which is the ortholog of Rv1818c is predicted to harbour a mitochondria targeting sequence (Table 3) and experimentally failed to do so (Figure 2). A possible explanation for this is that PE_PGRS proteins are unreliably alignable or differences in the annotation for the strain CDC1551 as compared for the H37Rv stain of M. tuberculosis. In addition the similarity score between RV1818c and MT_1866 is just 53.4%, a very low score for considering true homology between these two proteins (Table 3).

All in all, our results show, in another experimental system (J774 murine macropheges exposed to the recombinant protein), that PE_PGRS33 targets host cell mitochondria, thus confirming Cadieux et al. [9] findings.

In addition, based on bioinformatic analyses, we present evidence that Rv1411c (p27) which is the M. tuberculosis H37Rv ortholog of M. tuberculosis CDC1551 MT_1455 and a known virulence factor, targets mitochondria (Figure 2).

Identifying other M. tuberculosis proteins that target mitochondria would help to clarify the structural requirements for mitochondrial targeting and more important, the role of those proteins in the M. tuberculosis–host cell communication. This work provided a list of 19 candidate proteins, and after experimentally testing some of them and identifying p27 as a mitochondria targeting protein, still leaves some candidate proteins for further analysis.


Over the past decade it has been shown that in addition to their role in ATP production, mitochondria also function as signal transducers platforms and have an essential role in immune responses and host–pathogen interactions.

In analysing the pathogen–host cell crosstalk, it has been shown that several pathogen proteins target mitochondria thus controlling the host cell fate. Here, we conducted a bioinformatic analysis aimed at identifying M. tuberculosis proteins that, based on aminoacid sequence, could possible be secretory, and target mitochondria.

Our results singled out 19 out of 4, 246 M. tuberculosis predicted proteins as likely to be secretory and harbour mitochondrial targeting sequences. Interestingly, several proteins considered to be virulence factors are included and over–represented in this 19 protein list. Four such proteins were experimentally tested and while confirming that PE_PGRS33 targets mitochondria, a new M. tuberculosis mitochondria targeting protein was identified (p27).

This work suggests a correlation between M. tuberculosis protein mitochondrial targeting potential and virulence. The interaction between M. tuberculosis and its host cell is complex and, as we think, understanding the role of mitochondria in mycobacterial infections and the role of M. tuberculosis specific proteins in the crosstalk with host cells at the level of mitochondria will lead us to a better explanation of phenomena such as tuberculosis latency.


Bioinformatic analysis of predicted M. tuberculosisproteins

M. tuberculosis predicted protein sequences were downloaded from the M. tuberculosis CDC1551 Genome Project from J. Craig Venter Institute/Comprehensive Microbial Resource (JCVI/CMR version 1.0) (–scripts/CMR/GenomePage.cgi?database=gmt).

Protein sequences were then analyzed by using the MitoProt II–v1.101 Software (http://ihg2.helmholtz–, which calculates the N–terminal protein region that can support a mitochondrial targeting sequence and the cleavage site, as described [34]. M. tuberculosis putative proteins with a probability of at least 0.8500 of having a mitochondrial targeting sequence were selected and re–analyzed by means of the PSORT II prediction algorithm for subcellular localization (http://psort.ims.u– which uses the k nearest neighbours classifier [35]. Those proteins that in addition of scoring well for MitoProt II had the highest probability of being targeted to mitochondria by PSORT II were further analyzed for secretory signal peptides by using SignalP 3.0 (, which predicts the presence and location of signal peptide cleavage sites in amino acid sequences from Gram–positive prokaryotes, Gram–negative prokaryotes, and eukaryotes. The method incorporates a prediction of cleavage sites and a signal peptide/non–signal peptide prediction based on a combination of several artificial neural networks and hidden Markov models, as described [4244].

The amino acid sequences of some proteins likely to harbor mitochondrial targeting sequences, and to be secreted by the CDC1551 strain of M. tuberculosis, were compared against the homologous proteins of M. tuberculosis H37Rv, M. tuberculosis H37Ra, and M. bovis BCG Pasteur 1173P2 strain, by using the “Protein vs. All Alignment” tool at the JCVI/CMR webpage. The retrieved amino acid sequences were then analyzed by MitoProt II–v1.101, PSORT II, and SignalP 3.0 algorithms as previously described.

M. tuberculosisrecombinant proteins

The cloning, expression and purification of the M. tuberculosis Rv1411c (p27) protein was performed as previously described [45]. The full length of Rv1818c (PE_PGRS33) gene cloned into PET15b fused to a histidine tag was a kind gift from Dr. M.J. Brennan (CBER, FDA, Bethesda, MD, USA). The coding regions for the mycobacterial antigens were amplified by PCR with the high fidelity DNA polymerase Pfx (Invitrogen) from M. tuberculosis H37Rv genomic DNA (Rv1411c or p27; Rv0109 or PE_PGRS1; Rv1818c or PE_PGRS33) or M. tuberculosis CDC1551 (MT_1866) with specific oligonucleotide primers. PCR products were ligated into the pCR4 Blunt–TOPO vector (Invitrogen), and amplified in JM110 Escherichia coli strain (Stratagene) and then subcloned into pET15b vector, that was used for the transformation of the E. coli strains Rosetta (DE3) (Novagen) or C41 (DE3) (Avidis S.A.). Expression of the proteins was induced in logarithmic phase cultures. Bacterial pellets were suspended in PBS, sonicated and centrifuged. His–tagged recombinant proteins were purified in an AKTA FLPC system (GE Healthcare), using 1 ml Histrap columns (GE Healthcare). Pooled purified recombinant proteins were dialized against 0.3 M NaCl, 20 mM NaH2PO4, pH 8.0. In addition, the M. tuberculosis Rv2031 protein (also known as 16 kDa antigen, hsp 16.3, hspX, and α–crystallin), obtained from E. coli and purified by histidin–affinity chromatography was used as a negative control on the basis of its low probability score for targeting mitochondria (Mi MitoProt II–v1.101, and PSORT II). In all cases, protein concentration was quantified and recombinant proteins were stored at −70°C until use.

Mitochondrial targeting of M. tuberculosisrecombinant proteins

J774.A1 murine macrophages were plated on glass cover–slips in 6–well culture plates (Corning) at a density of 5×105 cells/well and incubated overnight at 37°C in 5% CO2 atmosphere. After washing, cells were incubated in the presence of the indicated M. tuberculosis recombinant proteins (p27, MT_1866. PE_PGRS1, PE_PGRS33, or α–crystallin used as a negative control) for 2 h. Cells were then labelled with MitoTracker red CMXRos (Invitrogen) at a final concentration of 100 nM, for 30 min and subsequently fixed and permeabilized (Cytofix/Cytoperm, BD Biosciences) and stained with FITC–labelled anti–His antibody (Miltenyi Biotec). After further washing, cells were mounted in DAPI–containing Vectashield (Vector). Imaging was performed in a LSM5 Pascal (Zeiss) confocal microscope, using a 100x oil immersion objective. Online data acquisition and further fluorescence analysis were carried out using LSM5 image browser (Zeiss).


M. tuberculosis :

Mycobacterium tuberculosis


Prolin–Glutamic acid motif


Prolin–Prolin–Glutamic acid motif


Polymorphic GC–rich repetitive sequence.


  1. WHO: report 2010,,

  2. Kenny B, Jepson M: Targeting of an enteropathogenic Escherichia coli (EPEC) effector protein to host mitochondria. Cell Microbiol. 2000, 2: 579-590. 10.1046/j.1462-5822.2000.00082.x.

    Article  CAS  PubMed  Google Scholar 

  3. Nougayrede J-P, Donnenberg MS: Enterophatogenic Escherichia coli EspF is targeted to mitochondria and is required to initiate the mitochondrial death pathway. Cell Microbiol. 2004, 6: 1097-1111. 10.1111/j.1462-5822.2004.00421.x.

    Article  CAS  PubMed  Google Scholar 

  4. Kozjak-Pavlovic V, Ross K, Rudel T: Import of bacterial pathogenicity factors into mitochondria. Curr Op Microbiol. 2008, 11: 9-14. 10.1016/j.mib.2007.12.004.

    Article  CAS  Google Scholar 

  5. Papatheodorou P, Domanska G, Oxle M, Mathieu J, Selchow O, Kenny B, Rassow J: The enterophatogenic Escherichia coli (EPEC) Map effector is imported into the mitochondrial matrix by the TOM/Hsp70 system and alters organelle morphology. Cell Microbiol. 2006, 8: 677-689. 10.1111/j.1462-5822.2005.00660.x.

    Article  CAS  PubMed  Google Scholar 

  6. Duan L, Gan H, Golan DE, Remold HG: Critical role of mitochondrial damage in determining outcome of macrophage infection with Mycobacterium tuberculosis. J Immunol. 2002, 169: 5181-5187.

    Article  PubMed  Google Scholar 

  7. Abarca-Rojano E, Rosas-Medina P, Zamudio-Cortéz P, Mondragón-Flores R, Sánchez-García FJ: Mycobacterium tuberculosis virulence correlates with mitochondrial cytochrome c release in infected macrophages. Scand J Immunol. 2003, 58: 419-427. 10.1046/j.1365-3083.2003.01318.x.

    Article  CAS  PubMed  Google Scholar 

  8. Chen M, Gan H, Remold HG: A mechanism of virulence: virulent Mycobacterium tuberculosis strain H37Rv, but not attenuated H37Ra, causes significant mitochondrial inner membrane disruption in macrophages leading to necrosis. J Immunol. 2006, 176: 3707-3716.

    Article  CAS  PubMed  Google Scholar 

  9. Cadieux N, Parra M, Cohen H, Maric D, Morris SL, Brennan MJ: Induction of cell death after localization of the host cell mitocondria by the Mycobacterium tuberculosis PE_PGRS33 protein. Microbiol. 2011, 157: 793-804. 10.1099/mic.0.041996-0.

    Article  CAS  Google Scholar 

  10. Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C, Harris D, Gordon SV, Eiglmeier K, Gas S, Barry CE, Tekaia F, Badcock K, Basham D, Brown D, Chillingworth T, Connor R, Davies R, Devlin K, Feltwell T, Gentles S, Hamlin N, Holroyd S, Hornsby T, Jagels K, Krogh A, McLean J, Moule S, Murphy L, Oliver K, Osborne J, Quail MA, Rajandream MA, Rogers J, Rutter S, Seeger K, Skelton J, Squares R, Squares S, Sulston JE, Taylor K, Whitehead S, Barrell BG: Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature. 1998, 393: 537-544. 10.1038/31159.

    Article  CAS  PubMed  Google Scholar 

  11. Fleischmann RD, Alland D, Eisen JA, et al: Whole–genome comparison of Mycobacterium tuberculosis clinical and laboratory strains. J Bacteriol. 2002, 184: 5479-5490. 10.1128/JB.184.19.5479-5490.2002.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  12. Kapopoulou A, Lew JM, Cole ST: The Mycobrowser portal: a comprehensive and manually annotated resource for Mycobacterial genomes. Tubercul. 2010, 91: 8-13.

    Article  Google Scholar 

  13. Cole ST: Learning from the genome sequence of Mycobacterium tuberculosis H37Rv. FEBS Lett. 1999, 452: 7-10. 10.1016/S0014-5793(99)00536-0.

    Article  CAS  PubMed  Google Scholar 

  14. Banu S, Honore N, Saint-Joanis B, Philpott D, Prevost M-C, Cole ST: Are the PE_PGRS proteins of Mycobacterium tuberculosis variable surface antigens?. Mol Microbiol. 2002, 44: 9-19. 10.1046/j.1365-2958.2002.02813.x.

    Article  CAS  PubMed  Google Scholar 

  15. Zheng H, Lu L, Wang B, Pu S, Zhang X, Zhu G, Shi W, Zhang L, Wang H, Wang S, Zhao G, Zhang Y: Genetic basis of virulence attenuation revealed by comparative genomic analysis of Mycobacterium tuberculosis strain H37Ra versus H37Rv. PLoS One. 2008, 3: e2375-10.1371/journal.pone.0002375.

    Article  PubMed Central  PubMed  Google Scholar 

  16. Lucattini R, Likic VA, Lithgow T: Bacterial proteins predisposed for targeting to mitochondria. Mol Biol Evol. 2004, 21: 652-658. 10.1093/molbev/msh058.

    Article  CAS  PubMed  Google Scholar 

  17. Rohde K, Yates RM, Purdy GE, Russell DG: Mycobacterium tuberculosis and the environment within the phagosome. Immunol Rev. 2007, 219: 37-54. 10.1111/j.1600-065X.2007.00547.x.

    Article  CAS  PubMed  Google Scholar 

  18. Armstrong JA, Hart PD: Response of cultured macrophages to Mycobacterium tuberculosis, with observations on fusion of lysosomes with phagosomes. J Exp Med. 1971, 134: 713-740. 10.1084/jem.134.3.713.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  19. Walburger A, Koul A, Ferrari G, Nguyen L, Prescianotto-Baschong C, Huygen K, Kebl B, Thompson C, Bacher G, Pieters J: Protein kinase G from pathogenic mycobacteria promotes survival within macrophages. Science. 2004, 304: 1800-1804. 10.1126/science.1099384.

    Article  CAS  PubMed  Google Scholar 

  20. Axelrod S, Oschkinat H, Enders J, Schlegel B, Brinkmann V, Kaufmann SH, Hass A, Schaible UE: Delay of phagosome maturation by a mycobacterial lipid is reversed by nitric oxide. Cell Microbiol. 2008, 10: 1530-1545. 10.1111/j.1462-5822.2008.01147.x.

    Article  CAS  PubMed  Google Scholar 

  21. Ragno S, Romano M, Howell S, Pappin DJ, Jenner PJ, Colston MJ: Changes in gene expression in macrophages infected with Mycobacterium tuberculosis: a combined transcriptomic and proteomic approach. Immunol. 2001, 104: 99-108. 10.1046/j.1365-2567.2001.01274.x.

    Article  CAS  Google Scholar 

  22. Tailleux L, Waddell SJ, Pelizzola M, Mortellaro A, Withers M, Tanne A, Castagnoli PR, Gicquel B, Stoker NG, Butcher PD, Foti M, Neyrolles O: Probing host pathogen cross–talk by transcriptional profiling of both Mycobacterium tuberculosis and infected human dendritic cells and macrophages. PLoS One. 2008, 3: e1403-10.1371/journal.pone.0001403.

    Article  PubMed Central  PubMed  Google Scholar 

  23. Danelishvili L, McGarvey J, Li YJ, Bermudez LE: Mycobacterium tuberculosis infection causes different levels of apoptosis and necrosis in human macrophages and alveolar epithelial cells. Cell Microbiol. 2003, 5: 649-660. 10.1046/j.1462-5822.2003.00312.x.

    Article  CAS  PubMed  Google Scholar 

  24. Kelly DM, ten Bokum AMC, O’Leary SM, O’Sullivan MP, Keane J: Bystander macrophage apoptosis after Mycobacterium tuberculosis H37Ra infection. Infect Immun. 2008, 76: 351-360. 10.1128/IAI.00614-07.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  25. Keane J, Remold HG, Kornfeld H: Virulent Mycobacterium tuberculosis strain evades apoptosis of infected alveolar macrophages. J Immunol. 2000, 164: 2016-2020.

    Article  CAS  PubMed  Google Scholar 

  26. Sly LM, Hingley-Wilson SM, Reiner NE, McMaster WR: Survival of Mycobacterium tuberculosis in host macrophages involves resistance to apoptosis dependent upon induction of antiapoptotic Bcl–2 family member Mcl–1. J Immunol. 2003, 170: 430-437.

    Article  CAS  PubMed  Google Scholar 

  27. Derrick SC, Morris SL: The ESAT6 protein of Mycobacterium tuberculosis induces apoptosis of macrophages by activating caspase expression. Cell Microbiol. 2007, 9: 1547-1555. 10.1111/j.1462-5822.2007.00892.x.

    Article  CAS  PubMed  Google Scholar 

  28. Velmurugan K, Chen B, Miller JL, Azogue S, Gurses S, Hsu T, Glickman M, Jacobs WR, Porcelli SA, Briken V: Mycobacterium tuberculosis nuoG is a virulent gene that inhibits apoptosis of infected host cells. PLoS Pathog. 2007, 3: e110-10.1371/journal.ppat.0030110.

    Article  PubMed Central  PubMed  Google Scholar 

  29. Sanchez A, Espinosa P, Esparza MA, Colon M, Bernal G, Mancilla R: Mycobacterium tuberculosis 38–KDa lipoprotein is apoptogenic for human monocyte–derived macrophages. Scand J Immunol. 2009, 69: 20-28. 10.1111/j.1365-3083.2008.02193.x.

    Article  CAS  PubMed  Google Scholar 

  30. Green DR, Kroemer G: The pathophysiology of mitochondrial cell death. Science. 2004, 305: 626-629. 10.1126/science.1099320.

    Article  CAS  PubMed  Google Scholar 

  31. Boya P, Roques B, Kroemer G: New EMBO members′review: viral and bacterial proteins regulating apoptosis at the mitochondrial level. EMBO J. 2001, 20: 4325-4331. 10.1093/emboj/20.16.4325.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  32. Arnoult D, Carneiro L, Tattoli I, Girardin SE: The role of mitochondria in cellular defense against microbial infection. Semin Immunol. 2009, 21: 223-232. 10.1016/j.smim.2009.05.009.

    Article  CAS  PubMed  Google Scholar 

  33. Gomez M, Johnson S, Gennaro ML: Identification of secreted proteins of Mycobacterium tuberculosis by a bioinformatic approach. Infect Immun. 2000, 68: 2323-2327. 10.1128/IAI.68.4.2323-2327.2000.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  34. Claros MG, Vincens P: Computational method to predict mitochondrially imported proteins and their targeting sequences. Eur J Biochem. 1996, 241: 779-786. 10.1111/j.1432-1033.1996.00779.x.

    Article  CAS  PubMed  Google Scholar 

  35. Horton P, Nakai K: Better prediction of protein cellular localization sites with the k nearest neighbours classifier. Intell Syst Mol Biol. 1997, 5: 147-152.

    CAS  Google Scholar 

  36. Leversen NA, de Souza GA, Malen H, Prasad S, Jonassen I, Wiker H: Evaluation of signal peptide prediction algorithms for identification of mycobacterial signal peptides using sequence data from proteomic methods. Microbiol. 2009, 155: 2375-2383. 10.1099/mic.0.025270-0.

    Article  CAS  Google Scholar 

  37. Abdallah AM, Verboom T, Weerdenburg EM, van Pittius NC G, Mahasha PW, Jiménez C, Parra M, Cadieux N, Brennan MJ, Appelmelk BJ, Bitter W: PPE and PE–PGRS proteins of Mycobacterium marinum are transported via the type VII secretion system ESX–5. Mol Mcrobiol. 2009, 73: 329-340. 10.1111/j.1365-2958.2009.06783.x.

    Article  CAS  Google Scholar 

  38. Mészáros B, Tóth J, Vértessy BG, Dosztányi Z, Simon I: Proteins with complex architecture as potential targets for drug design: a case study of Mycobacterium tuberculosis. PLoS Comp Biol. 2011, 7: e1002118-10.1371/journal.pcbi.1002118.

    Article  Google Scholar 

  39. Espitia C, Laclette JP, Mondragon-Palomino M, Amador A, Campusano J, Martens A, Singh M, Cicero R, Zhang Y, Moreno C: The PE_PGRS glycine–rich proteins of Mycobacterium tuberculosis: a new family of fibronectin–binding proteins?. Microbiol. 1999, 145: 3487-3495.

    Article  CAS  Google Scholar 

  40. Brennan MJ, Delogu G, Chen Y, Bardarov S, Kriakov J, Alavi M, Jacobs WR: Evidence that Mycobacterial PE_PGRS proteins are cell surface constituents that influence interactions with other cells. Infect Immun. 2001, 69: 7326-7333. 10.1128/IAI.69.12.7326-7333.2001.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  41. Blaji KN, Goyal G, Narayana Y, Srinivas M, Chaturvedi R, Mohammad S: Apoptosis triggered by Rv1818c, a PE family gene from mycobacterium tuberculosis is regulated by mitochondrial intermediates in T cells. Microb Infect. 2007, 9: 271-281. 10.1016/j.micinf.2006.11.013.

    Article  Google Scholar 

  42. Nielsen H, Engelbrecht J, Brunak S, von Heijne G: Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Prot Engin. 1997, 10: 1-6. 10.1093/protein/10.1.1.

    Article  CAS  Google Scholar 

  43. Bendtsen JD, Nielsen H, Von Heijne G, Brunak S: Improved prediction of signal peptides: SignalP 3.0. J Mol Biol. 204, 340: 783-795.

    Article  Google Scholar 

  44. Emanuelsson O, Brunak S, von Heijne G, Nielsen H: Locating proteins in the cell using TargetP, SignalP and related tools. Nat Protocol. 2007, 2: 953-971. 10.1038/nprot.2007.131.

    Article  CAS  Google Scholar 

  45. Gonzalez-Zamorano M, Mendoza-Hernandez G, Xolapa W, Parada C, Vallecillo AJ, Bigi F, Espitia C: Mycobacterium tuberculosis glycoproteomics based on ConA–lectin affinity capture of mannosylated proteins. J Prot Res. 2009, 8: 721-733. 10.1021/pr800756a.

    Article  CAS  Google Scholar 

Download references


This work was financed in part by a Consejo Nacional de Ciencia y Tecnología grant (CB–2010/158340) and SIP–IPN (20080287), the funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. We thank Cristina Parada, for assistance on recombinant protein purification, Jacqueline Oliva–Ramírez for help with Confocal microscopy, and Professor Sir Salvador Moncada and Dr. César Hernández–Rodríguez for critical reading of the manuscript. MMBMA and FJSG are COFAA–IPN, EDI, and SNI fellows, CE and RHP are SNI fellows.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Francisco Javier Sánchez-García.

Additional information

Competing interest

The authors declare no competing interest.

Authors’ contributions

MMBMA conceived the idea, experimental design,analyzed data; ISPG performed experiments; CE provided special reagents, experimental design, analyzed data; MSM performed experiments; RHP provided special reagents, analyzed data; FJSG conceived the idea, experimental design, performed experiments, analyzed data, and wrote the paper. All authors read and approved the final manuscript.

Authors’ original submitted files for images

Rights and permissions

Open Access This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License ( ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Moreno-Altamirano, M.M.B., Paredes-González, I.S., Espitia, C. et al. Bioinformatic identification of Mycobacterium tuberculosis proteins likely to target host cell mitochondria: virulence factors?. Microb Informatics Exp 2, 9 (2012).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: