Bacterial strains

The two mycobacterial reference strains, M. tuberculosis H37Ra (MNC 16394) and M. tuberculosis H37Rv (ATCC 27294), used in this study were kindly provided by Dr Harleen Grewal, The Gade Institute, University of Bergen, Norway. The strains had undergone less than 3 passages in the laboratory before being used for this study. The bacilli were cultured on Middelbrook 7H10 agar plates with OADC enrichment (BD Difco) at 37°C and 5% CO₂ for 3-4 weeks. Bacterial colonies were harvested by using an extraction buffer consisting of phosphate-buffered saline (PBS), pH 7.4 with freshly added Roche Protease Inhibitor Cocktail (1 μg/ml) (Complete, EDTA-free, Roche Gmbh, Germany). Six hundred μl of this extraction buffer was added to each agar plate and the mycobacterial colonies were gently scraped off the agar surface using a cell scraper. Aliquots of the resulting pasty bacterial mass were transferred into 2 ml cryotubes with O-rings (Sarstedt, Norway) containing 250 μl of acid washed glass beads (≤106 μm; Sigma-Aldrich, Norway) and an additional 600 μl of extraction buffer containing a cocktail of protease inhibitors (1 μg/ml) (Roche Diagnostics GmbH), and stored at -80°C until further treatment. For protein extraction, the mycobacteria were disrupted mechanically by bead-beating in a Ribolyser (Hybaid, UK) at max speed (6.5) for 45 seconds.

Triton X-114 extraction of exported proteins from whole bacteria

Triton X-114 phase-partitioning was used to isolate lipophilic proteins following the method of Bordier [20] and a modified version for extraction of lipophilic proteins from whole bacilli [21]. Briefly, 3-4 week old bacilli were lysed by bead-beating and unbroken cells and cell-wall debris were removed by centrifugation at 2300 g for 5 minutes. Triton X-114 was added to the supernatant (final detergent concentration 2%, w/v) and the suspension was stirred at 4°C for 30 minutes. Residual insoluble materials were removed by centrifugation at 15700 g for 10 min at 4°C. For separation of the hydrophobic and hydrophilic proteins, the solution was incubated at 37°C for 15 minutes, the solution separated into two phases, an upper aqueous phase containing hydrophilic proteins, and a lower (detergent) phase containing the hydrophobic proteins. Proteins in the lower detergent phase were precipitated by acetone.

Gel electrophoresis and in-gel digestion of proteins

Extracted proteins, 50 μg from each strain, were mixed with 25 μl sodium-dodecyl-sulphate (SDS) loading buffer and boiled for 5 minutes before separation on a 10 cm long 1 mm thick 12% SDS polyacrylamide gel. The protein migration was allowed to proceed until the bromophenol dye had migrated to the bottom of the gel. The protein bands were visualized with Coomassie Brilliant Blue R-250 staining (Invitrogen, Carlsbad, CA, U.S.A.). Protein lanes were divided in 10 fractions by cutting between the visible protein bands ranging from ~3 kDa to ~188 kDa and washed twice for 15 minutes at room temperature (RT) with 50% acetonitrile (ACN) in 25 mM ammonium bicarbonate (NH₄HCO₃). The gel pieces were dehydrated by incubating them with 50 μl 100% ACN for 20 minutes at RT. The disulfide bonds in the proteins were reduced using 10 mM dithiotreitol and alkylated with 55 mM iodoacetamide; both in 100 mM NH₄HCO₃. The gel pieces were dehydrated by 100% ACN as described above, and rehydrated in 25 mM NH₄HCO₃. The proteins were digested by trypsin (Promega, Madison, U.S.A.) for 16-20 h at 37°C. The peptides were eluted stepwise from each gel piece using 1% formic acid (FA), then 0.1% FA in 50% ACN and the last one 100% ACN. Each incubation was performed for 20 minutes at RT in 100 μl volumes, and finally the 3 supernatants were pooled.

Mass spectrometry

Experiments were performed on a Dionex Ultimate 3000 nano-LC system (Sunnyvale CA, USA) connected to a linear quadrupole ion trap-Orbitrap (LTQ-Orbitrap) mass spectrometer (ThermoElectron, Bremen, Germany) equipped with a nanoelectrospray ion source. The mass spectrometer was operated in the data-dependent mode to automatically switch between Orbitrap-MS and LTQ-MS/MS acquisition. Survey full scan MS spectra (from m/z 400 to 2,000) were acquired in the Orbitrap with resolution R = 60,000 at m/z 400 (after accumulation to a target of 1,000,000 charges in the LTQ). The method used allowed sequential isolation of the most intense ions (up to five, depending on signal intensity) for fragmentation on the linear ion trap using collisionally induced dissociation at a target value of 100,000 charges.

For accurate mass measurements the lock mass option was enabled in MS mode and the polydimethyilcyclosiloxane (PCM) ions generated in the electrospray process from ambient air (protonated (Si(CH₃)₂O)6; m/z 445.120025) were used for internal recalibration during the analysis [22]. Target ions already selected for MS/MS were dynamically excluded for 30 seconds. General mass spectrometry conditions were: electrospray voltage, 1.9 kV. Ion selection threshold was 500 counts for MS/MS, an activation Q-value of 0.25 and activation time of 30 milliseconds was also applied for MS/MS.

All acquired data were processed and analyzed using MaxQuant (version 1.0.13.13), a software script specifically developed for data acquired using high-resolution instrumentation [23]. MS/MS peak lists from 60 individual RAW files were generated using the Quant.exe tool from the MaxQuant package. Protein identification was performed by searching combined data from each fraction against an in-house developed M. tuberculosis complex database (4,643 protein sequences) [24]. The database was also modified to contain reversed sequences of all entries as a control of false-positive identifications during analysis [25]. In addition, common contaminants such as keratins, bovine serum albumin and trypsin were also added to the database (database final size of 9,308 protein sequences). We used MASCOT Deamon for submission of multiple searches to a local Mascot server v2.2 (Matrix Science). The search parameters were: Enzyme: trypsin with no proline restriction; Maximum missed cleavages: 3; Carbamidomethyl (C) as fixed modification; N-acetyl (Protein), oxidation (M), Pyr-Q (Gln to 2-pyrrolidone-5-carboxylic acid-Glu) and Pyr-E (Glu to 2-pyrrolidone-5-carboxylic acid-Glu) as variable modifications; Peptide mass tolerance of ± 15 parts per million; MS/MS mass tolerance of 0.5 Da.

Protein identification and validation was performed with Identify.exe from MaxQuant using the following parameters: peptide and protein false discovery rate: 0.01 (1%), minimal peptide length was 7, and to guarantee a high confidence identification rate, the maximal posterior error probability was set to 0.1 (from a range of 0 to 1); minimal number of unique peptides per protein: 1. The average mass accuracy for the identified peptides was 400 parts per billion. The MS/MS fragments assignments for all identified peptide sequences (including for single peptide-based protein identifications) are freely available at the Tranche network http://proteomecommons.org (see Supporting Information Available section for more details).

Estimation of protein abundance

To determine differentially represented membrane proteins between the M. tuberculosis H37Rv and the M. tuberculosis H37Ra strains, we used MaxQuant peak intensity calculations as a parameter for protein abundance. Previous reports demonstrate a good correlation between peak intensity and protein levels in the sample [26, 27]. To avoid variation due to loading differences between samples on the instrument, individual intensity values of each protein were divided by the sum of all intensities in the sample as a normalization procedure.

Proteins were divided in two categories as follows: I) for proteins identified in both samples, the difference in relative abundance between the strains had to be higher than 5 fold; II) for a protein identified in only one of the strains, we required that it had to be identified with a minimal of four different peptides. Such stringent criteria are required to guarantee that a protein identified in only one sample is most probably due to differences in abundance between the samples, and not because parent ions were not identified (but still present) in the MS analysis due to random fluctuation of the MS/MS data-dependant acquisition procedure.

Primary sequence analysis

The primary sequence analysis of the observed proteins to identify exported proteins were performed using the publically available algorithms: TMHMM version 2 for identification of transmembrane helixes (TMH) in membrane proteins http://www.cbs.dtu.dk/services/TMHMM/, SignalP for prediction of secreted proteins http://www.cbs.dtu.dk/services/SignalP/, and PROSITE for prediction of lipoproteins http://au.expasy.org/prosite/.

Triton X-114 detergent extracted proteins

The aim of this study was to perform a proteomic analysis on protein expression of two closely related lineages of M. tuberculosis, the virulent H37Rv and the avirulent H37Ra strains, with a main focus on membrane- and membrane-associated proteins. For this purpose, cultured bacilli were mechanically disrupted and proteins extracted by Triton X-114 detergent phase separation. Proteins were then precipitated by acetone, separated by SDS-PAGE, and analysed by high resolution mass spectrometry. Additional Figure 1 gives an example of the quality of the mass spectrometry data gathered in this work, which illustrates the full sequence obtained for ion m/z 1476.82, which was identified by Mascot as peptide LVLGSADGAVYTLAK from Rv2138, probable conserved lipoprotein LppL, with a Mascot score of 118 and contains fragmentation data for all the expected y-series daughter ions. In total, 1771 different protein groups were identified, with 1578 proteins identified in the M. tuberculosis H37Rv strain, and 1493 were observed in the H37Ra strain. The additional files 1 & 2 include peak lists, information about the criteria of protein identifications, such as number of peptides matching each protein, score and identification threshold.

Among the 1771 proteins observed in this study, there were 1300 proteins that were common to both strains. However, 278 proteins were exclusively identified in the M. tuberculosis H37Rv, while another 193 proteins were solely observed in the H37Ra strain. Further, to ascertain the validity of the comparison analysis of the two strains due to technical error margins, we have only taken into account the proteins observed with 4 or more different peptides. Using these stringent criteria, we reduced the number of the observed strain specific proteins drastically to only 4 identified in M. tuberculosis H37Rv but not observed in H37Ra. Two of them were predicted with 3 (Rv3479) and 13 transmembrane regions (Rv3792), one hypothetical protein (Rv2319c) and one secreted protein (R1184c). No such examples were found in M. tuberculosis H37Ra.

The data obtained in this study, was searched for membrane and membrane-associated proteins by using the TMHMM v2.0 algorithm http://www.cbs.dtu.dk/services/TMHMM/. In M. tuberculosis H37Rv 371 proteins were identified that were predicted to have 1 or more TMH regions, while in M. tuberculosis H37Ra 357 proteins were identified predicted to be anchored to the membrane by 1 or more TMHs. As it appears from Figure 1, the distributions of proteins identified with different number TMHs were similar for the two strains, with proteins with only 1 TMH as the largest group. Three hundred and twenty one of all the membrane proteins were common for both strains, while 36 membrane proteins were only observed in M. tuberculosis H3Ra and 51 membrane proteins only observed in M. tuberculosis H37Rv (Additional file 3).

Another interesting group of proteins that are associated with the membrane is lipoproteins. These are proteins translocated to the cell membrane and retained there by post-translational lipid modification. They are functionally diverse, and are suggested to be involved in host-pathogen interactions [28, 29]. They are also interesting with respect to development of serodiagnostic tests for detection of TB due to their strong immunogenicity [30, 31]. Lipoproteins represent a subgroup of secreted proteins characterized by the presence of a lipobox. The lipobox motif is located in the distal C-terminal part of the N-terminal signal peptide [32]. This motif functions as a recognition signal for lipid modification, which is made on the conserved and essential cysteine residue. Precursor lipoproteins are mainly translocated in a Sec-dependent manner across the plasma membrane and are subsequently modified [33]. The proteins identified in this study were analysed by PROSITE for prediction of lipoproteins http://au.expasy.org/prosite/. Seventy-six of them were predicted as potential lipoproteins, based on the presence of a cleavable signal peptide and signal peptidase II recognition motif. Sixty six of all the lipoproteins were common for both strains, while 7 lipoproteins were only observed in M. tuberculosis H3Ra and 3 lipoproteins only observed in M. tuberculosis H37Rv (Additional file 4).

Estimation of relative abundance

Using MaxQuant software that provide quantitative information about proteins and peptides using the spectra generated during the LC runs the relative abundance of each protein observed in both M. tuberculosis H37Rv and M. tuberculosis H37Ra were examined after normalization. Our data showed that most of the proteins identified in both strains had similar relative abundance. Using Pearson's method for correlation, the relative abundance of proteins observed in the two strains were significantly correlated with a correlation coefficient of 0.887 (p < 0.001), and R² = 0.78 (Figure 2). However, there were some proteins that had different relative abundance between the two strains. To ensure the relative protein abundance for these proteins were real and not due to technical error margins, we only focused on the ones with a 5 fold difference or higher. To this end, there were 121 proteins from both strains that belonged to different functional groups (Additional file 5). In order to reduce the amount of data required to be analysed, and due to the anticipated important biological role of membrane- and membrane-associated proteins, we chose to focus only on membrane- and lipoproteins. This further reduced the number of proteins to only 19 and 10 proteins in M. tuberculosis H37Rv and M. tuberculosis H37Ra, respectively (Table 1). Among the proteins observed with a 5 fold or higher relative abundance in M. tuberculosis H37Rv strain, 17 were predicted to be directly anchored to the membrane by TMH retention regions in their primary sequences, while 2 others are thought to be retained on the membrane through N-terminal lipidation.