Skip to main content
  • Loading metrics

The Mycobacterium tuberculosis PE15/PPE20 complex transports calcium across the outer membrane

  • Vishant Boradia,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Center for Global Infectious Disease Research, Seattle Children’s Research Institute, Seattle, Washington, United States of America

  • Andrew Frando,

    Roles Investigation

    Affiliation Center for Global Infectious Disease Research, Seattle Children’s Research Institute, Seattle, Washington, United States of America

  • Christoph Grundner

    Roles Conceptualization, Formal analysis, Funding acquisition, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing

    Affiliations Center for Global Infectious Disease Research, Seattle Children’s Research Institute, Seattle, Washington, United States of America, Department of Pediatrics, University of Washington, Seattle, Washington, United States of America, Department of Global Health, University of Washington, Seattle, Washington, United States of America


The mechanisms by which nutrients traverse the Mycobacterium tuberculosis (Mtb) outer membrane remain mostly unknown and, in the absence of classical porins, likely involve specialized transport systems. Calcium ions (Ca2+) are an important nutrient and serve as a second messenger in eukaryotes, but whether bacteria have similar Ca2+ signaling systems is not well understood. To understand the basis for Ca2+ transport and signaling in Mtb, we determined Mtb’s transcriptional response to Ca2+. Overall, only few genes changed expression, suggesting a limited role of Ca2+ as a transcriptional regulator. However, 2 of the most strongly down-regulated genes were the pe15 and ppe20 genes that code for members of a large family of proteins that localize to the outer membrane and comprise many intrinsically disordered proteins. PE15 and PPE20 formed a complex and PPE20 directly bound Ca2+. Ca2+-associated phenotypes such as increased ATP consumption and biofilm formation were reversed in a pe15/ppe20 knockout (KO) strain, suggesting a direct role in Ca2+ homeostasis. To test whether the PE15/PPE20 complex has a role in Ca2+ transport across the outer membrane, we created a fluorescence resonance energy transfer (FRET)-based Ca2+ reporter strain. A pe15/ppe20 KO in the FRET background showed a specific and selective loss of Ca2+ influx that was dependent on the presence of an intact outer cell wall. These data show that PE15/PPE20 form a Ca2+-binding protein complex that selectively imports Ca2+, show a distinct transport function for an intrinsically disordered protein, and support the emerging idea of a general family-wide role of PE/PPE proteins as idiosyncratic transporters across the outer membrane.


Second messengers are a class of signaling molecules that permit a fast response and swift amplification of signals intracellularly. Among second messengers, calcium ions (Ca2+) are a particularly versatile signal in eukaryotes [1]. The functions of hundreds of human proteins are directly regulated by Ca2+, and almost every cellular process is affected by Ca2+ [1,2]. Ca2+ signaling is initially facilitated by Ca2+ flux along membrane Ca2+ gradients that are carefully maintained by pumps and transporters.

Several bacteria also maintain a similar Ca2+ gradient between the inside and outside of the cell [3]. What’s more, Ca2+ has anecdotally been linked to bacterial processes such as motility, spore formation, gene expression [4,5], and, in the case of Yersinia, also to virulence [6], suggesting a common bacterial Ca2+ sense-and-response system. However, while some components of Ca2+ signaling have been identified in bacteria, their number remains small. Ca2+ transporters have been annotated but few experimentally tested. Similarly, Ca2+-binding proteins have been predicted, but few experimentally confirmed. As a result, Ca2+ remains a poorly understood signaling mechanism in bacteria, and many fundamental questions about Ca2+ homeostasis remain unanswered, from the triggers of Ca2+ influx, the Ca2+-binding proteins, to the eventual cellular outcomes.

Mycobacterium tuberculosis (Mtb) is surrounded by a highly impermeable outer cell wall that is composed primarily of the complex phthiocerol dimycocerosates (PDIMs) that form an outer membrane [7], creating a structure not unlike that of the outer membrane of gram-negative bacteria. While transport through the outer membrane in gram-negative bacteria is facilitated by characteristic beta barrel porins [8], no equivalent porins have been identified in Mtb. The impermeable outer membrane and lack of porins raise the question how Mtb transports small molecules such as nutrients, metabolites, but also Ca2+ across the outer membrane. In this way, the very first step in Mtb Ca2+ signaling remains unknown.

The mycobacterial PE/PPE proteins have long been a mystery. They are predominantly found in pathogenic mycobacteria, where they take up a substantial share of the coding capacity [9], and many are substrates of a type VII secretion system [10,11]. The PE/PPE proteins are associated with the outer membrane of the mycobacterial cell wall [12,13]. Many PE/PPE proteins have been implicated in aspects of tuberculosis pathogenesis, but the molecular mechanisms have not been conclusively identified [14]. A recent milestone study showed that several PE/PPE protein pairs function as channels for nutrient transport across the outer mycobacterial membrane [15], defining an idiosyncratic transport system and suggesting a new and perhaps shared family-wide function for the PE/PPE proteins as small molecule transporters.

Here, we sought to further explore Ca2+-mediated processes in Mycobacterium tuberculosis. We identified regulation of ATP levels and biofilm formation by Ca2+. The transcriptional response to Ca2+ was narrow, and the most highly regulated genes were pe15 and ppe20. We show that PE15/PPE20 form a complex, directly bind Ca2+, and facilitate the influx of Ca2+ into Mtb across the outer membrane. These data point to a functional Ca2+ sense-and-response system, identify physiologic processes regulated by Ca2+ and identify a new, specific PE/PPE Ca2+ import system across the outer membrane.


Ca2+ affects ATP levels and biofilm formation

The role of Ca2+ in Mtb physiology is almost entirely unknown. We initially tested for parallels with Ca2+ effects on other bacteria. Ca2+ transport in Escherichia coli depends on ATP, and Ca2+ in turn increases intracellular ATP levels [16]. To test if Ca2+ has a similar effect on ATP levels in Mtb, we exposed Mtb to increasing concentrations of extracellular Ca2+ and measured the intracellular ATP levels. ATP levels increased by >2.5-fold in the presence of 1 mM Ca2+ and by >4-fold in the presence of 10 mM Ca2+ (Fig 1A). These changes are larger than those observed in E. coli, where 10 mM Ca2+ resulted in 30% elevation in ATP [16]. To test the reverse effect, we cultured Mtb in the presence of EGTA, a Ca2+-specific chelator and observed a reduction in ATP levels by approximately 50% (Fig 1A).

Fig 1. Ca2+ affects Mtb cellular processes and down-regulates pe15 and ppe20 transcripts.

(A) Effect of Ca2+ on cellular ATP levels. Mtb was grown in detergent-free CTSM containing increasing concentration of Ca2+ (10 μm–10 mM) or 1 mM EGTA at 37°C, treated with 0.1% Tween-80 and ATP was quantified using BacTiter-Glo reagent. Fold change was calculated by comparing to the Ca2+-free condition. Data are from biological triplicates, error bars show standard deviation. (B) Ca2+ promotes biofilm formation. Biofilms were grown in the presence of increasing concentrations of Ca2+ (10 μm–10 mM) and 1 mM EGTA and were quantified using crystal violet. The experiment was repeated 3 times and a picture of a representative experiment is shown. Error bars show standard error. RNA-seq analysis of Mtb transcripts after exposure to 1 mM Ca2+ for (C) 1 h, (D) 6 h, (E) 1 day, (F) 3 days. Significant down-regulation of pe15 and ppe20 transcripts occurred at 1 and 3 days. Few other genes are significantly changed. The data underlying all the plots in this figure are included in S1 Data. Ca2+, calcium ion; CTSM, Chelex-treated Sauton’s medium.

In some bacteria, a link between biofilm formation and Ca2+ has been proposed [17,18]. To test this idea in Mtb, we measured the generation of biofilm in response to increasing concentrations of Ca2+ in vitro. Indeed, 1 mM Ca2+ led to a 4-fold increase in biofilm when compared to low Ca2+ conditions (Fig 1B). To rule out that this increase is simply a result of increased growth and thus biomass in high Ca2+ conditions, we tested for differences in growth of Mtb in the same Ca2+ concentrations as above. We detected no differences in growth (S1A and S1B Fig), showing that the biofilm effects are specifically due to biofilm generation, not to differences in biomass.

pe15/ppe20 transcripts are down-regulated in response to Ca2+

In eukaryotes and in some bacteria, Ca2+ affects the transcription of a large number of genes [19]. To test for transcriptional effects of Ca2+ in Mtb and to identify genes potentially involved in Ca2+ homeostasis, we grew Mtb with and without 1 mM Ca2+ and determined transcriptional effects by RNA-seq. At early time points, few transcripts changed abundance in response to Ca2+, although we detected a larger response after 3 days of Ca2+ exposure. All changes in transcript abundance are given in Table A in S1 Table. Interestingly, 2 genes with reduced transcript abundance were apparent as early as day 1 and were the most down-regulated genes by day 3: pe15 (Rv1386) and ppe20 (Rv1387) (Fig 1C–1F). This behavior in response to Ca2+ was reminiscent of metal transporter genes that are often regulated in response to changing metal concentrations, with importers typically down-regulated and exporters typically up-regulated in high metal concentrations.

PE15/PPE20 form a complex, bind Ca2+, and localize to the cell wall

Several PE/PPE proteins have been shown to form protein complexes, typically those coding in the same operon [20]. The pe15 and ppe20 genes are also co-operonic, suggesting that they could be a functional protein pair. According to previously published data, recombinant expression of PE15 and PPE20 individually in E. coli failed to produce soluble protein [20]. However, we could readily obtain soluble, recombinant protein when the 2 were expressed together, also indicating a potential interaction. To conclusively show a PE15 and PPE20 interaction, we tested for an association by reciprocal pulldowns with tagged proteins. For all biochemical experiments, we co-expressed His-tagged PE15 and Strep II-tagged PPE20 from a dual expression plasmid and precipitated separately with beads binding to each tag. Both proteins were efficiently pulled down by both beads, indicating binding between PE15 and PPE20 (Fig 2A). To further test the interaction, we co-expressed both proteins and visualized them by native PAGE. Both proteins migrated together, as shown by imaging for the respective tags, further confirming that they form a complex (Fig 2B).

Fig 2. PE15 and PPE20 form a complex and PPE20 binds Ca2+.

(A) Western blot of PE15 with an N-terminal His tag and PPE20 with a C-terminal Strep tag shows the 2 co-purify. W: wash, E: elution. (B) PE15 and PPE20 co-migrate on a native PAGE gel, also indicating complex formation. (C) Thermal shift experiment with western blot readout shows different thermal stability of PPE20 in the absence and presence of Ca2+. The PE15/PPE20 complex was used, but PE15 did not show interpretable difference. (D) PPE20 stability is Ca2+ dose dependent. (E) Band intensities in (D) plotted to estimate the KD of PPE20 for Ca2+ (383 μm). (F) Schematic of equilibrium dialysis experiment. (G) The PE15/PPE20 complex and a known Ca2+-binding protein (CBP) show Ca2+ retention in the protein chambers. Rv0831, a protein not known to bind Ca2+, does not show Ca2+ retention. A total of 10 μm of protein was used and Ca2+ measured by ICP-OES. The data underlying all the plots and uncropped images in this figure are included in S1 Data and S1 Raw images. Ca2+, calcium ion; CBP, calcium-binding protein; ICP-OES, inductively coupled plasma optical emission spectrometry.

To test whether PE15 and/or PPE20 directly bind Ca2+, we incubated recombinantly expressed PE15/PPE20 with Ca2+ and tested for protein stability using a thermal shift assay with a gel readout (Fig 2C). After heating and precipitation, PPE20 showed clear differential stability when incubated with Ca2+, indicating Ca2+ binding. By using different concentrations of Ca2+ in the same assay, we determined a Ca2+ denaturation curve and estimated a KD of 383.7 ± 34.08 μm (Fig 2D and 2E). Interestingly, in a departure from typical ligand-induced stability changes, Ca2+ binding decreased thermal stability of PPE20. PE15 did not produce interpretable results in this assay. Ligand binding in the context of transport suggests that the PE15/PPE20 complex may be a specific channel, as channels typically bind to their ligands [21]. To confirm Ca2+ binding of PE15/PPE20 in an orthogonal assay, we used equilibrium dialysis of Ca2+ in the presence and absence of the PE15/PPE20 complex (Fig 2F). After complete equilibration, the fluid chamber containing PE15/PPE20 retained 30% more Ca2+ than the fluid chamber without protein as determined by inductively coupled plasma optical emission spectrometry (ICP-OES) (Fig 2G). Controls showed a similar effect of a known Ca2+-binding protein and no effect for a protein that does not bind Ca2+ (Figs 2G and S2).

To test if PE15/PPE20 localize to the outer membrane as do other PE/PPE pairs [15,22], we tested for PPE20 expression in different Mtb cell fractions. PPE20 was only detected in the cell wall fraction (Fig 3A).

Fig 3. pe15/ppe20 affects Ca2+ associated processes and facilitates Ca2+ uptake.

(A) Western blot showing localization of PPE20 to the cell wall. Subcellular fractions from the complemented pe15/ppe20 KO strain expressing PPE20-FLAG were probed with α-FLAG antibody. Cyt: cytosolic fraction, CM: cell membrane fraction, CW: cell wall fraction. LAM is a cell envelope marker and FtsZ is a cytosolic marker. (B) PE15/PPE20 affect cellular ATP levels. Mtb was treated with 1 mM Ca2+ at 37°C and ATP quantified using BacTiter-Glo reagent. Fold change was calculated by comparing to the Ca2+-free condition. Data shown are from 4 experiments, error bars show standard deviation (nsp > 0.05, ***p < 0.001). (C) PE15/PPE20 affect biofilm formation. Cultures were grown with increasing concentrations of Ca2+ and biofilm was quantified by crystal violet (D). The experiment was repeated 3 times, error bars show standard error (nsp > 0.05, *p < 0.05, **p < 0.01). The data underlying all the plots and uncropped images in this figure are included in S1 Data and S1 Raw images. Ca2+, calcium ion; Comp, complement; KO, knockout; WT, wild type.

PE15/PPE20 KO reverses Ca2+-dependent phenotypes

To test for a phenotypic link of the PE15/PPE20 protein complex in Ca2+-dependent processes, we generated an Mtb pe15/ppe20 knockout (KO) strain by recombineering [23]. We next tested the pe15/ppe20 KO strain for altered ATP levels. The KO showed reduced Ca2+-dependent increase in ATP production that was reverted to wild-type (WT) levels by complementation with pe15/ppe20 (Comp) (Fig 3B). We next tested whether the PE15/PPE20 complex also affects Ca2+-dependent biofilm formation. The KO blocked the effect of Ca2+ on biofilm formation, and complementation with pe15/ppe20 restored the phenotype (Fig 3C and 3D). These data show that the PE15/PPE20 complex is not only transcriptionally responsive to, but regulates functions related to Ca2+.

The PE15/PPE20 complex is a Ca2+ importer

A recent study showed selective channel function of PE/PPE proteins [15], and the decrease of transcript levels of pe15/ppe20 in the presence of Ca2+ was consistent with the behavior of a Ca2+ channel. To directly test the idea that PE15/PPE20 has a role in Ca2+ transport, we created an Mtb reporter strain using a fluorescence resonance energy transfer (FRET) system that detects intracellular Ca2+, Twitch [24]. Twitch is based on a minimal Ca2+-binding domain from troponin C that is optimized for maximal Ca2+ selectivity and ratiometric signal and that has a KD for Ca2+ of 200 nM [24] (Fig 4A). We ectopically expressed Twitch in the H37Rv background (Mtb-twitch) and tested for Twitch expression in different cell fractions. Twitch was expressed only in the cytoplasm, indicating that it only reports on cytoplasmic Ca2+ (S3A Fig). We could only obtain a robust FRET signal in Ca2+- free Sauton’s medium, not 7H9 medium which already contains Ca2+ (Figs 4B, S1C and S1D). We thus used Sauton’s medium for all Ca2+ measurements. To further test the Twitch reporter in Mtb, we measured the Ca2+ signal at different Ca2+ concentrations. The FRET signal was robust, dose dependent, and comparable to that of Twitch in previously described nonbacterial systems [24] (Fig 4C). We next tested the selectivity of the probe over the closest Earth alkali metal neighbor, Mg2+. Mg2+ did not generate a FRET signal (S3B Fig). These data show that Twitch is a sensitive probe to continuously measure intracellular Ca2+ in Mtb, establish conditions for detecting intracellular Ca2+ changes, show that Mtb readily takes up extracellular Ca2+, and reveal a similar Mtb response to extracellular Ca2+ to that previously observed in E. coli [25].

Fig 4. PE15/PPE20 facilitate Ca2+ uptake.

(A) Schematic of the Ca2+ FRET probe and its ratiometric signal. (B) Ca2+ FRET signals in different media show that Ca2+ presents in standard 7H9 compromise Ca2+ detection. FRET ratio was calculated by calculating the green:blue fluorescence ratio and was plotted against time. (C) The FRET probe generates a robust, dose dependent FRET signal. Emission scan of Mtb-twitch incubated with increasing concentrations of Ca2+ for 30 min at 37°C. (D) FRET trace over time shows reduced Ca2+ influx in the pe15/ppe20 KO strain. The data underlying all the plots in this figure are included in S1 Data. Ca2+, calcium ion; FRET, fluorescence resonance energy transfer; KO, knockout; WT, wild type.

To directly test for a role of the PE15/PPE20 complex in Ca2+ transport, we measured Ca2+ uptake in the Mtb-twitch strain and a strain expressing Mtb-twitch in the pe15/ppe20 KO background. The WT produced a robust FRET signal as before upon addition of 5 mM Ca2+. The KO, however, showed reduced influx, consistent with the loss of a selective import channel (Fig 4D). Complementation of the KO with pe15/ppe20 expressed from an extrachromosomal plasmid restored Ca2+ levels to those in WT. PDIM is one of the most abundant cell wall lipids and is often lost in strains grown in vitro [26]. To rule out PDIM differences between WT and mutant strains that might affect Ca2+ transport, we tested all strains for PDIM. All strains produced PDIM at comparable levels (S3C Fig). To further test if PE15/PPE20’s transport role is associated with the cell wall/outer membrane, we permeabilized the Mtb cell wall by treatment with lysozyme and Triton-X100 [27]. The difference in Ca2+ influx between WT and KO was reduced by permeabilization, indicating that PE15/PPE20 facilitate transport across the outer membrane (S3D Fig).


Ca2+ signaling is ubiquitous in eukaryotes [1] but in bacteria, only few components and functions of Ca2+ signaling have been described. In Mtb, the outer membrane generally prevents passage of charged solutes, presenting a hurdle for Ca2+ uptake not encountered in most bacteria. In the absence of typical porins, the transport processes that allow for transfer through the outer membrane have long been unknown. Here, we identify a specific Ca2+ channel that consists of PE15 and PPE20. In addition, we identify several phenotypes associated with elevated Ca2+ in Mtb: Ca2+ leads to an increase in cellular ATP concentrations, an effect that has also been reported in E. coli and was suggested to be a mechanism to sustain the increased activity of Ca2+ ATPase efflux pumps required to reset Ca2+ levels [16]. We also show a clear contribution of Ca2+ to biofilm formation. The role of biofilms for Mtb pathogenesis and tuberculosis treatment has long been unclear, but recent evidence supports the presence of biofilms in infected lungs of nonhuman primates and human patients and a role in Mtb pathogenesis and drug susceptibility [28].

The PE/PPE proteins have long been the subject of much speculation. They are specific not only to mycobacteria but are predominantly found in pathogenic or slow-growing mycobacteria. Although they make up approximately 10% of Mtb’s genetic coding potential, their function has long remained unclear [14]. Their large number and sequence variation is reminiscent of variable surface proteins that serve as antigenic decoys in other pathogens. Consistent with this idea, the PE/PPE proteins have generally been linked to the cell wall, although mass spectrometry-based proteomic studies not always identify PE/PPE proteins in cell wall fractions. The difficulties to detect PE/PPE proteins by MS are likely due to a combination of the relative lack of trypsin cleavage sites in the repetitive sequences and the challenges to assign repeat-derived peptides to individual PE/PPE proteins [2931]. However, phylogenetic analyses showed that sequence variation does not arise through immunogenic pressure and does not support a role in providing antigenic variation [14]. While mycobacterial ESX secretion systems have been associated with nutrient import before [32,33], the responsible proteins and mechanisms were not known. A recent hallmark study showed channel-like function of PE/PPE complexes specifically transporting several carbon sources, Mg2+, and phosphate across the outer mycobacterial cell membrane [15]. We now show a PE/PPE complex with such a channel-like function for Ca2+. What’s more, we show direct binding of PPE20 to the cargo, which is also more consistent with a channel than for example, a porin. The selectivity of PE15/PPE20 for Ca2+ over even the closely related Mg2+ further argues for a specific channel rather than a porin, which often show more indiscriminate transport [34]. In fact, another PE/PPE pair, PE20/PPE31 specifically transports Mg2+ [15]. Interestingly, several other PE-PGRS proteins bind Ca2+ [35,36], could also contribute to Ca2+ import, and could explain the residual Ca2+ import seen in the pe15/ppe20 KO strain. A link of PE15/PPE20 to pathogenicity is highly plausible given the association of PE/PPE proteins with the outer mycobacterial cell membrane [37], the presence of PE/PPE proteins primarily in pathogenic mycobacteria, their association with the type VII secretion system, and previous studies directly implicating other PE/PPE proteins in virulence [14]. What’s more, the source of Ca2+ for the PE/PPE channel is likely host Ca2+. In this way, Mtb may be able to eavesdrop on the many host cell signaling events that involve Ca2+, for example phagocytosis, which is Ca2+-dependent [38] and marks the beginning of Mtb’s intracellular stage.

How do PE/PPE proteins facilitate transport? Crystal structures show that these proteins, unlike classical porins, fold into long alpha helices, with no structural information available for the long repetitive C-terminal sequences of the PPE proteins that are intrinsically disordered. A potential model for how the alpha helical sections of PE/PPE proteins may form pores was recently suggested by structures of the ESX type VII secretion system component EspB [39], which is a naturally occurring fusion of a PE and a PPE protein that forms a donut-shaped heptamer with a large central pore. Although the helices that potentially traverse the cell membrane did not have the requisite outward-facing hydrophobic residues, other models in which the PPE C-termini provide a hydrophobic sheath or association with other membrane-spanning proteins of the ESX complex are possibilities. In our studies, Ca2+ binding led to an atypical decrease in thermal stability of PPE20. This change could indicate relaxation of the protein concomitant with channel opening. The precise nature of this unfolding event and how it impacts transport remains to be identified, as do the larger transport mechanics of the PE/PPE proteins and how the intrinsically disordered regions of the PPE proteins contribute to transport. Although our data show that PE15/PPE20 are necessary for efficient Ca2+ uptake, they may not be sufficient. The genetic and functional association of the PE/PPE proteins with the ESX secretion system suggests that other ESX components such as an ATPase could be required for PE/PPE transport functions.

The PE and PPE proteins can be further stratified into subfamilies by C-terminal sequence motifs [14] that perhaps also indicate distinct functions. While all currently known PE/PPE channels including PE15 contain a minimal PE-only protein, the PPE proteins identified as components of transporters prior to our study belong to the PPE-SVP subfamily. However, PPE20 belongs to the PPE-PPW subgroup and is the first example of a PE/PPE channel outside of the PPE-SVP group. Similarly, the previous PE/PPE pairs with channel function are associated with the ESX-5 secretion system [40], while PE15/PPE20 is associated with ESX-3 [37]. The PE/PPE channels appear to be highly specific, which could explain their large number, as Mtb requires access to many different nutrients without compromising the outer membrane’s barrier function. The full set of PE/PPE transporters and cargos remains to be identified. However, our data support the emerging theme that the PE/PPE protein family is a mycobacterial transporter family that solves the conundrum of an outer cell membrane so impermeable as to exclude vital nutrients.


Media and growth conditions

Mtb H37Rv was used as a parental strain for generation of all mutants and was cultured at 37°C in either Middlebrook 7H9 medium (Difco) with 10% (vol/vol) oleic acid-albumin-dextrose-catalase (OADC) enrichment (BBL; Becton Dickinson), 0.5% glycerol (referred to as “7H9+GO”), and 0.05% Tween 80 or 0.05% Tyloxapol (referred to as “7H9+GOT or 7H9+GOTy”) or 7H10 agar supplemented with 10% OADC and 0.5% glycerol or Chelex-treated Sauton’s medium (CTSM) with additional supplements as indicated. CTSM consisted of 0.5 g KH2PO4, 4g L-asparagine, 2 g citric acid, 6% glycerol, adjusted to pH 7, and was treated overnight with Chelex-100 resin (10 g/L) (Sigma) to remove trace metal ion contaminants including Ca2+. After filtration, 0.5 g/L of MgSO4, 0.05 g/L ferric ammonium citrate, 0.1 ml of 1% zinc sulfate, and 0.05% Tween 80 or 0.05% Tyloxapol was added. The pH was adjusted to 6.9 and the medium was sterilized by filtration. Strains bearing antibiotic cassettes were cultured with 50 μg/mL hygromycin or 30 μg/mL kanamycin or 25 μg/ml zeocin as appropriate. Ca2+ concentrations in media were determined using ICP-OES (Perkin Elmer Optima 8300). Media were treated with nitric acid (trace metal grade) for 1 h at 65°C and diluted with Chelex-treated water. Ca2+ intensities in each of the digested sample were measured 3 times by ICP-OES at 317.933 nm. Ca2+ standards were measured to generate a standard curve of absorbance versus concentration (S1C Fig). Intensities were converted to Ca2+ concentration using the standard curve.

Biofilm formation

Mtb biofilms were generated and quantified according to a published protocol [41] in 48-well polystyrene plates. Briefly, cells were grown to an OD600 of 0.8–1 in 7H9+GOT medium, washed twice with CTSM, and diluted to an OD600 of 0.01 in CTSM without detergent and added to each well supplemented with varying concentrations of Ca2+. Outer wells were filled with water and the plate was incubated at 37°C for 4 weeks without shaking. The plates were photographed and the processed for crystal violet staining. The medium was removed, biofilms were dried and incubated with 500 μl of 1% crystal violet for 10 min. Wells were washed 3 times with water and dried again. Absolute ethanol (1 ml) was added to each well and incubated for 10 min. Then, 3-fold serial dilutions were read at A600 on a spectrophotometer in a 96-well plate. The represented bar graph is the average readings of 4 biological replicates and Welch’s t test was applied to determine the significance (p-value).

ATP measurement

Mtb cells from a 7H9+GOT medium culture (OD600∼1) were washed twice and diluted to an OD600 of 0.01 in CTSM without detergent supplemented with different concentrations of CaCl2 or 1 mM EGTA. Cultures were grown in a 48-well plate at 37°C for 14 days. Cells were treated with 0.1% Tween-80 overnight to create a homogenous suspension. ATP was quantified by incubating 50 μl of the bacteria in triplicate with 50 μl of the BacTiter-Glo reagent for 15 min followed by a luminescence reading. ATP production in the Ca2+-treated cultures were calculated as fold change compared to the calcium-free condition. The represented bar graph is the average fold change of 4 biological replicates and Welch’s t test was applied to determine the significance (p-value).

RNA sequencing

Mtb H37Rv was grown to an OD600 of approximately 1 in 7H9+GOT medium. Cells were washed twice with CTSM, subcultured in CTSM starting at an OD600 of 0.05, and grown to an OD600 of approximately 1. The culture was again subcultured in CTSM starting at an OD600 of 0.05 and incubated at 37°C to an OD600 of 0.2. The cultures were supplemented with or without 1 mM CaCl2 in triplicate. At the indicated times, cells were pelleted at 4,000 g for 5 min at 4°C, resuspended in Trizol and lysed by bead beating for 30 s at 6 m/s for 3 cycles with intermittent cooling on ice. Cell debris were pelleted at 20,000 g for 1 min, and the supernatant was transferred to a heavy phase lock gel tube containing 300 μl chloroform. The tubes were inverted for 2 min and centrifuged at 20,000 g for 5 min. RNA in the aqueous phase was precipitated using 300 μl isopropanol and 300 μl high salt solution (0.8 M Na citrate, 1.2 M NaCl). RNA was purified using QIAGEN RNeasy kit and ribosomal RNA was depleted using the Ribo-Zero rRNA removal magnetic kit (Illumina). The cDNA library was generated using the NEBNext Ultra II RNA Library Prep Kit, and each replicate was barcoded in the DNA library using the NEBNext Multiplex Oligos for Illumina. Libraries were quantified using the KAPA qPCR quantification kit, pooled, and sequenced at the University of Washington Northwest Genomics Center with the Illumina NextSeq 500 High Output v2 Kit. Read alignment was performed using the Bowtie 2 custom processing pipeline (, Gene expression changes were identified using a combination of 5 differential expression (DE) tools within DuffyTools. The 5 DE tools included round robin, RankProduct, significance of microarrays (SAMs), EdgeR, and DeSeq2. Each DE tool measurement was combined using the weighted average of fold change and significance (p-value). Genes with averaged absolute fold change more than 1.5-fold and p-value <0.01 were considered differentially expressed. RNA-seq data is available at NCBI-GEO (accession no. GSE214266).

Cloning, co-expression, and purification of recombinant PE15 and PPE20 proteins

pETDuet dual expression plasmid was used to co-express PE15 and PPE20. PE15 was cloned in the MCS-1 region with N-terminal His tag and PPE20 was cloned in the MCS-2 region with a C-terminal Strep II tag. Mtb Rv1386 (PE15) and Rv1387 (PPE20) genes were amplified from Mtb H37Rv genomic DNA using the primers Duet 1–5 that included Gibson overlap sequence (Table B in S1 Table). The linearized pETDuet plasmid and the region between MCS-1 and MCS-2 were prepared by PCR amplification using the primers Duet 6,7 and Duet 8,9, respectively (Table B in S1 Table). Finally, all 4 purified PCR products, i.e., PE15, PPE20, region between MCS-1/MCS-2, and linearized pETDuet plasmid were ligated using the Gibson Assembly (NEB) to generate the pETDuet pe15/ppe20 plasmid. Cloning was confirmed by sequencing. The vector was transformed into E. coli BL21(DE3) and a single colony was picked and was grown at 37°C in Terrific Broth medium containing ampicillin. Protein expression was induced by addition of 0.5 mM isopropyl-β-d-thiogalactoside (IPTG) at OD600 of approximately 0.4; culture was then maintained at 16°C for 16 h. Cells were harvested, and pellets were processed for purification. For complex purification through His-PE15, pellets were resuspended in buffer A (50 mM Tris (pH 8.0), 150 NaCl, and 10% glycerol) containing 20 mM imidazole and 1 mM AEBSF and lysed by sonication. Lysate was centrifuged at 35,000 g for 30 min at 4°C and the supernatant was loaded on Ni-NTA column. The column was thoroughly washed with buffer A containing 20 mM imidazole followed by elution with buffer A containing 250 mM imidazole. For PPE20-strep II tag, pellets were resuspended in buffer A containing 1 mM AEBSF and lysed by sonication. Lysate was centrifuged at 35,000 g for 30 min at 4°C and the supernatant was loaded on Strep-tactin column. The column was thoroughly washed with buffer A followed by elution with buffer A containing 2.5 mM desthiobiotin. Both the purified proteins were immediately dialyzed against buffer A and stored at −80°C. Interaction of PE15/PPE20 was confirmed by western blot. The wash and elution fraction from the Ni-NTA and strep tag purification were loaded on either SDS-PAGE or Native-PAGE, and the proteins were transferred onto a nitrocellulose membrane. Blots were probed with mouse α-His antibody followed by IRDye 680RD Goat α-Mouse IgG Secondary Antibody (LI-COR) to detect the His tagged PE15 and were also probed with IRDye 800CW Streptavidin to detect the PPE20-strep tag. The blots were scanned on LI-COR Odyssey platform using the 700 nm and 800 nm channel.

Thermal shift assay and equilibrium dialysis to determine Ca2+ binding

Purified PE15/PPE20 protein was incubated with or without 1 mM CaCl2 in a thermocycler for 5 min at 25°C followed by incubation at varying temperature (55 to 70°C) for 3 min. The reaction was cooled to 4°C and centrifuged at 20,000 g for 30 min at 4°C to separate the native (supernatant) and the denatured (pellet) fractions. Pellet fraction was solubilized by boiling with 2xSDS loading buffer. PPE20 in each of the supernatant and pellet fraction was detected by western blot using rabbit α-strep II tag antibody followed by IRDye 800CW Goat α-Rabbit IgG Secondary Antibody. To determine the KD value, PPE20 was incubated with varying concentration of CaCl2 (12.5 μm to 3.2 mM) for 5 min at 25°C, 3 min at 65°C followed by incubation at 4°C. PPE20 in the pellet fraction was detected by western blot as described above. The band intensities were calculated using the Image studio software and converted to fraction denatured relative to the total protein. The graph of fraction denatured v/s log Ca2+ concentration was plotted using GraphPad Prism 9.3, and the KD value was determined by fitting a dose-response curve using the least square regression method. For equilibrium dialysis, Ca2+-free PE15/PPE20 was prepared by EGTA treatment and dialysis against Chelex-treated buffer A. Protein (10 μm) was placed in the upper chamber and dialyzed against 50 mM Tris (pH 7.4), 150 mM NaCl buffer containing 200 μm of Ca2+ in the lower chamber for 24 h and the Ca2+ concentration in both chambers was determined by ICP-OES. Samples (100 μl) were digested with 100 μl of nitric acid (trace metal grade) for 1 h at 65°C and diluted with 2.8 ml of Chelex-treated water. Ca2+ intensities were measured 3 times by ICP-OES and converted to Ca2+ concentration using the standard curve described above. Percentage ΔC was calculated as per the following formula:

The thermal shift and equilibrium dialysis experiments were repeated with buffer control (no protein), calcium-binding protein (CBP) from Encephalitozoon cuniculi as a positive control (obtained from SSGCID # EncuA.01276.a.A1.PS00484), and Rv0831c as a negative control. The bar graph in Fig 2G is the average %ΔC of 3 technical replicates. Student’s t test was applied to determine the p-value.

Creation of pe15/ppe20 knock out and complemented strain in Mtb H37Rv

The KO strain was created by recombineering as described previously [23]. Initially 500 bp upstream of pe15 (primers KO1 and KO2), 500 bp downstream of ppe20 (primers KO3 and KO4) and hygromycin cassette (primers KO5 and KO6) were amplified separately (Table B in S1 Table). Both PCR fragments were Gibson ligated to the 5′ and 3′ of a hygromycin-resistance cassette, respectively, to generate the recombineering cassette. This linear recombineering cassette was PCR amplified, purified, and electroporated into Mtb H37Rv strain carrying recombineering plasmid pNIT [42]. Hygromycin positive colonies were screened, and the positive clone was confirmed by DNA sequencing. For the complemented strain, both pe15 (His-tag) and ppe20 (FLAG-tag) genes were separately PCR-amplified using primers Comp 1–3 and Comp 4, 5, respectively (Table B in S1 Table), and Gibson cloned in pDTCF plasmid (Zeocin) separately under an anhydrotetracycline (ATc)-inducible promoter. The ATc promoter along with the ppe20 gene was amplified using primers Comp 6, 7 (Table B in S1 Table) from pDTCF-PPE20 vector and Gibson cloned in pDTCF-PE15 vector downstream of pe15 gene. The resulting plasmid pDTCF-pe15/ppe20 carries both genes with individual ATc promoters. The plasmid was electroporated into the KO strain and positive clones were selected by growth in hygromycin and zeocin. This strain expressed His-PE15 and PPE20-FLAG proteins when induced with 100 ng/ml ATc.

Creation of Mtb Ca2+ FRET reporter strain

The Ca2+ FRET reporter Twitch-2B was PCR amplified from the plasmid Twitch-2B pRSETB obtained from Addgene [24] using the primers T2B1 and T2B2 (Table B in S1 Table) that included Gibson overlap sequence. The PCR product was Gibson cloned in the E. coli-mycobacterial episomal shuttle vector pDTCF (kanamycin) under the anhydrotetracycline (ATc)-inducible promoter. The plasmid was electroporated into wild type Mtb H37Rv (WT), pe15/ppe20 KO and pe15/ppe20 complemented strain (Comp), and the expression of Twitch was induced using 100 nM ATc. The expression of Twitch was confirmed by measuring the fluorescence of uninduced and induced cells using a plate reader at 3 different wavelengths (432/475 nm, 432/532 nm, and 488/532 nm). The localization of Twitch and PPE20 was confirmed by western blot. Briefly, 100 ml culture of uninduced and induced Mtb-twitch complemented with the pe15/ppe20 genes was grown in 7H9+GO medium. Cells were washed twice in PBS and resuspended in lysis buffer (50 mM Tris (pH 7.4), 150 mM NaCl) containing protease inhibitor cocktail. Cells were lysed by bead beating and the following fractions were prepared according to published protocol [43]: cytosolic (CYT), cell membrane (CM), and cell wall (CW) fraction. Briefly, the lysate was sequentially centrifuged at 3,000 g to separate the beads and cell debris, followed by spin at 20,000 g to pellet the CW fraction and finally at 100,000 g to separate the pellet, i.e., CM fraction and supernatant CYT fraction. CM and CW fractions were washed once and resuspended in lysis buffer. The expression of Twitch protein (His-tagged) was detected by western blot using a mouse α-His antibody, PPE20 was detected by mouse α-Flag antibody and the purity of cellular fractions was confirmed by using α-LAM (CM and CW fraction) and α-FtsZ (CYT) antibodies (obtained from BEI resources, NIAID, NIH).

PDIM analysis

Cultures were grown in 50 ml 7H9+GO for 7 days and washed 3× with PBS. Cell pellets were resuspended in 9 ml chloroform-methanol (2:1) and incubated overnight at room temperature with shaking. The extract was filtered through 0.2 μm PTFE filters and dried under a stream of nitrogen. The total lipid was weighed and resuspend at a concentration of 20 mg/ml in chloroform-methanol (2:1). Lipids (500 μg) were spotted onto aluminum-backed TLC plate and run in petroleum ether-ethyl acetate (98:2) as the solvent system. Control PDIM was obtained from BEI resources, solubilized in chloroform-methanol (2:1), and 100 μg was run on TLC. The TLC plate was dried, and the lipids were stained by spraying 5% sulfuric acid in methanol followed by heating.

Ca2+ uptake assay

Mtb-twitch was grown in either 7H9+GOT or CTSM to an OD600 of 1. Cells (50 μl) were aliquoted in duplicate wells and spiked with Ca2+ (5 mM) and EGTA (5 mM) at indicated time points and fluorescence was continuously measured at wavelengths 432/475 nm (blue) and 432/532 nm (green) at 37°C. Fluorescence readings were corrected by subtracting the readings from wells containing uninduced culture. The FRET ratio was calculated as the green:blue fluorescence ratio and plotted against time. An increase in the FRET ratio indicates an increased intracellular Ca2+ concentration. To determine the dose-dependent effect of Ca2+ on the FRET signal, Mtb-twitch was incubated with increasing concentration of Ca2+ for 30 min at 37°C and an emission scan (from Em450-550 nm) was recorded using excitation at 432 nm.

Mtb H37Rv (WT), pe15/ppe20 KO and complemented strains (Comp) expressing twitch were grown to OD600 of approximately 0.5 in CTSM/tyloxapol with or without 100 nM ATc. Cells were washed twice and resuspended in CTSM/tyloxapol at a dilution of 1 OD/ml. Cells were spiked with Ca2+ (5 mM) at indicated time point and blue and green fluorescence were measured continuously at 37°C. FRET ratios were calculated and plotted against time. The assay was repeated by treating cells with permeabilizing reagent [27] (0.1% Triton-X100 and 2 μg/μl lysozyme) at time 0 min. Experiments were performed using 3 biological replicates and 2 technical replicates.

Supporting information

S1 Fig. Growth curves of Mtb H37Rv and Ca2+ quantification in different media.

(A) CTSM and (B) 7H9 medium containing glycerol, OADC, and Tween-80 (7H9+GOT) supplemented with different concentrations of CaCl2 shows no effect of Ca2+ on growth. (C) ICP-OES standard curve for the determination of Ca2+ concentrations. (D) Ca2+ concentrations in standard 7H9 and Sauton’s media. The data underlying all the plots in this figure are included in S1 Data. Ca2+, calcium ion; CTSM, Chelex-treated Sauton’s medium; ICP-OES, inductively coupled plasma optical emission spectrometry.


S2 Fig. Thermal shift-based analysis of Ca2+ binding to control proteins.

(A) A known CBP from Encephalitozoon cuniculi shows a Ca2+-dependent increase in the melting temperature and concomitant precipitation. (B) A protein not known to bind Ca2+ shows no difference in melting temperature and resulting precipitation. Uncropped images in this figure are included in S1 Raw images. Ca2+, calcium ion; CBP, calcium-binding protein.


S3 Fig. Transport by PE15/PPE20 is associated with the cell wall and is specific for Ca2+.

(A) Western blot showing localization of Twitch-2B (His tagged) in the cytosolic fraction. The same cell fractions as in Fig 3A were used, see controls therein. Cyt: cytosolic fraction, CM: cell membrane fraction, CW: cell wall fraction. (B) The FRET signal is highly specific for Ca2+ over the congener Mg2+. The His-tagged Twitch protein was purified from the cytosolic fraction using Ni-NTA column, treated with EDTA and desalted. An emission scan of the purified protein was recorded upon incubation with either Ca2+ or Mg2+. (C) Thin-layer chromatography of the cell wall component PDIM shows no difference in PDIM content between WT and pe15/ppe20 KO strains. (D) Permeabilization of the cell wall diminishes PE15/PPE20’s effect on Ca2+ import. WT, pe15/ppe20 KO, and complemented strain were transformed with Twitch, permeabilized (P) with lysozyme and Triton-X100, and the Ca2+-FRET signal measured. The data underlying all the plots and uncropped images in this figure are included in S1 Data and S1 Raw images. Ca2+, calcium ion; FRET, fluorescence resonance energy transfer; KO, knockout; PDIM, phthiocerol dimycocerosate; WT, wild type.


S1 Table. List of differentially expressed genes and primers.

(A) Differentially expressed genes in response to Ca2+ in Mtb and (B) primers used in this study.


S1 Data. Numerical data points underlying presented graphs.



Recombinant calcium-binding protein from Encephalitozoon cuniculi was provided by the Seattle Structural Genomics Center for Infectious Disease ( that is supported Federal Contract No. HHSN272201700059C from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services.


  1. 1. Clapham DE. Calcium signaling. Cell. 2007;131(6):1047–1058. pmid:18083096
  2. 2. Carafoli E, Krebs J. Why calcium? How calcium became the best communicator. J Biol Chem. 2016;291(40):20849–20857. pmid:27462077
  3. 3. Domínguez DC, Guragain M, Patrauchan M. Calcium binding proteins and calcium signaling in prokaryotes. Cell Calcium. 2015;57(3):151–165. pmid:25555683
  4. 4. Norris V, Chen M, Goldberg M, Voskuil J, McGurk G, Holland IB. Calcium in bacteria: a solution to which problem? Mol Microbiol. 1991;5(4):775–8. Epub 1991/04/01. pmid:1857203.
  5. 5. Smith RJ. Calcium and bacteria. Adv Microb Physiol. 1995;37:83–133. Epub 1995/01/01. pmid:8540424.
  6. 6. Straley SC, Plano GV, Skrzypek E, Haddix PL, Fields KA. Regulation by Ca2+ in the Yersinia low-Ca2+ response. Mol Microbiol. 1993;8(6):1005–10. Epub 1993/06/01. pmid:8361348.
  7. 7. Jackson M. The mycobacterial cell envelope-lipids. Cold Spring Harb Perspect Med. 2014;4(10):a021105. pmid:25104772
  8. 8. Galdiero S, Falanga A, Cantisani M, Tarallo R, Elena Della Pepa M, D’Oriano V, et al. Microbe-host interactions: structure and role of Gram-negative bacterial porins. Curr Protein Pept Sci. 2012;13(8):843–854. pmid:23305369
  9. 9. Cole S, Brosch R, Parkhill J, Garnier T, Churcher C, Harris D, et al. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature. 1998;396(6707):190–190.
  10. 10. Abdallah AM, Gey van Pittius NC, DiGiuseppe Champion PA, Cox J, Luirink J, Vandenbroucke-Grauls CMJE, et al. Type VII secretion—mycobacteria show the way. Nat Rev Microbiol. 2007;5(11):883–891. pmid:17922044
  11. 11. Ates LS, Houben ENG, Bitter W. Type VII secretion: a highly versatile secretion system. Microbiol Spectr. 2016;4(1):4.1:12. pmid:26999398
  12. 12. Burggraaf MJ, Ates LS, Speer A, Van Der Kuij K, Kuijl C, Bitter W. Optimization of secretion and surface localization of heterologous OVA protein in mycobacteria by using LipY as a carrier. Microb Cell Fact. 2019;18(1):1–14.
  13. 13. Cascioferro A, Delogu G, Colone M, Sali M, Stringaro A, Arancia G, et al. PE is a functional domain responsible for protein translocation and localization on mycobacterial cell wall. Mol Microbiol. 2007;66(6):1536–1547. pmid:18028308
  14. 14. Fishbein S, Van Wyk N, Warren RM, Sampson SL. Phylogeny to function: PE/PPE protein evolution and impact on M ycobacterium tuberculosis pathogenicity. Mol Microbiol. 2015;96(5):901–916. pmid:25727695
  15. 15. Wang Q, Boshoff HIM, Harrison JR, Ray PC, Green SR, Wyatt PG, et al. PE/PPE proteins mediate nutrient transport across the outer membrane of Mycobacterium tuberculosis. Science. 2020;367(6482):1147–1151. pmid:32139546
  16. 16. Naseem R, Wann KT, Holland IB, Campbell AK. ATP regulates calcium efflux and growth in E. coli. J Mol Biol. 2009;391(1):42–56.
  17. 17. Bilecen K, Yildiz FH. Identification of a calcium-controlled negative regulatory system affecting Vibrio cholerae biofilm formation. Environ Microbiol. 2009;11(8):2015–2029. pmid:19397680
  18. 18. Sarkisova S, Patrauchan MA, Berglund D, Nivens DE, Franklin MJ. Calcium-induced virulence factors associated with the extracellular matrix of mucoid Pseudomonas aeruginosa biofilms. J Bacteriol. 2005;187(13):4327–4337.
  19. 19. Oomes S, Jonker MJ, Wittink FRA, Hehenkamp JO, Breit TM, Brul S. The effect of calcium on the transcriptome of sporulating B. subtilis cells. Int J Food Microbiol. 2009;133(3):234–242. pmid:19552981
  20. 20. Strong M, Sawaya MR, Wang S, Phillips M, Cascio D, Eisenberg D. Toward the structural genomics of complexes: crystal structure of a PE/PPE protein complex from Mycobacterium tuberculosis. Proc Natl Acad Sci U S A. 2006;103(21):8060–8065.
  21. 21. Nikaido H. Porins and specific channels of bacterial outer membranes. Mol Microbiol. 1992;6(4):435–42. Epub 1992/02/01. pmid:1373213.
  22. 22. Mitra A, Speer A, Lin K, Ehrt S, Niederweis M. PPE surface proteins are required for heme utilization by Mycobacterium tuberculosis. mBio. 2017;8:e01720–e01716. pmid:28119467
  23. 23. Van Kessel JC, Hatfull GF. Recombineering in Mycobacterium tuberculosis. Nat Methods. 2007;4(2):147–152.
  24. 24. Thestrup T, Litzlbauer J, Bartholomäus I, Mues M, Russo L, Dana H, et al. Optimized ratiometric calcium sensors for functional in vivo imaging of neurons and T lymphocytes. Nat Methods. 2014;11(2):175–182. pmid:24390440
  25. 25. Jones HE, Holland IB, Baker HL, Campbell AK. Slow changes in cytosolic free Ca2+ inEscherichia colihighlight two putative influx mechanisms in response to changes in extracellular calcium. Cell Calcium. 1999;25(3):265–274. pmid:10378087
  26. 26. Domenech P, Reed MB. Rapid and spontaneous loss of phthiocerol dimycocerosate (PDIM) from Mycobacterium tuberculosis grown in vitro: implications for virulence studies. Microbiology (Reading). 2009;155(Pt 11):3532–43. Epub 2009/08/08. pmid:19661177; PubMed Central PMCID: PMC5154741.
  27. 27. Cimino M, Alamo L, Salazar L. Permeabilization of the mycobacterial envelope for protein cytolocalization studies by immunofluorescence microscopy. BMC Microbiol. 2006;6(1):1–3. pmid:16620389
  28. 28. Chakraborty P, Bajeli S, Kaushal D, Radotra BD, Kumar A. Biofilm formation in the lung contributes to virulence and drug tolerance of Mycobacterium tuberculosis. Nat Commun. 2021;12(1):1–17.
  29. 29. Malen H, Pathak S, Softeland T, de Souza GA, Wiker HG. Definition of novel cell envelope associated proteins in Triton X-114 extracts of Mycobacterium tuberculosis H37Rv. BMC Microbiol. 2010;10:132. Epub 2010/05/01. pmid:20429878; PubMed Central PMCID: PMC2874799.
  30. 30. Mawuenyega KG, Forst CV, Dobos KM, Belisle JT, Chen J, Bradbury EM, et al. Mycobacterium tuberculosis functional network analysis by global subcellular protein profiling. Mol Biol Cell. 2005;16(1):396–404. Epub 2004/11/05. pmid:15525680; PubMed Central PMCID: PMC539182.
  31. 31. Wolfe LM, Mahaffey SB, Kruh NA, Dobos KM. Proteomic definition of the cell wall of Mycobacterium tuberculosis. J Proteome Res. 2010;9(11):5816–26. Epub 2010/09/10. pmid:20825248; PubMed Central PMCID: PMC2995261.
  32. 32. Ates LS, Ummels R, Commandeur S, van de Weerd R, Sparrius M, Weerdenburg E, et al. Essential Role of the ESX-5 Secretion System in Outer Membrane Permeability of Pathogenic Mycobacteria. PLoS Genet. 2015;11(5):e1005190. Epub 2015/05/06. pmid:25938982; PubMed Central PMCID: PMC4418733.
  33. 33. Siegrist MS, Unnikrishnan M, McConnell MJ, Borowsky M, Cheng TY, Siddiqi N, et al. Mycobacterial Esx-3 is required for mycobactin-mediated iron acquisition. Proc Natl Acad Sci U S A. 2009;106(44):18792–7. Epub 2009/10/23. pmid:19846780; PubMed Central PMCID: PMC2774023.
  34. 34. Prajapati JD, Kleinekathöfer U, Winterhalter M. How to enter a bacterium: bacterial porins and the permeation of antibiotics. Chem Rev. 2021;121(9):5158–5192. pmid:33724823
  35. 35. Sharma T, Singh J, Grover S, PM, Firdos F, Alam A, et al. PGRS Domain of Rv0297 of Mycobacterium tuberculosis Functions in A Calcium Dependent Manner. Int J Mol Sci. 2021;22(17). Epub 2021/09/11. pmid:34502303; PubMed Central PMCID: PMC8430768.
  36. 36. Yeruva VC, Kulkarni A, Khandelwal R, Sharma Y, Raghunand TR. The PE_PGRS Proteins of Mycobacterium tuberculosis Are Ca(2+) Binding Mediators of Host-Pathogen Interaction. Biochemistry. 2016;55(33):4675–87. Epub 2016/08/03. pmid:27483162.
  37. 37. Tufariello JM, Chapman JR, Kerantzas CA, Wong K-W. Vilchà ze C, Jones CM, et al. Separable roles for Mycobacterium tuberculosis ESX-3 effectors in iron acquisition and virulence. Proc Natl Acad Sci U S A. 2016;113(3):E348–E357.
  38. 38. Melendez AJ. Calcium signaling during phagocytosis. Molecular mechanisms of phagocytosis. Springer; 2005. p. 117–32.
  39. 39. Jrm P, Pojer F, Wakatsuki S, Gati C, Cole ST. High resolution CryoEM structure of the ring-shaped virulence factor EspB from Mycobacterium tuberculosis. J Struct Biol X. 2020;4:100029. pmid:32875288
  40. 40. Ehtram A, Shariq M, Ali S, Quadir N, Sheikh JA, Ahmad F, et al. Teleological cooption of Mycobacterium tuberculosis PE/PPE proteins as porins: Role in molecular immigration and emigration. Int J Med Microbiol. 2021;311(3):151495.
  41. 41. Pang JM, Layre E, Sweet L, Sherrid A, Moody DB, Ojha A, et al. The polyketide Pks1 contributes to biofilm formation in Mycobacterium tuberculosis. J Bacteriol. 2012;194(3):715–721.
  42. 42. Murphy KC, Papavinasasundaram K, Sassetti CM. Mycobacterial recombineering. Chromosomal. Mutagenesis. 2015:203–215.
  43. 43. Rezwan M, Lanéelle M-A, Sander P, Daffé M. Breaking down the wall: fractionation of mycobacteria. J Microbiol Methods. 2007;68(1):32–39. pmid:16839634