The P-type ATPase CtpF is a plasma membrane transporter mediating calcium efflux in Mycobacterium tuberculosis cells

Abstract

Among the 12 P-type ATPases encoded by the genome of Mycobacterium tuberculosis(Mtb), CtpF responds to the greatest number of stress conditions, including oxidative stress, hypoxia, and infection. CtpF is the mycobacterial homolog of the sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA) of higher eukaryotes. Its expression is regulated by the global regulator of latency, DosR. However, the role that CtpF plays in the mycobacterial plasma membrane remains unknown. In this study, different functional analyses showed that CtpF is associated with calcium pumping from mycobacterial cells. Specifically, Mtb CtpF expression in Mycobacterium smegmatis cells prevents Ca²⁺ accumulation compared with wild type (WT) cells. In addition, plasma membrane vesicles from recombinant membranes, in which the direction of ion transport is inverted, accumulate more Ca²⁺ compared with vesicles obtained from the WT strain. This findings support the hypothesis that CtpF contributes to calcium efflux from mycobacterial cells. Accordingly, Mtb cells defective in ctpF (MtbΔctpF) accumulate more Ca²⁺ compared with WT cells, while the Ca²⁺-dependent ATPase activity is significantly lower in the mutant cells. Interestingly, the deletion of ctpF in Mtb impairs the tolerance of the bacteria to oxidative and nitrosative stress. Overall, our results indicate that CtpF is associated with calcium pumping from mycobacterial cells and the response to oxidative stress.

Bacteria; Biochemistry; Biomolecules; Microbiology; Microorganism; Bioinformatics; Mycobacterium tuberculosis; Mycobacterium smegmatis; P-type ATPases; Plasma membrane; CtpF; Ca2+ efflux.

1. Introduction

Tuberculosis (TB) is produced by the acid-fast bacillus Mtb and is one of the top 10 causes of death worldwide. In 2017, there were 6.4 million new cases reported and 1.6 million deaths by TB [1]. The incidence of TB has increased as a result of the emergence of multidrug and extensively resistant (MDR and XDR) mycobacterial strains, Mtb-HIV coinfection, and the ineffectiveness of the Bacillus Calmette–Guérin (BCG) vaccine [1, 2]. Therefore, the search for alternative control strategies is a priority that relies on a better understanding of the molecular mechanism used by Mtb to succeed as intracellular pathogen. In this sense, the role played by cell membrane proteins and transporters in the mycobacterial interaction with the host cell environment is pivotal. Previous studies have suggested the relevance of P-type ATPases in the mycobacterial physiology and host-pathogen interaction [3].

P-type ATPases are a large family of membrane proteins relevant for maintaining cellular homeostasis and generating appropriate electrochemical gradients for cell survival. These enzymes use the energy released by ATP hydrolysis to catalyze the transport of cations across the cell membrane [3, 4, 5, 6]. In fact, P-type ATPases are expressed during mycobacterial infection as a response to the toxicity produced by high levels of metals in human macrophages [7, 8, 9]. There are reports of diminished vacuolar concentration of Ca²⁺ (1.8 $\pm$ 1.3 mM) and K⁺ (19.5$\pm$16.9 mM) in the early phagosome (first hour after phagocytosis) as compared with extracellular bacteria [7, 10]. However, the concentrations of Ca²⁺ (7.1 $\pm$3.3mM) and K⁺ (51.0 $\pm$ 28.6mM) as well as of other metals such as copper, zinc, and iron are replenished or increased 24 h post-infection [7, 10]. Therefore, P-type ATPases, among other systems, play a critical role in maintain mycobacterial metal homeostasis during infection [7, 8, 9, 11, 12, 13, 14].

P-type ATPases are classified into five subfamilies (P₁–P₅), based on ionic specificity and structural characteristics [4, 6]: P_1A-type bacterial potassium transporters; P_1B-type heavy metal pumps; P₂-type alkaline/alkaline earth metal transporters; P_3A-type H⁺ pumps; P_3B-type bacterial Mg²⁺ pumps; P₄-type putative lipid flippases; and the uncharacterized P₅-type ATPase pumps [4, 5, 6, 15]. Bioinformatic studies identified 12 P-type ATPases in the Mtb genome, namely: seven P_1B-type, four P₂-type, and one P_1A-type ATPases [16, 17, 18]. Regarding the functional characterization of these Mtb plasma membrane pumps, CtpA is a Cu⁺ transporter [19], CtpC transports Mn²⁺ and/or Zn²⁺ across the plasma membrane [8, 13], CtpJ and CtpD are Fe²⁺, Co²⁺, and Ni²⁺ transporters [12, 14], CtpV is a Cu⁺ pump [9], and CtpG is a Cd²⁺ transporter [20]. However, the ion specificity of P₂-type ATPases and/or their possible roles in mycobacteria are unknown.

Unlike eukaryotic cells, there is not a wide variety of calcium-mediated processes in bacteria [21]. However, critical physiological processes such as cellular growth, motility, quorum sensing, sporulation, and the development of different bacterial structures are regulated by the cytosolic Ca²⁺ concentration in bacteria [21]. All forms of life, even mycobacteria, developed mechanisms such as passive and active transporters, ion channels, and non-protein channels to regulate calcium homeostasis [21], including Ca²⁺-ATPases that mediate calcium homeostasis [22, 23].

Mtb CtpF is the mycobacterial P₂-type ATPase most closely related to the sarco/endoplasmic reticulum calcium ATPase (SERCA1a) from eukaryotes [16]. The expression of the ctpF gene is regulated by the global mycobacterial dormancy regulator DosR [24, 25, 26, 27, 28, 29, 30, 31]. Interestingly, ctpF is activated under conditions similar to the phagosome environment and during infection [28, 32]. Specifically, ctpF responds when Mtb cells are treated with toxic substances, such as isoxyl, tetrahydrolipstatin [33], reactive nitrogen species (RNS), reactive oxygen species (ROS) [28, 29, 30, 34, 35], and under hypoxia [27, 29, 36, 37, 38]. This transcriptional behavior suggests that CtpF could be part of the strategies of the tubercle bacillus uses to face the environmental conditions encountered during infection [39]. However, the actual role of CtpF in mycobacterial ion homeostasis and its biology remains unknown. Which is the cation transported by CtpF? Are there any phenotypic consequences of ctpF overexpression and deletion in mycobacteria?

In this work, we assessed the role of CtpF in Ca²⁺ pumping from mycobacterial cells. Initially, the ATPase activity in the plasma membrane together with the calcium accumulation in whole cells and plasma membrane vesicles from recombinant M. smegmatis overexpressing Mtb CtpF and Mtb cells defective in ctpF (MtbΔctpF) confirmed that this transporter is associated with Ca²⁺ pumping from mycobacterial cells. In addition, MtbΔctpF cells were more susceptible to oxidizing agents, suggesting a link between Ca²⁺ transport and the mycobacterial response to oxidative stress.

2. Materials and methods

2.1. Bacterial strains and growth conditions

The bacterial strains, plasmids, and primers used in this study are listed in Tables 1 and 2. Mtb strains were grown in Middlebrook 7H9 broth supplemented with oleic acid-albumin-dextrose-catalase (OADC) (50 μg/mL oleic acid, 0.5 % Bovine albumin Fraction V, 0.2 % dextrose and 0.004 % catalase) and 0.5 % glycerol at 37 °C with gentle agitation (80 rpm) until OD₆₀₀ = 0.5–0.8, or on 7H10 and 7H11 agar plates supplemented with OADC and glycerol. For the experiments of bacterial tolerance to cations, the mycobacteria were grown in Sauton's medium (pH = 7.4) supplemented with 0.05 % Tween 80 and 0.2 % glucose at 37 °C and 80 rpm. Escherichia coli DH5α and TOP10, used for plasmid propagation, were cultured at 37 °C in LB broth with agitation (180 rpm) or on LB agar plates. When required, 7H9, 7H10, and 7H11 were supplemented with 20 μg/mL kanamycin (Kan), 100 μg/mL hygromycin (Hyg), while LB was supplemented with 100 μg/mL ampicillin (Amp), 0.5 mM isopropyl-β-D-thiogalactopyranoside (IPTG), 80 μg/mL X-gal, or 20 μg/mL kanamycin (Kan). Mtb genomic DNA was isolated as previously reported [40].

Table 1.

Bacterial strains and plasmids used in this study.

Strains	Relevant features	Reference
Mycobacterium tuberculosis
H37Ra	Slow-growing attenuated strain, AmpR, ChxR, CbR	ATCC 25177
H37Ra:pJV53	Recombineering strain (with pJV53), AmpR, ChxR, CbR, KmR	This study
H37RaΔctpF	ΔctpF, gene replaced by a HygR cassette	This study
Mycobacterium smegmatis
mc2155	Fast-growing, usually non-pathogenic Mycobacterium	ATCC 700084
Escherichia coli
DH5α	recA-, endA-, Blue/white color screening with lacZΔM15	Thermo Fisher Scientific
TOP10	F– mcrA Δ(mrr-hsdRMS-mcrBC) ϕ80lacZΔM15 ΔlacX74 recA1 araD139 Δ(araA-leu)7697 galU galK rpsL endA1 nupG	Thermo Fisher Scientific
Plasmids
pMV261	E. coli-Mycobacterium shuttle vector, hsp60 promoter, KmR	[57]
pLNA29	Mtb Rv1997 (ctpF) cloned into pMV261, KmR	This study
pJV53	Derivative of pLAM12 with Che9c 60–61 genes under control of the acetamidase promoter	Gift from Unizar [43]
pYUB854	HygR cassette is flanked by the γδ-res sites and by two MCSs for directional cloning of the homologous recombination substrates	Gift from Unizar [58]
pLNA22	607 bp upstream and 520 bp downstream of Mtb Rv1997 (ctpF) in pYUB854	This study

Table 2.

List of oligonucleotides used in this study.

Primer	Sequence (5′-3′)
F-RT Dir	CAGTGATCTTCGGTGTGGTG
F-RT Rev	GATTGAGCGTGAACGAGTCA
pMVComp Up	CAGCGAGGACAACTTGAGC
pMVComp Down	CGACTGCCAGGCATCAAATA
Fcm2 Dir	TTTTGGGATCCATTGTCGGCGTCAGTGTCTG
Fpmv His Rev	TTTTTAAGCTTCAATGATGATGATGATGATGTGGCGGTTGCGCCCGTA
pJV53dir	GTCAGTCACCAACCCTCCAC
pJV53rev	GAATCCTGCTTGGTGACAGC
ctpF_interno_dir	CTATGCACCCGACGTCCT
cpF_interno_rev	GAACCTGGTATCACGTTTTCG
Comp_Up-ctpF	TCGTCGAACACTCGTACCTG
Comp_Down_ctpF	CGTCCGCAACCTAGTTGAAT
primerpYUB854	GTGGCTCCCTCACTTTCTGG
Hyg_dir_out	ACTTCGAGGTGTTCGAGGAG
Adir2013	TTTTCTCGAGCGGATGGCAAGACC
Arev2013	TTTTGCTAGCGCGCGTTACCACC
Bdir2013	TTTTTCTAGATATCGGGGTGTGGGTGC
Brev2013	TTTTTCATGATACCACCAGCACGATCCAG

2.2. Bioinformatic analysis

The amino acid sequence of CtpF was pairwise aligned with 12 P₂-type Ca²⁺-ATPase sequences retrieved from UniProt and the alignments were visualized using Jalview 2.10.5 software [41].

2.3. Mtb ctpF gene cloning and expression in M. smegmatis

The ctpF gene (Rv1997) was amplified by PCR from the template genomic DNA of Mtb H37Ra using the primer pair Fcm2Dir/FpmvHis-Rev (Table 2) and the Phusion DNA Polymerase (Thermo Scientific). The reverse primer added a His₆-tag to the C-term of the protein. The amplimer, flanked by the BamHI and HindIII restriction sites, was cloned into the shuttle vector pMV261 to obtain the pLNA29 plasmid, whose integrity was confirmed by PCR, restriction mapping, and DNA sequencing. M. smegmatis mc²155 cells were transformed with the pLNA29 plasmid and the transformant colonies were verified by colony PCR. Recombinant colonies were grown in LB-Kan at 37 °C with agitation at 180 rpm until an OD₆₀₀ = 0.5–0.6. Protein expression was induced by heat shock at 45 °C for 1 h [42]. Protein extracts were analyzed in 10% polyacrylamide gels (SDS-PAGE) and immunostaining dot-blots using rabbit anti–His polyclonal primary antibody and goat anti–rabbit secondary antibody HRP-conjugate (Thermo Scientific, USA).

2.4. Construction of the ctpF-defective Mtb strain

The deletion of the ctpF gene in Mtb was performed using the Che9c recombineering system [43]. Briefly, the allelic exchange substrate (AES) was generated by separately cloning 500 bp of the upstream and downstream sequences of the ctpF gene into the pYUB854 vector flanking a hygromycin-resistance cassette. The upstream (section A) and downstream (section B) regions of the ctpF gene were amplified by PCR and separately cloned into the pGEM®-T Easy vector (Promega). The A and B sections were subcloned into pYUB854 to generate the pLNA22 plasmid, from which the AES was released by restriction enzymes digestion. Simultaneously, the recombineering strain (Mtb transformed with the pJV53 plasmid) was cultured in 7H9 supplemented with 0.2 % succinate, 0.05 % Tween 80 and Kan, until an OD₆₀₀ = 0.5–0.8. Cells were then supplemented with 0.2 M glycine and induced with 0.2% acetamide. The Recombineering cells were electroporated with 100 ng AES and plated on 7H11-OADC-Kan-Hyg [43]. Mtb colonies defective in ctpF (MtbΔctpF) were screened by colony PCR (primers listed in Table 2). Genomic DNA was isolated from the selected Mtb mutant to confirm the targeted gene replacement by PCR, using primers matching within the Hyg^R cassette and flanking the sections A and/or B of the AES (primers listed in Table 2). Finally, the PCR products were sequenced to confirm the integrity of the targeted gene replacement.

2.5. Preparation of plasma membrane vesicles

Cultures of mycobacterial cells (5 L) were grown until OD₆₀₀ = 0.5–0.8. The entire procedure was performed at 4 °C. Cells were harvested by centrifugation and washed twice with buffer A (10 mM MOPS, 0.08 g/mL sucrose, pH = 7.4), resuspended in lysis buffer (10 mM MOPS, 1 mM EDTA, 0.3 mM PMFS, pH = 7.4), and mechanically lysed with a Mini Bead Beater (Biospec) by eight 1-minute cycles. Cellular debris were removed by centrifugation at 25,000 x g for 30 min in a Megafuge 16R Centrifuge (Thermo Scientific). Supernatants (membrane and cytoplasmic fractions) were isolated by centrifugation at 100,000 x g for 90 min in a Sorvall WX Floor Ultra Centrifuge (Thermo Scientific). The remaining supernatant was discarded and the pellet containing the membrane fraction was resuspended in buffer A [44]. The protein concentration was assessed with the Bradford–Zor–Selinger [45] or the BCA methods using the Pierce BCA Protein Assay Kit (Thermo scientific). The protein extracts were finally adjusted to 1 mg/mL, aliquoted, and stored at -20 °C until use.

2.6. Metal accumulation assays

Cultures of mycobacterial cells were grown until OD₆₀₀ = 0.5–0.8. Then, the cells were harvested and washed three times with washing buffer (10 mM MOPS, 140 mM choline chloride, 0.5 mM DTT, 250 mM sucrose). The cell pellet was resuspended in 5 mL in washing buffer and separately supplemented with cations using the following final concentration of salts: 10 mM CaCl₂, 240 mM NaCl or 240 mM NaCl/40 mM KCl, followed by incubation at 37 °C for 1 h with agitation. Subsequently, cells were harvested and washed twice with washing buffer. Pellets were dried at 37 °C to constant weight and mineralized with 500 μL of HNO₃ (trace metal grade) for 1 h at 80 °C, followed by overnight incubation at 20 °C. Sample digestions were stopped by adding 30% H₂O₂ and were completed with 2% high purity HNO₃ in H₂O (18 MOhm*cm). The metal content of the samples was measured by flame absorption spectroscopy using a 300 Atomic Absorption Spectrometer (Perkin Elmer, USA). The calcium accumulation was normalized against the amount of calcium measured in the WT cells (dry mass) [9, 12, 13].

To assess the calcium accumulation inside membrane vesicles, 10 μg of protein (membrane vesicles) were mixed with reaction buffer (25 mM MOPS, 250 mM sucrose, 3 mM MgSO₄, 150 mM KCl, 0.05 % Brij-58, 3 mM Na₂ATP) and incubated at 37 °C for 1 min. Reactions were initiated by adding 100 μM Ca²⁺, incubated at 37 °C for 30 min, stopped by filtration through 0.22 μM pore size filters (Millipore, USA), and the filters were washed two times with washing buffer (25 mM MOPS, 250 mM sucrose, 3 mM MgSO₄, 150 mM KCl, 0.05 % Brij-58, 1 mM EGTA). The filters were then mineralized as above. The metal content of the samples was measured by flame absorption spectroscopy using a contrAA 700 Jena Analytik Atomic Absorption Spectrometer.

2.7. ATPase activity assays

The ATPase activity of the plasma membrane vesicles was measured according to the Fiske-Subbarow method with modifications, as previously described [19, 20, 46]. Enzymatic reactions (final volume 50 μL) were performed in 96-well plates using 10 μg of protein in reaction buffer (3 mM MgCl₂, 10 mM MOPS, pH = 7.4) and supplemented with 0.02 % Brij-58, 0.29 mM Ca²⁺ and 0.25 mM EGTA (final concentrations) to control the amount of free calcium. Maxchelator software was used to calculate the free Ca²⁺ [47]. The enzymatic reactions were initiated by adding 3 mM Na₂ATP and incubating at 37 °C for 30 min. The reactions were stopped by adding 100 μL of stopping solution (3 % ascorbic acid, 0.5 % ammonium molybdate, 3 % SDS and 2 M HCl). Subsequently, the samples were kept 10 min at 4 °C and then 150 μL of stabilizing solution (3.5 % bismuth citrate, 3.5 % sodium citrate, 2 M HCl) were added. Afterward, the samples were incubated at 37 °C for 10 min. Finally, the Pi released was quantified by measuring the OD₆₉₀. The ATPase activity is reported as Pi nmol produced by mg protein by min of reaction (nmol Pi. mg⁻¹. min⁻¹) from three independent experiments [19, 20, 46].

2.8. Mycobacterial tolerance to metal cations

The culture was harvested and washed three times with Sauton's medium (pH = 7.4) supplemented with 0.05 % Tween 80 and 0.2 % glucose. The cell pellet was resuspended in Sauton's medium and diluted until OD₆₀₀=0.05–0.06. Subsequently, 100 μL of bacterial suspension were separately mixed in 96-well plates with 100 μL of serial dilutions of cations in a range of previously determined concentrations: Ca²⁺ (0.4 mM - 9.5mM), Na⁺ (10 mM–150 mM) and K⁺ (5 mM–150 mM).

Cultures were incubated at 37 °C for 21 days at 80 rpm and the final OD₆₀₀ of cultures was measured in an iMARKTM Microplate Reader (Bio-Rad, CA, USA). Cells grown in the same medium without cation or supplemented only with 10 μg/mL isoniazid were considered controls for 100 % and 0 % growth, respectively [19, 20]. The IC₅₀ was determined using GraphPad Prism 8.0 software using a nonlinear regression with log (inhibitor) vs. normalized response -Variable slope. Each experiment was assessed in triplicate from three biological replicates.

2.9. Oxidative and nitrosative stress assays

Cultures of mycobacterial cells were harvested and washed three times with 7H9 supplemented with oleic acid-albumin-dextrose (OAD) (50 μg/mL oleic acid, 0.5 % Bovine albumin Fraction V and 0.2 % dextrose) and 0.05 % Tween 80. The cell pellet was resuspended in 7H9-OAD and diluted until an OD₆₀₀ = 0.05–0.06. Subsequently, 100 μL of bacterial suspension were separately mixed in 96-well plates with 100 μL of serial dilutions of redox agents in a range of previously determined concentrations: H₂O₂ (0.5 mM–25 mM) and sodium nitroprusside (SNP) (0.01 mM–1 mM) [29, 34, 48]. Cultures were incubated at 37 °C for 8 days at 80 rpm and the final OD₆₀₀ of cultures were measured in an iMARKTM Microplate Reader (Bio-Rad, CA, USA). Cells grown in the same medium without an oxidant agent or supplemented only with 10 μg/mL isoniazid were considered controls for 100 % and 0 % growth, respectively. The IC₅₀ was determined in GraphPad Prism 8.0 software using nonlinear regression with log (inhibitor) vs. normalized response -Variable slope. The results are representative of three independent experiments.

3. Results

3.1. Mtb ctpF encodes a putative Ca²⁺ P-type ATPase

Among the 12 ORFs that encode P-type ATPases in the Mtb genome, four of them have been classified as P₂-type ATPases: CtpE, CtpF, CtpI, and CtpH [16]. This subclass of membrane transporters includes calcium transport-associated enzymes such as SERCA, PMCA1, SPCA and LMCA1, together with Na⁺/K⁺-ATPases [49]. SERCA is the best structurally characterized P₂-type ATPase, and multiple conformations comprising its entire catalytic cycle have been resolved and reported in the Protein Data Bank. In addition, SERCA is highly conserved in higher eukaryotes and displays high homology with some prokaryotic organisms. In contrast, no X-ray structures of bacterial Ca²⁺ P-type ATPases have been resolved. There are only biochemical characterizations of some bacterial Ca²⁺ P-type ATPases from Listeria monocytogenes and Streptococcus pneumoniae [22, 50].

Members of the P₂-type ATPases group share the characteristic PEGL-motif and the conserved amino acids of the ion-binding pocket [51]. As shown in Fig. 1, CtpF is closely related to SERCA by sharing 8 of the 10 residues involved in calcium binding in SERCA [52]. The calcium coordination sites in SERCA are named Site I (Asn768, Glu771, Thr799, Asp800, Glu908) and Site II (Val304, Ala305, Ile307, Glu309, Asn796, Asp800) [52]. We observed that CtpF displays all of these residues, except for Ala305 that is substituted by a Gly residue and Asp800 that is substituted by an Ala residue (Fig. 1). Thus, even when the primary structures of CtpF (905 aa; 95 kDa) and SERCA (994 aa; 110 kDa) display just a 33% overall identity, they share an 80% identity with respect to the residues involved in Ca²⁺ coordination.

Fig. 1.

Mtb CtpF possesses the characteristic Ca²⁺ binding residues of P-type ATPases. Multiple sequence alignment of 12 P₂-type ATPases in MEGA X using Muscle. All sequences were retrieved from UniProt. Jalview 2.10.5 was used to visualize the final result. The arrows show the 10 key calcium-binding residues of SERCA, 8 of which are conserved in CtpF. Blocks in top represent the transmembrane segments (TM) were the cation binding amino acids are located.

3.2. CtpF prevents calcium accumulation in mycobacterial cells

In order to determine if CtpF is involved in calcium transport across the mycobacterial cell membrane, we evaluated calcium accumulation in recombinant obtained from M. smegmatis cells expressing CtpF, Mtb WT and MtbΔctpF cells. The latest were obtained by homologous recombination using the Che9c system (Fig. 2A) [43]. The allelic exchange replacement of the ctpF locus in the MtbΔctpF mutant cells was confirmed by PCR (Fig. 2B and C). Nucleotide sequencing of the amplimers showed that the Hyg cassette indeed inserted into the ctpF gene, by showing the AES insertion into the desired site of the Mtb genome and the presence of the γδ resolvase sites, allowing the further removal of the Hygromycin resistance cassette to eliminate the antibiotic resistance gene [43].

Fig. 2.

Targeted allelic exchange of the Mtb ctpF locus. A) Schematic representation of the homologous recombination process to generate a ctpF knockout mutant of Mtb. B) Thirteen colonies were examined by PCR using primers a and b (508 pb for the WT locus and no product for the mutant). The recombinant strain showing the absence of the ctpF gene was selected and named MtbΔctpF. DNA from Mtb WT was used as control. C) A PCR analysis was performed using different combinations of primers c and d or e and f to confirm the homologous recombination in the target locus; no PCR product is expected from WT cells. Primers d and e are located within the Hyg^R cassette and only match in the chromosome of mutant strain.

When exposing the M. smegmatis cells expressing CtpF to a high calcium concentration in the extracellular medium (10 mM) during 1 h, the recombinant cells accumulate 2-fold less Ca²⁺ than the control cells (WT transformed with empty pMV261). This suggests that CtpF may be involved in the active Ca²⁺ transport from mycobacterial cells against the concentration gradient (Figs. 3A and 3B). To confirm that it was an ion-specific transport, this experiment was also performed using 240 mM sodium and 240 mM/40 mM sodium/potassium treatment. No significant differences were observed between recombinant and control cells under those conditions (Fig. 3C). To further confirm that CtpF is associated with calcium efflux from mycobacterial cells, we determined the calcium accumulation in the MtbΔctpF strain. In agreement, MtbΔctpF cells accumulated 4-fold more Ca²⁺ than the Mtb WT cells (Fig. 3D). These results are consistent with the reduced calcium accumulation observed for the recombinant strain and supports the hypothesis that CtpF is a calcium efflux pump (Fig. 3A).

Fig. 3.

Calcium accumulation in mycobacterial cells. A) Hypothesized function of CtpF in the mycobacterial plasma membrane. In the presence of a high extracellular calcium concentration, CtpF transports calcium against its concentration gradient. B) Calcium accumulation inside M. smegmatis WT and the recombinant cells expressing CtpF. The amount of calcium accumulated was internally normalized against the amount of calcium measured in the dry mass of pellet from the WT strain. C) Sodium accumulation in the absence and presence of potassium in the WT and recombinant strains normalized with respect to the metal accumulation in the WT strain. D) Calcium accumulation in Mtb cells. The amount of calcium accumulated was internally normalized against the amount of calcium measured in the dry mass of pellet from the WT strain. The data shown are representative of three independent experiments. Unpaired two-tailed t test, ****P < 0.0001.

3.3. CtpF promotes calcium accumulation in right-side-out mycobacterial plasma membrane vesicles

To further confirm that CtpF transports calcium from inside mycobacterial cells against the concentration gradient, we measured calcium accumulation in an inverted membrane vesicle model supplemented with Brij 58 to obtain an "inside-out" configuration (Fig. 4A) [44]. In this experimental model, the cytoplasmic domains of CtpF are exposed to the reaction medium, the direction of transport is inverted, and calcium accumulation depends on the availability of ATP and Ca²⁺in the reaction milieu. As shown in Fig. 4B, vesicles obtained from recombinant M. smegmatis cells accumulated more Ca²⁺ than vesicles obtained from WT cells, confirming that CtpF transports the metal from the cytoplasm.

Fig. 4.

Calcium accumulation in membrane vesicles obtained from M smegmatis cells. A) Model of inverted membrane vesicles where the cytoplasmic domains of CtpF are exposed to the reaction medium. B) Calcium accumulation in WT and recombinant membrane vesicles from cells expressing ctpF. All normalized against the calcium accumulation in membrane vesicles of the WT strain. Data corresponds to mean ± SEM from three independent replicates. Unpaired two-tailed t test, **P < 0.01.

3.4. Ca²⁺ ions stimulate the CtpF ATPase activity

We assessed the Ca²⁺-dependent ATPase activity on plasma membrane vesicles obtained from mycobacterial cells. The enzyme reactions were supplemented with ATP as energy source, Mg²⁺ as cofactor, and Brij-58. Then, the ATPase activity was measured by quantifying the Pi released from ATP hydrolysis [44, 46]. As expected, the Ca²⁺-ATPase activity was higher in vesicles from M. smegmatis cells expressing CtpF (2.5-fold) than in vesicles obtained from M. smegmatis WT cells (Fig. 5A). In agreement, Ca²⁺ ATPase activity increased only 7 % in the MtbΔctpF plasma membrane vesicles, when the increment was 30 % in vesicles obtained from the Mtb WT cells, compared to the basal ATPase activity (Fig. 5B). This decreased Ca²⁺-dependent ATPase activity in vesicles from MtbΔctpF cells is in agreement with the proposed role of CtpF, as Ca²⁺-efflux pump. Importantly, since the ATPase activity was measured on crude membrane extracts, part of the Ca²⁺-stimulated ATPase activity in vesicles from MtbΔctpF cells should be attributed to other Ca²⁺ATPases present in the mycobacterial plasma membrane. Corroborating CtpF is a Ca²⁺ P-type ATPase, its activity was susceptible to vanadate (Fig. 5B), a known inhibitor of this kind of transporters [44, 46, 53].

Fig. 5.

Ca²⁺ P-type ATPase mediated by CtpF. A) Ca²⁺-dependent ATPase activity of membrane vesicles of recombinant M. smegmatis normalized against the control strain (M. smegmatis WT). B) Ca²⁺-ATPase activity of membrane vesicles of Mtb WT and MtbΔctpF against the basal ATPase activity. The dotted lines represent the percentage of the ATPase activity stimulated by Ca²⁺. Values correspond to the mean ± SEM from three independent replicates. Unpaired two-tailed t test, ***P < 0.001.

3.5. CtpF confers Ca²⁺ tolerance to mycobacterial cells

Since Mtb activates P-type ATPases to face high concentrations of metals inside macrophages [7, 8, 9], it is possible that CtpF may contribute to restoring intracellular calcium levels in Mtb. Therefore, we assessed the tolerance of Mtb cells to different concentrations of Ca²⁺, K⁺, and Na⁺ (Fig. 6). Assuming that CtpF is a potential calcium transporter, MtbΔctpF should be sensitive to high concentrations of calcium. As observed in Fig. 7, MtbΔctpF cells were more susceptible to Ca²⁺ than WT. Specifically, the IC₅₀ value of calcium in MtbΔctpF cells (1.7 mM) was significantly lower than in WT cells (2.5 mM). Even though the tolerance of MtbΔctpF to Na⁺ was significantly lower than WT, no difference in the tolerance to of K⁺ was observed (Fig. 7). These results reinforce the idea that CtpF contributes to maintaining the physiological level of calcium in the mycobacterial cytosol.

Fig. 6.

Growth of Mtb cells in presence of cations and oxidizing agents. Mtb strains were grown in Sauton's media supplemented with varying concentrations of A) CaCl₂. B) NaCl. C) KCl. 7H9-OAD media supplemented with varying concentrations of D) H₂O₂. E) SNP. The OD₆₀₀ of mycobacteria growing in absence of cations or oxidizing agents was considered as positive control (100% of cell growth).

Fig. 7.

Tolerance of Mtb cells to metal cations. Mycobacterial cells were grown in Sauton media supplemented with varying concentrations of CaCl₂, KCl, and NaCl. OD₆₀₀ of cultures supplemented without cations were considered as 100% of cell growth. Values represent the IC₅₀. Data are mean ± SEM from three independent experiments. Unpaired two-tailed t test, *P < 0.05, **P < 0.01.

3.6. CtpF is associated with the oxidative stress response in mycobacterial cells

Being part of the DosRS regulon [24, 25, 26, 27, 28, 29, 30, 31], we suspected a possible association between the transcriptional response of ctpF and oxidative/nitrosative stress. Testing this, we evaluated the sensitivity of MtbΔctpF and Mtb WT to ROS/RNS stress conditions (Figs. 6 and 8). As observed in Fig. 8, MtbΔctpF cells were more sensitive to oxidative and nitrosative stresses compared to Mtb WT. With IC₅₀ values of 1.6 mM H₂O₂ and 85 μM SNP for MtbΔctpF cells and 2.9 mM H₂O₂ and 214 μM SNP for Mtb WT, confirming that MtbΔctpF cells displayed hyper-susceptibility to ROS/RNS stresses compared to the WT strain.

Fig. 8.

Response of Mtb WT and MtbΔctpF to oxidative and nitrosative stresses. Bacteria were grown in 7H9-OAD media supplemented with varying concentrations of H₂O₂ and SNP. OD₆₀₀ of cultures in absence of oxidant agent are considered the 100%. Values represent the IC₅₀. Data are mean ± SEM from three independent experiments. Unpaired two-tailed t test, ****P < 0.0001.

4. Discussion

Calcium is a pivotal messenger for different physiological processes and signaling cascades in bacteria [50, 54]. Calcium is directly involved in membrane transport mechanisms, chemotaxis, cell division, and differentiation processes [54, 55]. The intracellular calcium content in bacteria increases when cells are surrounded by natural environments containing high doses of this metal (millimolar calcium concentrations). However, this increase in calcium levels should be transient in order to maintain bacterial viability [50, 54]. Therefore, a calcium homeostatic system is essential for bacterial survival [50, 54]. In this sense, Ca²⁺ P-type ATPases may play a relevant essential role in bacterial integrity.

CtpF is the most closely related mycobacterial transporter to the well-studied calcium eukaryotic transporter SERCA. Accordingly, our bioinformatics predictions showed that both proteins share 8 of 10 amino acids from the calcium coordination sites. Since other bacterial calcium ATPases also share key calcium-binding residues with SERCA, such as LMCA1 from L. monocytogenes [22], we hypothesized that CtpF is a Ca²⁺-transporting P-type ATPase. Testing this, calcium accumulation experiments were conducted on whole mycobacterial cells and plasma membrane vesicles. M. smegmatis was used as an expression host for CtpF since this environmental species is an easier-to-handle model with similar envelope and membrane properties to Mtb [56]. The calcium accumulation experiments indicated that M. smegmatis plasma membrane with MtbCtpF embedded responded to increased extracellular calcium concentration. Even though, this could be attributable not only to P-type ATPases but also to the activity of different metal transport systems [21], we consider that this reduced calcium accumulation in M. smegmatis recombinant cells to the presence of CtpF in the membrane, which is the relevant difference between the recombinant and control cells. To further discard that CtpF is a sodium/potassium P₂-type ATPase [46], we tested sodium accumulation. The results showed that there is no significant difference in the accumulation of sodium between recombinant and control cells. These results suggest a calcium efflux activity mediated by CtpF.

Further evidence for CtpF-mediated calcium transport to the extracellular medium was provided by experiments using a membrane vesicle model, in which, hypothetically, the direction of calcium transport by CtpF is reversed [44]. This approach was complemented by measuring ATPase activity. As expected, calcium accumulation and Ca²⁺-dependent ATPase activity were higher in the membrane vesicles from the recombinant cells compared to control cells. This validates the hypothesis that the direction of transport was inverted in plasma membrane vesicles. The mutant phenotype of the MtbΔctpF corroborates that CtpF is a calcium efflux pump. This is: 1) MtbΔctpF cells have an impared capacity to restore cytoplasmic calcium levels upon extracellular Ca²⁺ exposure; 2) Vesicles isolated from MtbΔctpF exhibited a reduced Ca²⁺-dependent ATPase activity and 3) MtbΔctpF cells are more susceptible to Ca²⁺ than WT cells. Therefore, CtpF plays an important role in calcium homeostasis by preventing a toxic metal overload and favoring mycobacterial survival under infection conditions.

Undoubtedly, Mtb requires an efflux system to maintain calcium homeostasis when the tubercle bacillus is surviving in highly enriched calcium environments, such as lungs and mucous membranes [50]. The deletion of P-type ATPases in mycobacteria is known to cause unbalanced cation transport across the plasma membrane and impaired capacity to respond to toxic substances [8, 9, 12, 13, 14]. It is likely that different concerted cellular events contribute to maintaining the optimal levels of intracellular calcium, including: 1) calcium influx; 2) an increase in the intracellular calcium concentration; 3) calcium binding of target proteins, and 4) restoration of the calcium concentration by inducing a metal efflux mechanisms, in which CtpF could be relevant.

There are reports that intracellular calcium accumulation in S. pneumoniae activates molecular systems in response to oxidative stress; indeed, S. pneumonia defective in Ca²⁺-ATPase is more sensitive to oxidative stress compared with the WT strain [50]. In agreement, our data showing the hypersensitivity to oxidizing agents of the MtbΔctpF strain suggest a correlation between calcium pumping and oxidative stress response in mycobacteria. This is in agreement with ctpF expression being regulated by the dormancy regulator DosR in response to low levels of O₂ and nitric oxide exposure [26, 27, 29, 34, 38]. We speculate that CtpF is activated in response redox stress, a characteristic stress condition faced by pathogens during infection. This implies a critical function of CtpF in Mtb; however, experiments evaluating the importance of CtpF during infection are necessary to further confirm this hypothesis.

5. Conclusions

CtpF resembles SERCA, the well-studied P₂-type Ca²⁺-ATPase in higher eukaryotes. Our data show that CtpF is in fact a bacterial Ca²⁺-ATPase. The expression of CtpF in M. smegmatis allows recombinant cells to withstand a transient increase in calcium concentration. Accordingly, MtbΔctpF cells exhibit higher calcium accumulation and increased susceptibility to the cation. The increased susceptibility of the MtbΔctpF strain to oxidative stress, suggest a link of the efflux pump function of CtpF with more complex cellular processes, as the response to toxic substances encountered during infection, or bacterial signaling in other to respond to the arsenal of the host cell. In conclusion, our experiments indicate that CtpF transports calcium from mycobacterial cells to the extracellular environment against the concentration gradient.

Declarations

Author contribution statement

Milena Maya-Hoyos, Cristian Rosales: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Lorena Novoa-Aponte: Conceived and designed the experiments; Wrote the paper.

Eliana Castillo: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Carlos Yesid Soto: Conceived and designed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This work was supported by the Colciencias Grant 110171250419 and the División de Investigación Bogotá (DIB)-Universidad Nacional de Colombia, grant 41646. MMH and CR are fellows of Colciencias-Colfuturo Doctorado Nacional 647 (2015–2019) and the Vicerrectoría de Investigación/Universidad Nacional de Colombia, respectively. LNA was fellow of Colciencias-Colfuturo Doctorado Nacional 567.

Competing interest statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.