Advertisement

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Cholesterol metabolism and intrabacterial potassium homeostasis are intrinsically related in Mycobacterium tuberculosis

  • Yue Chen,

    Roles Conceptualization, Funding acquisition, Investigation, Writing – original draft, Writing – review & editing

    Affiliation Department of Molecular Biology and Microbiology, Tufts University School of Medicine, Boston, Massachusetts, United States of America

    ⨯
  • Berge Hagopian,

    Roles Investigation, Writing – review & editing

    Affiliation Department of Molecular Biology and Microbiology, Tufts University School of Medicine, Boston, Massachusetts, United States of America

    ⨯
  • Shumin Tan

    Roles Conceptualization, Funding acquisition, Investigation, Supervision, Writing – original draft, Writing – review & editing

    shumin.tan@tufts.edu

    Affiliation Department of Molecular Biology and Microbiology, Tufts University School of Medicine, Boston, Massachusetts, United States of America

    ⨯

Abstract

Potassium (K+) is the most abundant intracellular cation, but much remains unknown regarding how K+ homeostasis is integrated with other key bacterial biology aspects. Here, we show that K+ homeostasis disruption (CeoBC K+ uptake system deletion) impedes Mycobacterium tuberculosis (Mtb) response to, and growth in, cholesterol, a critical carbon source during infection, with K+ augmenting activity of the Mtb ATPase MceG that is vital for bacterial cholesterol import. Reciprocally, cholesterol directly binds to CeoB, modulating its function, with a residue critical for this interaction identified. Finally, cholesterol binding-deficient CeoB mutant Mtb are attenuated for growth in lipid-rich foamy macrophages and in vivo colonization. Our findings raise the concept of a role for cholesterol as a key co-factor, beyond its role as a carbon source, and illuminate how changes in intrabacterial K+ levels can act as part of the metabolic adaptation critical for bacterial survival and growth in the host.

Author summary

Ions are fundamental to both bacterial and mammalian biology, with potassium (K+) the most abundant positively-charged ion present within cells. However, how the maintenance of K+ balance is related to other important aspects of bacterial biology remains poorly understood. Mycobacterium tuberculosis (Mtb) is the causative agent of tuberculosis, which remains the largest cause of mortality due to an infectious disease. Here, we show that disrupting Mtb K+ balance via deletion of the CeoBC K+ uptake system impedes its ability to grow in cholesterol, a critical lipid nutrient source during infection, with K+ found to play a role in bacterial cholesterol import. Intriguingly, cholesterol directly bound to the CeoB protein that is part of the CeoBC K+ uptake system. Preventing the binding of cholesterol to CeoB affected K+ uptake activity in the presence of cholesterol and resulted in attenuation of the ability of Mtb to grow and survive in lipid-rich host conditions. Our work uncovers a role for cholesterol beyond nutrition, with cholesterol serving as a key co-factor for function of a Mtb K+ uptake system, and reveals how changes in the balance of K+ within the bacterial cell are incorporated in the adaptation process of Mtb.

Citation: Chen Y, Hagopian B, Tan S (2025) Cholesterol metabolism and intrabacterial potassium homeostasis are intrinsically related in Mycobacterium tuberculosis. PLoS Pathog 21(5): e1013207. https://doi.org/10.1371/journal.ppat.1013207

Editor: Francis Alonzo, University of Illinois at Chicago College of Medicine, UNITED STATES OF AMERICA

Received: January 23, 2025; Accepted: May 12, 2025; Published: May 22, 2025

Copyright: © 2025 Chen et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files.

Funding: This work was supported by grants R01 AI143768 and R21 AI171356 from the National Institutes of Health (https://www.niaid.nih.gov/) to ST, and by an American Lung Association (https://www.lung.org/research) Catalyst Award (CA-1268681) to YC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Ions are fundamental to cellular physiology [1–6], and in the context of host-pathogen interactions, extensive research has focused on metal ions such as iron, zinc, and manganese, due to their scarcity and the competition between host and pathogen for their acquisition [3,7,8]. However, abundant ions, such as potassium (K+) and chloride (Cl−), also play critical roles in host-pathogen interactions, both in driving bacterial transcriptional responses and adaptation, and in their roles in cellular homeostasis [6,9–15]. Of pertinence here, K+ is the most abundant intracellular cation in both host and bacterial cells, and its levels must be carefully regulated for proper cellular function [15,16]. In the host, K+ plays myriad functions, including as a signal in the induction of immune responses, with for example low K+ concentrations ([K+]) triggering activation of the NLRP3 inflammasome [17,18]. In bacteria, beyond its most often studied role in osmoprotection [15,19], disruption of K+ homeostasis has also been reported to affect aspects ranging from Salmonella effector protein secretion [20], to Streptococcus mutans acid stress adaptation [21]. These studies highlight the critical role of K+ in bacterial biology and pathogenicity, but much remains unknown regarding how intrabacterial [K+] may be regulated in response to other environmental cues, and the underlying mechanisms that account for the impact of K+ homeostasis disruption on bacterial biology phenotypes.

Mycobacterium tuberculosis (Mtb), the causative agent of tuberculosis, is a bacterial pathogen highly adapted for colonization of the human host, and remains the leading cause of death from an infectious disease worldwide [22]. The ability of Mtb to respond to ionic signals and maintain intrabacterial ionic homeostasis is critical for its survival within the host. For example, Mtb is adept at maintaining intrabacterial pH near neutrality even in the presence of acidic environmental pH levels, with disruption of this ability resulting in attenuated host colonization [23]. In the case of K+, we have previously shown that disruption of Mtb K+ homeostasis by deletion of ceoBC, encoding the constitutive, low-medium affinity Trk K+ uptake system [24], significantly impaired Mtb response to acidic pH and high [Cl−] in its local environment, without affecting intrabacterial pH or membrane potential [13]. ∆ceoBC Mtb is consequently attenuated for host colonization in both macrophage and murine infection models [13], underscoring the importance of K+ homeostasis in the biology of Mtb-host interactions.

Crucially, host colonization by bacterial pathogens entails not just adaptation to changing ionic signals, but also integration of these responses to the availability of different nutrient sources during infection. Intriguingly, we recently discovered that a reduction in environmental [K+] dampened the transcriptional response of Mtb to cholesterol, while the presence of cholesterol conversely increased induction of K+ regulon genes [25]. Lipids, including cholesterol, are a vital carbon source for Mtb during infection, and deletion of Mtb genes needed for cholesterol utilization results in significantly attenuated host colonization [26,27]. How Mtb K+ homeostasis might impact bacterial cholesterol metabolism and vice versa remain open questions.

Here, we interrogate this interplay between Mtb K+ homeostasis and cholesterol uptake and metabolism. Our work reveals that disruption of K+ homeostasis via deletion of the CeoBC Trk K+ uptake system impedes Mtb response to, and growth in, cholesterol. This impairment likely arises from decreased intrabacterial [K+] in ∆ceoBC Mtb diminishing the activity of the ATPase MceG, which is vital for Mtb import of cholesterol [28–30]. Reciprocally, we find that cholesterol directly binds to CeoB, modulating its function, and identify a residue critical for this interaction. The interplay between Mtb K+ homeostasis and cholesterol uptake and metabolism is vital for the bacterium’s virulence, as disruption of the cholesterol-binding ability of CeoB results in significant attenuation of Mtb growth in lipid-rich foamy macrophages, and in a murine infection model that recapitulates canonical necrotic granulomas observed during human disease. Our findings raise the concept of a role for cholesterol as a key co-factor, beyond its role as a carbon source, and illuminate how changes in Mtb intrabacterial K+ levels act as part of the metabolic adaptation critical for Mtb survival and growth in the host.

Results

Disruption of K+ homeostasis inhibits Mtb cholesterol response

To examine how K+ homeostasis affects Mtb cholesterol response, we tested cholesterol regulon gene expression levels upon exposure of ∆ceoBC Mtb to cholesterol. Intriguingly, deletion of ceoBC resulted in reduced induction of cholesterol regulon genes [25,27,31,32] as compared to WT Mtb, which was restored upon complementation (ceoBC*) (Fig 1A). In contrast, disruption of the high affinity Kdp K+ uptake system, which is induced only in the presence of limiting K+, had no effect on the bacterium’s cholesterol response (S1A Fig). In accord with the dampening of the cholesterol transcriptional response, growth of ∆ceoBC Mtb was significantly reduced in cholesterol medium, but not in standard 7H9 rich medium (glucose and glycerol as carbon sources; Fig 1B). No difference in growth between WT and ∆ceoBC Mtb was similarly observed in control medium where the only difference to cholesterol medium was the use of glucose and glycerol as carbon sources (S2 Fig). We next tested if ∆ceoBC Mtb were altered in their ability to import cholesterol, utilizing assays with an intrinsically fluorescent cholesterol analog, dehydroergosterol (DHE). DHE fluorescence is limited in the aqueous phase but increases upon binding to cholesterol-binding proteins [33–35], and has been effectively used previously to demonstrate cholesterol binding to bacterial proteins [36,37]. As expected, in 7H9 medium, DHE signal was low and not different in WT, ∆ceoBC or ceoBC* Mtb (Fig 1C). In contrast, DHE signal was significantly higher in cholesterol medium, reflecting uptake of DHE into Mtb (Fig 1C). Notably, there was reduced DHE signal in ∆ceoBC Mtb as compared to WT and ceoBC* Mtb, indicating lower levels of DHE uptake into the mutant bacteria (Fig 1C). As a further control for this assay, we tested a Mtb mutant deleted for mceG, which encodes an ATPase important for import of fatty acids and cholesterol through the Mce1 and Mce4 transporters, respectively [28–30]. DHE signal in ∆mceG Mtb was 41% reduced compared to WT Mtb in cholesterol medium (S3 Fig), similar to the ~ 50% reduction in cholesterol uptake efficiency previously reported for this mutant [29], supporting that DHE signal differences observed reflect cholesterol uptake differences.

thumbnail
Download:
Fig 1. Disruption of K+ homeostasis inhibits Mtb cholesterol response.

(A) Mtb response to cholesterol is dampened in ∆ceoBC Mtb. Log-phase WT, ∆ceoBC, and ceoBC* (complemented mutant) Mtb were exposed to 7H9 or cholesterol media for 4 hours, before RNA extraction for qRT-PCR analysis. Fold change is as compared to the 7H9 condition, with sigA as the control gene. (B) ∆ceoBC Mtb is attenuated for growth in cholesterol medium. WT, ∆ceoBC, and ceoBC* Mtb were grown in 7H9 or cholesterol media, and OD600 monitored over time. (C) Cholesterol uptake is reduced in ∆ceoBC Mtb. Log-phase WT, ∆ceoBC, and ceoBC* Mtb were exposed to 7H9 or cholesterol media, supplemented with 100 µM dehydroergosterol (DHE), for 24 hours. DHE uptake into Mtb was measured via analysis of DHE fluorescence on a microplate reader, normalized against OD600. (D) Cholesterol increases intrabacterial K+ levels in a CeoBC-dependent manner. WT, ∆ceoBC, and ceoBC* Mtb, each carrying the P606’::GINKO2 reporter, were subcultured to OD600 = 0.3 into the indicated media (“chol” = cholesterol), and GINKO2 fluorescence measured by flow cytometry 6 days post-assay start. Reporter signal from 10,000 Mtb cells per sample per experimental run were obtained for determination of the mean GINKO2 fluorescence for each sample. (E) MceG exhibits ATPase activity. His-tagged MceG at indicated concentrations was tested for ATPase activity using an ADP-Glo kinase assay kit. His-tagged KstR1, a transcription factor with no ATPase activity, was also tested. ATP and ADP controls served as negative and positive controls, respectively. Luminescence (relative light units, RLU) was read on a microplate reader. (F) K+ increases MceG ATPase activity. 25 µM MceG was incubated with 200 mM KCl, 100 mM K2SO4 or 200 mM NaCl and tested for ATPase activity as in (E). All data are shown as means ± SEM from three independent experiments. Statistical analyses were performed using an unpaired t-test with Welch’s correction and Holm-Sidak multiple comparisons in (A) and (B). For (B), comparisons were of ∆ceoBC to WT in the cholesterol condition. A two-way ANOVA with Tukey’s multiple comparisons test was used in (C) and (D). An unpaired t-test with Welch’s correction was used in (E) and (F), with comparisons to the ATP control in (E) and to the no additive control within each group (MceG, ATP, or ADP) for (F). No significance was found for any comparisons in the ATP or ADP control sets in (F). N.S. not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. The numerical data underlying the graphs shown in this figure are provided in S1 Data.

https://doi.org/10.1371/journal.ppat.1013207.g001

To test if cholesterol reciprocally affects K+ homeostasis in Mtb, we adapted the genetically encoded GINKO2 K+ sensor, which consists of a circularly permuted enhanced GFP integrated with the Escherichia coli K+ binding protein Kbp [38], for expression in Mtb. Binding of K+ to GINKO2 triggers a conformational change, resulting in a K+ concentration-dependent increase in GFP fluorescence [38]. Utilizing Mtb strains constitutively expressing the GINKO2 reporter, we observed that WT and ceoBC* Mtb effectively maintained their intrabacterial K+ levels, even after exposure to K+-free medium for 6 days (Fig 1D). As expected, ∆ceoBC Mtb exhibited significantly reduced GINKO2 reporter signal after growth in K+-free conditions, demonstrating disrupted K+ homeostasis in the mutant strain (Fig 1D). Interestingly, cholesterol significantly increased intrabacterial [K+] in WT and ceoBC* Mtb, a phenotype that was lost in the ∆ceoBC mutant (Fig 1D). This was similarly observed in the K+-free cholesterol condition (Fig 1D). Disruption of the inducible Kdp high affinity K+ uptake system did not affect the increase in intrabacterial [K+] in cholesterol medium (S1B Fig), reinforcing that the relationship between cholesterol and K+ is specific to basal K+ homeostasis and the CeoBC K+ uptake system.

K+ can serve key roles in enzyme activation [39], and has been found to be important in activity of ATPases present in systems ranging from archaea to mammalian [39–41]. Markedly and as noted above, MceG is a Mtb ATPase critical for driving the import of fatty acids and cholesterol through the Mce1 and Mce4 transporters, respectively [28–30]. Given our results above, we thus hypothesized that increased intrabacterial [K+] levels stimulate the activity of MceG during Mtb growth in cholesterol medium, enabling the effective uptake/utilization of cholesterol. As expected, purified MceG exhibited ATPase activity, while a control protein, the transcription factor KstR1 that is involved in cholesterol regulon gene expression control [27,42], did not (Fig 1E). Intrabacterial [K+] has been reported to be in the range of hundreds of millimolar [43], and strikingly, we found that the presence of increased [K+] indeed resulted in higher MceG ATPase activity (Fig 1F). In contrast, addition of a different cation, sodium, had no effect on MceG ATPase activity (Fig 1F).

Collectively, these data demonstrate that disruption of K+ homeostasis impedes the uptake and response of Mtb to cholesterol, consequently decreasing the ability of the bacteria to grow in cholesterol medium. Mechanistically, our findings further indicate that the attenuation of Mtb growth on cholesterol medium upon disruption of intrabacterial K+ homeostasis results at least in part from the inability of the mutant Mtb to raise the intrabacterial [K+] setpoint in the presence of cholesterol, as K+ acts to boost the activity of the MceG ATPase that is crucial in function of the Mce4 cholesterol uptake system.

Cholesterol directly acts on the CeoBC K+ uptake system

Intriguingly, cholesterol has been shown to directly bind to and modulate the activity of the Kir family of K+ uptake systems in mammalian cells and the E. coli inwardly-rectifying K+ channel KirBac1.1 [44–47]. Additionally, studies have demonstrated that cholesterol can regulate the function of voltage-gated potassium channels (Kv) in various mammalian systems, including alveolar epithelial and lung cells [48–52]. We thus investigated whether cholesterol directly interacts with CeoBC by employing a thermostability shift assay, which has previously been successfully used to show Mtb protein binding to other factors, such as glycerol and magnesium [53,54]. As shown in Fig 2A, the presence of cholesterol resulted in an increase in CeoB thermostability, which was not observed in the presence of glycerol or glucose. In contrast, no thermostability shift was observed with CeoC in the presence of cholesterol (Fig 2B), demonstrating the specificity of the cholesterol interaction with CeoB. To further verify this interaction, we pursued a second independent approach using DHE. As previously described, DHE fluorescence increases upon binding to a cholesterol-binding protein [33–35]. In accord with the thermostability shift assay, incubation of CeoB with DHE resulted in significantly higher DHE fluorescence signal than with CeoC or the buffer only control (Fig 2C), supporting the specific interaction between cholesterol and CeoB.

thumbnail
Download:
Fig 2. Cholesterol directly acts on the CeoBC K+ uptake system.

(A and B) CeoB, but not CeoC, shows increased thermostability in the presence of cholesterol. Purified CeoB (A) or CeoC (B) were incubated with 5 µM cholesterol, glycerol, or glucose as noted, at room temperature for 20 minutes, before exposure to indicated temperatures for 5 minutes. Samples were then centrifuged and supernatant aliquots run on SDS-PAGE gels and analyzed by Western blot. Graphs show quantification of band intensity at 65ºC/50ºC for CeoB (A), or 95ºC/80ºC for CeoC (B). (C) CeoB, but not CeoC, binds to the fluorescent cholesterol analog dehydroergosterol (DHE). Purified CeoB or CeoC were incubated with 1 µM DHE for 30 minutes, and DHE fluorescence measured on a microplate reader. (D) Y100 and Y108 residues are important for increased CeoB thermostability in the presence of cholesterol. Purified CeoB, CeoB-Y100S, and CeoB-Y108S were tested for thermostability ± cholesterol as in (A). (E) CeoB Y100S and Y108S point mutants are unable to bind DHE. Purified CeoB, CeoB-Y100S and CeoB-Y108S proteins were tested for DHE binding as in (C). Data are shown as means ± SEM from 3 experiments for all graphs. p-values were obtained with unpaired t-tests for (A) and (B), a one-way ANOVA (Brown-Forsythe and Welch) with Dunnett’s T3 multiple comparisons test for (C) and (E), and a two-way ANOVA with Tukey’s multiple comparisons test for (D). N.S. not significant, ** p < 0.01. The numerical data underlying the graphs shown in this figure are provided in S1 Data.

https://doi.org/10.1371/journal.ppat.1013207.g002

Previous reports, primarily in the context of mammalian cells, have shown cholesterol interaction with proteins at a motif called the “cholesterol recognition amino acid consensus” (“CRAC”) motif (L/V-X1-5-Y-X1-5-K/R) or the inverted “CARC” version (K/R-X1-5-Y-X1-5-L/V) [55–57]. Intriguingly, the CeoB sequence contains one CRAC and one CARC motif (S4 Fig), which CeoC lacks. To determine whether mutations at these tyrosine (Y) sites disrupt cholesterol binding, we mutated the key Y residues (Y100 for the CRAC motif and Y108 for the CARC motif) to serine (S). Thermostability shift assays showed that both the CeoB Y100S and Y108S mutants lost the ability to bind cholesterol (Fig 2D). Similarly, DHE failed to bind effectively to the CeoB Y100S and Y108S mutant proteins, with DHE fluorescence unchanged from the buffer only control (Fig 2E).

Together, these data demonstrate that cholesterol binds directly to CeoB, part of the CeoBC Trk K+ uptake system that is critical for K+ homeostasis in Mtb. It further identifies CRAC and CARC motifs, and the Y100 and Y108 residues, as potential key sites of cholesterol interaction with CeoB.

Cholesterol binding to CeoB is critical for Mtb response and adaptation to cholesterol

Having established that CeoB is able to directly bind to cholesterol, we next sought to determine the biological consequences of this binding to CeoBC function and Mtb biology. We had previously demonstrated that deleting ceoBC leads to a significant reduction in Mtb response to acidic pH and high [Cl−] [13]. To test the impact of loss of CeoB cholesterol binding on this response, we introduced the pH/Cl-responsive reporter rv2390c’::GFP into ∆ceoBC complemented with ceoBC alleles where CeoB contained either the Y100S or Y108S point mutations (ceoB(Y100S)C* or ceoB(Y108S)C*, respectively). As shown in Fig 3A, the mutations at Y100 and Y108 did not affect rv2390c’::GFP reporter signal induction under high [Cl−] conditions. This result supports the conclusion that these mutations specifically disrupt the cholesterol binding of CeoB, and do not affect its function in contexts absent of cholesterol.

thumbnail
Download:
Fig 3. Cholesterol binding to CeoB is critical for Mtb response and adaptation to cholesterol.

(A) The ability of CeoB to bind to cholesterol does not affect Mtb response to Cl−. WT, ∆ceoBC, ceoBC*, ceoB(Y100S)C* and ceoB(Y108S)C* Mtb each carrying the Cl−-responsive rv2390c’::GFP reporter were grown in 7H9 medium ± 250 mM NaCl. Samples were taken at indicated time points and fixed for analysis of reporter expression by flow cytometry. (B) ceoB(Y108S)C* Mtb fails to exhibit increased intrabacterial [K+] in the presence of cholesterol. WT, ∆ceoBC, ceoBC*, ceoB(Y100S)C* and ceoB(Y108S)C* Mtb each carrying the P606’::GINKO2 reporter were subcultured at OD600 = 0.3 into the indicated media (“chol” = cholesterol), and GINKO2 fluorescence measured by flow cytometry 6 days post-assay start. Reporter signal from 10,000 Mtb cells per sample per experimental run were obtained for determination of the mean GINKO2 fluorescence for each sample. (C) Cholesterol uptake is reduced in ceoB(Y108S)C* Mtb. Indicated Mtb strains were exposed to 7H9 or cholesterol media, supplemented with 100 µM DHE, for 24 hours. DHE uptake into Mtb was measured via analysis of DHE fluorescence on a microplate reader, normalized against OD600. (D) ceoB(Y108S)C* Mtb is attenuated for growth in cholesterol. Indicated Mtb strains were cultured in 7H9 or cholesterol media and growth tracked by OD600 over time. (E) ceoB(Y108S)C* Mtb has a dampened response to cholesterol. Indicated Mtb strains were exposed to 7H9 or cholesterol media for 4 hours, before RNA extraction for qRT-PCR analysis. Fold change is as compared to the 7H9 condition, with sigA as the control gene. Data in all panels are shown as means ± SEM from 3 experiments. p-values were obtained with an unpaired t-test with Welch’s correction and Holm-Sidak multiple comparisons in (A), (D), and (E). For (A), comparisons were of the mutant/complement strains to WT in the 250 mM Cl− condition. For (D), comparisons were of the mutant/complement strains to WT in the cholesterol condition. For (E), comparisons made were for the mutant/complement strains to WT for each condition. A two-way ANOVA with Tukey’s multiple comparisons test was used in (B) and (C), with significant statistics for WT versus ∆ceoBC or ceoB(Y108S)C* shown. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. The numerical data underlying the graphs shown in this figure are provided in S1 Data.

https://doi.org/10.1371/journal.ppat.1013207.g003

In agreement with the continued functionality of the CeoB Y100S and Y108S proteins in non-cholesterol-related contexts, the point mutants also largely did not impair the ability of Mtb to maintain intrabacterial [K+] under K+-limiting conditions (Fig 3B). However, unlike WT and ceoBC* Mtb, the ceoB(Y108S)C* strain exhibited a phenotype similar to ∆ceoBC in the presence of cholesterol, failing to increase its intrabacterial [K+] (Fig 3B). Surprisingly, despite the purified protein assays indicating that CeoB Y100S is also unable to bind to cholesterol, ceoB(Y100S)C* Mtb showed similar increased intrabacterial [K+] as WT and ceoBC* in cholesterol-containing conditions (Fig 3B). This finding suggests that the ability of the CeoB Y100S protein to bind cholesterol might be rescued in the context of intact bacteria, where other protein partners, such as CeoC, are present and may affect overall structure of the CeoBC complex. Similarly, cholesterol uptake as indicated by DHE fluorescence showed that only the ceoB(Y108S)C* mutant phenocopied ∆ceoBC Mtb, with decreased DHE signal versus WT when the bacteria were grown in cholesterol medium (Fig 3C).

Examination of the effect of the CeoB point mutations on Mtb growth in, and response to, cholesterol medium reinforced the importance of the Y108 residue for proper cholesterol interaction with CeoB. In particular, the ceoB(Y108S)C* mutant exhibited a growth defect identical to ∆ceoBC Mtb in cholesterol medium (Fig 3D), with a corresponding reduction in its cholesterol transcriptional response (Fig 3E).

These data identify the Y108 residue as vital for the interaction of cholesterol with CeoB, and demonstrate the critical importance of CeoB binding to cholesterol for proper adaptation of Mtb to cholesterol.

Cholesterol binding to CeoB is important for Mtb host infection

Levels of cholesterol experienced by Mtb during host infection vary spatiotemporally, with lipid-rich foamy macrophages observed ringing the caseous necrotic lesions that form as infection progresses, but not at earlier time points [25,58]. We had previously shown that deletion of ceoBC resulted in attenuation for Mtb growth in murine bone marrow-derived macrophages (BMDMs) [13]. To test how the presence of lipids may further alter the ability of ∆ceoBC to colonize host macrophages, we infected untreated or oleate-treated BMDMs (to induce formation of foamy macrophages [25,59,60]) with WT, ∆ceoBC, and the various ceoBC complementation strains. Strikingly, we found that ∆ceoBC Mtb exhibited even greater attenuation in foamy macrophages compared to untreated BMDMs (Fig 4A). As expected, ceoB(Y108S)C* Mtb, but not ceoB(Y100S)C* Mtb, phenocopied ∆ceoBC Mtb in exhibiting reduced growth both in untreated and foamy BMDMs (Fig 4A).

thumbnail
Download:
Fig 4. The ability of CeoB to bind to cholesterol is important for host colonization.

(A) ΔceoBC and ceoB(Y108S)C* Mtb exhibit increased attenuation for growth in foamy versus untreated macrophages. Murine bone marrow-derived macrophages untreated or pre-treated with oleate for 24 hours to induce foamy macrophages were infected with the indicated Mtb strains and colony forming units (CFUs) tracked over time. Data are shown as means ± SD from 3 wells, representative of 3 independent experiments. p-values comparing each strain to WT in the untreated macrophages were obtained with a 2-way ANOVA with Tukey’s multiple comparisons test (*). Unpaired t-tests with Welch’s correction were further applied to compare ∆ceoBC and ceoB(Y108S)C* infections in foamy versus untreated macrophages (#). Only comparisons with significant p-values are indicated. *, # p < 0.05, **, ## p < 0.01, ### p < 0.001. (B) ceoB(Y108S)C* Mtb is attenuated for colonization in a murine infection model when foamy macrophages are present. C3HeB/FeJ mice were infected with the indicated Mtb strains, and lung homogenates plated for CFUs 2 or 6 weeks post-infection. p-values were obtained with a Mann-Whitney statistical test. N.S. not significant, * p < 0.05, *** p < 0.001, **** p < 0.0001. The numerical data underlying the graphs shown in this figure are provided in S1 Data.

https://doi.org/10.1371/journal.ppat.1013207.g004

Finally, to assess the role of cholesterol binding to CeoB in the context of Mtb infection of a whole animal host, C3HeB/FeJ mice were infected with WT, ∆ceoBC, ceoBC*, or ceoB(Y108S)C* Mtb and bacterial loads determined at 2 and 6 weeks post-infection. Hallmark necrotic lesions are formed in this mouse strain upon Mtb infection, with lipid-rich foamy macrophages present at 6 weeks, but not 2 weeks, post-infection [25]. At both time points examined, a significant reduction in bacterial load was observed for the ∆ceoBC mutant compared to WT (Fig 4B), consistent with our previous findings of attenuation in host colonization of ∆ceoBC Mtb in the C57BL/6J murine infection model [13]. In contrast, a defect in host colonization of ceoB(Y108S)C* Mtb was observed only at 6 weeks, but not 2 weeks, post-infection (Fig 4B). Together, these data demonstrate the importance of cholesterol binding to CeoB for Mtb growth and survival in lipid-rich environments during host infection.

Discussion

While K+ has been well-appreciated as the most abundant cation present in bacterial cells [15], how maintenance of K+ homeostasis may relate to bacterial metabolism adaptation, and how K+ homeostasis itself is regulated in response to changing nutrient conditions, has remained open questions. Our findings here reveal the critical interplay between K+ homeostasis and cholesterol metabolism in Mtb, and support two concepts that are likely to have broad pertinence across bacterial species, given the fundamental role of K+ in bacterial biology.

First, the observation that the intrabacterial [K+] setpoint is increased in the presence of cholesterol, with K+ acting to stimulate the activity of the ATPase MceG needed for cholesterol uptake and utilization by Mtb (Fig 5), supports the concept that K+ homeostasis is dynamic and integrated with environmental signals, with K+ serving a role in the regulation of key downstream pathways. Ions as critical cofactors for enzymes is well-appreciated in all kingdoms, with scarce divalent cations such as iron, manganese, and zinc being the most intensely studied [61,62]. In mammalian and plant biology however, the important role that K+ can play in enzyme activity has also been recognized [39,63]. This includes in kinases such as branched-chain ⍺-ketoacid dehydrogenase and pyruvate dehydrogenase kinase, where K+ acts to critically stabilize parts of the protein [64,65], and in ATPases such as Hsc70 and the plant plasma membrane H+-ATPase proton pump, which exhibit higher ATPase activity in the presence of K+ [40,66]. Our results here with MceG and the change in intrabacterial [K+] in response to changes in external environment, in combination with the status of K+ as the most abundant intracellular cation in bacterial cells [15], suggest that similar dependencies on K+ are likely to also exist more widely in the bacterial kingdom. Interestingly, while MceG acts as an ATPase for both the Mce1 (fatty acid) and Mce4 (cholesterol) uptake systems [29], ∆ceoBC Mtb grew as well as WT Mtb in media with fatty acids as a sole carbon source (S5 Fig). One possibility raised by this finding is that different levels of MceG activity are required for efficient Mce1 versus Mce4 substrate import. Further studies will be required for mechanistic insight into such differences, and for understanding any impact that fatty acids may have on CeoBC and intrabacterial K+ levels. More broadly, we propose that future studies examining the role of K+ will yield vital insight into a new facet of regulation of bacterial enzymatic activities. They will also provide understanding of whether K+ acts in concert with other divalent cations such as magnesium or manganese, as is often the case in the K+-regulated mammalian enzymes studied to date [39,63].

thumbnail
Download:
Fig 5. Schematic model of cholesterol and K+ interplay in Mtb.

1. Cholesterol (“Chol”) binds to CeoB, affecting K+ uptake through the Trk K+ uptake system and increasing the intrabacterial K+ set point. 2. K+ increases MceG activity, increasing cholesterol uptake through the Mce4 transport system in a feed forward loop. 3. Increased cholesterol uptake enables appropriate Mtb growth and response to cholesterol. In the absence of cholesterol in the local environment, the system resets, with intrabacterial K+ returning to a baseline level.

https://doi.org/10.1371/journal.ppat.1013207.g005

Second, our finding that cholesterol directly binds to CeoB, a component of the Trk K+ uptake system in Mtb, affecting its function and Mtb host colonization in the context of lipid-rich environments, raises the concept of cholesterol as not just a carbon source, but also a molecule capable of regulating key facets of bacterial biology. Cholesterol binding to K+ transport systems in mammalian cells can result in either upregulation or downregulation of K+ transport [44,67]. Here, cholesterol binding to CeoB appears to increase activity of the Trk K+ uptake system, given the observed increase in intrabacterial K+ levels. The one previous example, to our knowledge, of a bacterial K+ uptake system affected by cholesterol is KirBac1.1 from E. coli, where cholesterol was found to inhibit channel activity [46,47]. Those studies were however conducted with purified proteins incorporated into liposomes, with radioactive rubidium (86Rb+) as a proxy for K+ transport measurement [46,47]. Unlike mammalian K+ transport systems, bacterial K+ transport systems often discriminate against Rb+ [68–70]; our establishment of the GINKO2 reporter for relative measurement of intrabacterial K+ in intact bacteria opens the path for future studies examining how cholesterol, or other potential co-factors, affect bacterial K+ uptake in physiological context. Further, our identification of the Y108 residue, part of a CARC motif, as essential for cholesterol binding to CeoB and the effects of cholesterol on CeoB function in intact Mtb cells sets the foundation for mechanistic understanding. Future studies could be aimed at unveiling the precise mechanism by which cholesterol binding to CeoB elevates K+ levels within Mtb; perhaps, for example, by structural-based changes to the uptake system, as has been identified in mammalian systems [44].

The CRAC cholesterol recognition motif was originally put forth from a study focused on the mammalian peripheral-type benzodiazepine receptor that regulates the transport of cholesterol across the mitochondrial outer and inner membranes [71], with later studies describing the inverted CARC motif, and the presence and role of both motifs for various mammalian cholesterol-binding proteins [55,72–74]. In bacterial systems, studies that have examined the CRAC/CARC motifs have largely centered on toxins that interact with cholesterol present in the host cell membrane during the process of target cell intoxication, such as ⍺-hemolysin from E. coli, cytolethal distending toxins from Campylobacteri jejuni and Aggregatibacter actinomycetemcomitans, and leukotoxin from A. actinomycetemcomitans [56,75–77]. For bacteria such as Mtb, Rhodococcus sp., and Gordonia sp. that can utilize cholesterol as a carbon source [26,27,78–81], the possibility of cholesterol functioning in diverse aspects of its biology are greatly expanded, given the active uptake of cholesterol into the bacterium and the consequent presence of cholesterol within the bacterial cell. Intriguingly, examination of the Mtb genome reveals the presence of CRAC and CARC motifs in a wide variety of proteins, spanning transporters to transcription factors. However, in studies with mammalian systems and bacterial toxins, not all previously identified CRAC and CARC motifs have been found to impact cholesterol binding [45,55,82], which is perhaps unsurprising given the relatively loose sequence definition of the motifs. We propose that further study of the possible role of cholesterol in the function of Mtb and other bacterial proteins that encode CRAC or CARC motifs will aid in better defining these motifs, and importantly lead to new discoveries regarding cholesterol-driven regulation of bacterial biology.

Ionic homeostasis and environmental and metabolic adaptation are critical facets for all bacteria, and our results here open new avenues in the understanding of how these facets are integrated. Application and extension of the concepts raised here thus hold exciting potential for revealing both fundamental insight into bacterial biology and key integration nodes that can be targeted to disrupt successful adaptation to local niches by a pathogen.

Materials and methods

Ethics statement

Animal protocols in this research adhered to The National Institutes of Health “Guide for Care and Use of Laboratory Animals” guidelines. The animal protocol (#B2024-90) was reviewed and approved by the Institutional Animal Care and Use Committee at Tufts University, following Association for Assessment and Accreditation of Laboratory Animal Care, the US Department of Agriculture, and the US Public Health Service guidelines.

Mtb strains and culture

Mtb cultures were propagated in standing T25 or T75 flasks with filter caps, in standard 7H9 medium (7H9 base + 10% OADC supplement + 0.2% glycerol + 0.05% Tween 80; hereon referred to as “7H9 medium”), with all media buffered to pH 7 with 100 mM MOPS and antibiotics added to the media as needed at the following concentrations: 100 µg/ml streptomycin, 50 µg/ml hygromycin, 50 µg/ml apramycin, and 25 µg/ml kanamycin. Strains for in vitro assays were in the CDC1551 background, and those for in vivo assays were in the Erdman background. The rv2390c’::GFP, ∆ceoBC mutant and its complement strain, with and without the rv2390c’::GFP reporter, have all been previously described [13]. The ∆kdpBC and ∆mceG mutants and their complements were constructed as previously described [12], with the ∆kdpBC mutation consisting of a deletion beginning at nucleotide 136 of the kdpB open reading frame through nucleotide 302 of the kdpC open reading frame, and the ∆mceG mutation consisting of a deletion beginning at nucleotide 34 through nucleotide 989 of the mceG open reading frame (both as annotated in the Erdman Mtb strain). Complementation of the ∆kdpBC mutant was with a construct containing the kdpBC operon driven by the kdpF promoter, introduced in single copy into the Mtb genome via the pMV306 integrating plasmid. Complementation of the ∆mceG mutant was with a construct containing the mceG open reading frame (as annotated in the Erdman Mtb strain), driven by the hsp60 promoter, introduced in single copy into the Mtb genome via the pMV306 integrating plasmid. The ceoB(Y100S)C* and ceoB(Y108S)C* point mutations were constructed using QuikChange mutagenesis (Agilent). The P606’::GINKO2 construct was generated by cloning an Mtb codon-optimized GINKO2 (GenScript) [38], driven by the P606 promoter, into the destination Gateway vector pDE43-MEK using the Gateway system [83,84].

qRT-PCR analyses

For qRT-PCR analyses, log-phase Mtb cultures (OD600 ~ 0.6) were used to inoculate standing T25 flasks with filter caps at an OD600 = 0.3, containing 10 ml of 7H9 or cholesterol media. Cholesterol medium with 200 µM cholesterol (7H9 base + 0.5 g/l fatty acid-free bovine serum albumin + 14.5 mM NaCl + 200 µM cholesterol + 0.1% tyloxapol) was prepared as previously described [25,85]. Bacteria were incubated in each medium type for 4 hours, before RNA was extracted as previously described [86]. qRT-PCR experiments were conducted and analyzed according to previously established protocols [13]. Briefly, cDNA was synthesized from 250 ng of extracted RNA using the iScript cDNA synthesis kit (Bio-Rad). qRT-PCR was performed using the iTaq Universal SYBR Green Supermix kit (Bio-Rad) on an Applied Biosystems StepOnePlus real-time PCR system, with each sample run in triplicate. The housekeeping gene sigA served as the control, and fold induction was determined using the ∆∆CT method [87].

Dehydroergosterol uptake assays

For dehydroergosterol (DHE; ergosta-5,7,9(11),22-tetraen-3β-ol, Sigma-Aldrich) uptake assay in Mtb, bacteria were cultured to log-phase (OD600 ~ 0.6) in standing T25 flasks with filter caps in 7H9 medium. Strains were then subcultured to OD600 = 0.3 in 7H9 or cholesterol medium, supplemented with 100 µM DHE. 100 μl triplicate aliquots per strain/condition were taken and placed in a clear bottom black 96-well plate (Corning Costar), and incubated for 24 hours. Fluorescence intensity was subsequently measured on a Biotek Synergy Neo2 microplate reader, with excitation 338 nm, emission 381 nm.

Recombinant protein expression and purification

To purify CeoB, CeoB-Y100S, and CeoB-Y108S, the genes were cloned into the pET23a plasmid to generate C-terminal 6xHis-tagged proteins. mceG (rv0655) and kstR1 (rv3574) were each cloned into the pET28a plasmid, generating N-terminal 6xHis-tagged proteins. The CeoC protein was purified by adding a C-terminal Flag tag via PCR and then cloning it into the pET23a vector. Expression plasmids were transformed into Escherichia coli BL21(DE3) for recombinant expression and purification. 1 ml of an overnight E. coli culture started from frozen stock was used to inoculate 1 L of LB medium + 50 μg/ml ampicillin or kanamycin. Cultures were grown at 37°C, 160 rpm, to an OD600 of ~0.6. Protein production was induced with 1 mM IPTG, and cultures were grown for an additional 4 hours at 37°C, 160 rpm. The supernatants were removed, and cell pellets stored at -80°C prior to further processing.

Purification of the CeoC-FLAG-tagged protein and CeoB, CeoB-Y100S and CeoB-Y108S 6xHis-tagged proteins followed previously described protocols [13,88]. MceG 6xHis-tagged protein was present in the insoluble fraction and was purified via treatment of the insoluble fraction with 5M urea buffer, followed by the standard 6xHis-tagged protein purification protocol [13]. CeoB, CeoB-Y100S, CeoB-Y108S, CeoC, and KstR1 proteins were dialyzed into phosphate buffered saline (PBS) buffer. MceG protein was dialyzed into ATPase reaction buffer (50 mM Tris-HCl, 1 mM MgCl2, pH 7.5). Protein concentrations were quantified using a Bradford assay (Bio-Rad).

ATPase activity assay

MceG ATPase activity was measured using the ADP-Glo Kinase assay kit (Promega) following the manufacturer’s instruction. Briefly, purified MceG at the indicated concentrations, or 25 µM KstR1, was mixed with 10 µM ATP and ATPase reaction buffer (50 mM Tris-HCl, 1 mM MgCl2, pH 7.5) in a total volume of 25 µl and incubated at room temperature for 30 minutes. For the treated groups, indicated concentrations of different compounds (KCl, NaCl, or K2SO4) were included in the ATPase reaction buffer. After incubation, 25 µl of ADP-Glo reagent was added to each reaction mixture and incubated at room temperature for a further 30 minutes to stop the reaction and deplete unused ATP. Then, 50 µl of kinase detection reagent was added to convert ADP to ATP and introduce luciferin for ATP measurement, followed by a 30 minute incubation in the dark. Triplicate 90 µl aliquots of the final reaction were transferred into a clear-bottom white 96-well plate (Corning Costar) for each sample and luminescence measured with a BioTek H1 multimode microplate reader.

Cholesterol binding assays

For thermostability shift cholesterol binding assays, purified proteins (CeoB, CeoB-Y100S, CeoB-Y108S at 15 µg and CeoC at 2.5 µg) were incubated with 5 µM cholesterol, glycerol, or glucose at room temperature for 20 minutes. The mixtures were then incubated at the indicated temperatures for 5 minutes (for CeoB and its point mutants) or 20 minutes (for CeoC). After incubation, the samples were centrifuged at 4ºC for 30 minutes and the supernatant then run on 12% SDS-PAGE gels. Gels were transferred to Millipore Immobilon-FL PVDF membranes and blocked overnight at 4°C with LI-COR Odyssey blocking buffer. The membranes were then incubated with either mouse anti-6x-His-tag antibody (Invitrogen) or mouse anti-FLAG antibody (Sigma-Aldrich) as needed, at a dilution of 1:1000 for 1 hour at room temperature. Following incubation, the membranes were washed three times for 5 minutes each with PBS + 0.1% Tween 20. They were then incubated with goat anti-mouse IRDye 680RD (LI-COR) at a dilution of 1:3000 for 1 hour at room temperature. After incubation, the membranes were washed three times for 5 minutes each with PBS + 0.1% Tween 20 and then rinsed with deionized water. Protein bands were visualized using a LI-COR Odyssey CLx imaging system, and signal intensities quantified using Image Studio software (LI-COR).

Protein binding assays with DHE (Sigma) were carried out following a previously published protocol [34,36]. Briefly, purified proteins (10 µM) were mixed with 1 µM DHE in DHE buffer (10 mM HEPES, pH 7.4, and 150 mM NaCl) in a 200 µl volume. The mixture was transferred into a black, clear bottom 96-well plate and incubated at room temperature for 30 minutes in the dark. Fluorescence was measured on a BioTek H1 multimode microplate reader with excitation 338 nm, emission 381 nm.

Mtb reporter assays and growth assays

For GINKO2 reporter assays, strains carrying the P606’::GINKO2 were propagated to log phase and subcultured to an OD600 = 0.3 in (i) 7H9, (ii) K+-free 7H9, (iii) cholesterol medium, or (iv) K+-free cholesterol medium, in standing T25 flasks with filter caps. K+-free 7H9 was prepared as previously described [13], where 7H9 was made from individual components, replacing monopotassium phosphate with monosodium phosphate, and all other supplements unchanged from standard 7H9 medium. K+-free cholesterol medium was similarly prepared with 7H9 base made from individual components, replacing monopotassium phosphate with monosodium phosphate, and all other supplements unchanged from cholesterol medium. Prior to resuspending the cultures in the final assay media, an additional wash step with K+-free 7H9 medium was performed. All cultures were then incubated at 37°C for 6 days. After 6 days, aliquots were taken and fixed in 4% paraformaldehyde (PFA) in PBS. The fixed samples were pelleted and then resuspended in PBS + 0.1% Tween 80 for flow cytometry analysis on a BD FACSCalibur. Just prior to running, each sample was passed six times through a tuberculin syringe (25G × 5/8” needle) to disrupt clumps. Reporter signal from 10,000 Mtb cells per sample per experimental run were obtained, with three independent runs conducted. Mean fluorescence value for each sample was determined using FlowJo software (BD).

rv2390c’::GFP reporter assays were performed as previously described [12]. In brief, strains were propagated to log phase and subcultured to an OD600 = 0.05 in 7H9 medium ± 250 mM NaCl, in standing T25 flasks with filter caps. At each time point, aliquots were taken and fixed in 4% PFA in PBS. Reporter signal was analyzed via flow cytometry as described above.

For growth assays, log-phase Mtb cultures were used to inoculate 10 ml of 7H9 or cholesterol medium at a starting OD600 = 0.05 in standing T25 flasks with filter caps. OD600 was measured at indicated time points. Where indicated, a control medium = 7H9 base + 0.5 g/l fatty acid-free bovine serum albumin + 14.5 mM NaCl + 0.2% glycerol + 2 g/l D-glucose (dextrose) + 0.1% tyloxapol was also utilized (carbon source as the only difference to cholesterol medium). Media with fatty acids as a sole carbon source consisted of 7H9 base + 0.5 g/l fatty acid-free bovine serum albumin + 14.5 mM NaCl + 0.05% tyloxapol, with either 1 mM sodium acetate or 0.05 mM sodium palmitate added [89]. For palmitate, a 100 mM stock was first made in a 1:1 ethanol:tyloxapol solution [89].

Macrophage culture and infections

Bone marrow-derived macrophages were isolated from C57BL/6J wild type mice procured from Jackson Laboratories. The cells were cultured in DMEM containing 10% FBS, 10% L929-cell conditioned media, 2 mM L-glutamine, 1 mM sodium pyruvate, and antibiotics (penicillin/streptomycin) as needed. They were maintained in a 37°C incubator with 5% CO2. To generate foamy macrophages, cells were pre-treated with macrophage medium supplemented with oleate/albumin complexes (0.42 mM sodium oleate, 0.35% BSA) 24 hours before infection, as previously described [59,60]. Infections of macrophages with Mtb were performed as previously described [12,90]. For colony forming unit (CFU) enumeration, macrophages were lysed in water containing 0.01% sodium dodecyl sulfate (SDS), and serial dilutions plated on 7H10 agar plates containing 100 μg/ml cycloheximide.

Mouse Mtb infections

C3HeB/FeJ wild-type mice from Jackson Laboratories were intranasally infected with 103 CFUs of Mtb in a 35 μl volume while under light anesthesia using 2% isoflurane [25,85,91]. At 2 or 6 weeks post-infection, the mice were sacrificed using CO2. The left lobe and accessory right lobe of the lungs were then homogenized in PBS + 0.05% Tween 80. Serial dilutions of the lung homogenates were plated on 7H10 agar plates supplemented with 100 μg/ml cycloheximide to quantify CFUs.

Quantification and statistical analysis

All statistical analyses were performed using GraphPad Prism. Specific statistical tests conducted are described in the figure legends. A p-value of less than 0.05 was considered statistically significant.

Supporting information

S1 Fig. Disruption of the inducible high affinity Kdp K+ transport system does not affect Mtb response to cholesterol or intrabacterial [K+].

(A) Deletion of kdpBC does not affect Mtb response to cholesterol. WT, ∆kdpBC, and kdpBC* (complemented mutant) Mtb were exposed to 7H9 or cholesterol medium for 4 hours, before RNA extraction for qRT-PCR analysis. Fold change is as compared to the 7H9 condition, with sigA as the control gene. (B) Deletion of kdpBC does not affect intrabacterial [K+] in Mtb. WT, ∆kdpBC, and kdpBC* Mtb each carrying the P606’::GINKO2 reporter were subcultured to OD600 = 0.3 into: (i) 7H9 medium, (ii) K+-free 7H9 medium, (iii) cholesterol medium, or (iv) K+-free cholesterol medium. GINKO2 fluorescence was measured by flow cytometry 6 days post-assay start. Reporter signal from 10,000 Mtb cells per sample per experimental run were obtained for determination of the mean GINKO2 fluorescence for each sample. Data in both panels are shown as means ± SEM from 3 experiments. p-values were obtained with an unpaired t-test with Welch’s correction and Holm-Sidak multiple comparisons in (A), and a two-way ANOVA with Tukey’s multiple comparisons test in (B). N.S. not significant, * p < 0.05. The numerical data underlying the graphs shown in this figure are provided in S1 Data.

https://doi.org/10.1371/journal.ppat.1013207.s001

(TIF)

S2 Fig. Disruption of K+ homeostasis inhibits Mtb growth in cholesterol but not glucose/glycerol media.

WT, ∆ceoBC, and ceoBC* Mtb were grown in media with either 2 g/l glucose + 0.2% glycerol or 200 µM cholesterol as the sole carbon source, and OD600 monitored over time. Data are shown as means ± SEM from three experiments. Statistical analyses were performed using an unpaired t-test with Welch’s correction and Holm-Sidak multiple comparisons. Comparisons of ∆ceoBC to WT in the cholesterol condition are shown. ** p < 0.01, *** p < 0.001. The numerical data underlying the graphs shown in this figure are provided in S1 Data.

https://doi.org/10.1371/journal.ppat.1013207.s002

(TIF)

S3 Fig. DHE reports decreased cholesterol uptake by ∆mceG Mtb.

Log-phase WT, ∆mceG, mceG* (complemented mutant), ∆ceoBC, and ceoBC* (complemented mutant) Mtb were exposed to 7H9 or cholesterol media, supplemented with 1 µM dehydroergosterol (DHE), for 24 hours. DHE uptake into Mtb was measured via analysis of DHE fluorescence on a microplate reader, normalized against OD600. Data are shown as means ± SEM from three experiments. p-values were obtained with a two-way ANOVA with Tukey’s multiple comparisons test. All comparisons in the 7H9 condition were non-significant. **** p < 0.0001. The numerical data underlying the graphs shown in this figure are provided in S1 Data.

https://doi.org/10.1371/journal.ppat.1013207.s003

(TIF)

S4 Fig. CeoB contains cholesterol recognition amino acid consensus (CRAC/CARC) motifs.

(A) shows the amino acid sequence of CeoB (from the genome of the Erdman strain, accession #AP012340.1 [92]) with the CRAC motif boxed in blue and the CARC motif boxed in red. The key tyrosine residue in each of the motifs are bolded and marked in the AlphaFold-predicted structure [93,94] shown in (B).

https://doi.org/10.1371/journal.ppat.1013207.s004

(TIF)

S5 Fig. ceoBC deletion does not affect Mtb growth in fatty acid media.

WT, ∆ceoBC, and ceoBC* Mtb were grown in 7H9 or media with acetate (“C2”) or palmitate (“C16”) as a carbon source, and OD600 monitored over time. Data are shown as means ± SEM from three experiments. The numerical data underlying the graphs shown in this figure are provided in S1 Data.

https://doi.org/10.1371/journal.ppat.1013207.s005

(TIF)

S1 Data. Numerical data underlying the presented graphs.

Excel file with numerical data underlying graphed data presented in all figure files.

https://doi.org/10.1371/journal.ppat.1013207.s006

(XLSX)

Acknowledgments

We thank members of the Tan laboratory for helpful discussion.

References

  1. 1. Armstrong CM, Hollingworth S. A perspective on Na and K channel inactivation. J Gen Physiol. 2017;150(1):7–18. pmid:29233885
  2. 2. Cheng H, Lederer WJ. Calcium sparks. Physiol Rev. 2008;88(4):1491–545. pmid:18923188
  3. 3. Palmer LD, Skaar EP. Transition Metals and Virulence in Bacteria. Annu Rev Genet. 2016;50:67–91. pmid:27617971
  4. 4. Fontecave M. Iron-sulfur clusters: ever-expanding roles. Nat Chem Biol. 2006;2(4):171–4. pmid:16547473
  5. 5. Raut SK, Singh K, Sanghvi S, Loyo-Celis V, Varghese L, Singh ER, et al. Chloride ions in health and disease. Biosci Rep. 2024;44(5):BSR20240029. pmid:38573803
  6. 6. Tan S. Abundant Monovalent Ions as Environmental Signposts for Pathogens during Host Colonization. Infect Immun. 2021;89(4):e00641-20. pmid:33526568
  7. 7. Subashchandrabose S, Mobley HLT. Back to the metal age: battle for metals at the host-pathogen interface during urinary tract infection. Metallomics. 2015;7(6):935–42. pmid:25677827
  8. 8. Chandrangsu P, Rensing C, Helmann JD. Metal homeostasis and resistance in bacteria. Nat Rev Microbiol. 2017;15(6):338–50. pmid:28344348
  9. 9. Loughman JA, Caparon M. Regulation of SpeB in Streptococcus pyogenes by pH and NaCl: a model for in vivo gene expression. J Bacteriol. 2006;188(2):399–408. pmid:16385029
  10. 10. Cakar F, Zingl FG, Moisi M, Reidl J, Schild S. In vivo repressed genes of Vibrio cholerae reveal inverse requirements of an H+/Cl− transporter along the gastrointestinal passage. Proc Natl Acad Sci U S A. 2018;115(10):E2376–85. pmid:29463743
  11. 11. Yost S, Duran-Pinedo AE, Krishnan K, Frias-Lopez J. Potassium is a key signal in host-microbiome dysbiosis in periodontitis. PLoS Pathog. 2017;13(6):e1006457. pmid:28632755
  12. 12. Tan S, Sukumar N, Abramovitch RB, Parish T, Russell DG. Mycobacterium tuberculosis responds to chloride and pH as synergistic cues to the immune status of its host cell. PLoS Pathog. 2013;9(4):e1003282. pmid:23592993
  13. 13. MacGilvary NJ, Kevorkian YL, Tan S. Potassium response and homeostasis in Mycobacterium tuberculosis modulates environmental adaptation and is important for host colonization. PLoS Pathog. 2019;15(2):e1007591. pmid:30716121
  14. 14. Gries CM, Bose JL, Nuxoll AS, Fey PD, Bayles KW. The Ktr potassium transport system in Staphylococcus aureus and its role in cell physiology, antimicrobial resistance and pathogenesis. Mol Microbiol. 2013;89(4):760–73. pmid:23815639
  15. 15. Epstein W. The roles and regulation of potassium in bacteria. Prog Nucleic Acid Res Mol Biol. 2003;75:293–320. pmid:14604015
  16. 16. Zacchia M, Abategiovanni ML, Stratigis S, Capasso G. Potassium: From Physiology to Clinical Implications. Kidney Dis (Basel). 2016;2(2):72–9. pmid:27536695
  17. 17. Muñoz-Planillo R, Kuffa P, Martínez-Colón G, Smith BL, Rajendiran TM, Núñez G. K⁺ efflux is the common trigger of NLRP3 inflammasome activation by bacterial toxins and particulate matter. Immunity. 2013;38(6):1142–53. pmid:23809161
  18. 18. Pétrilli V, Papin S, Dostert C, Mayor A, Martinon F, Tschopp J. Activation of the NALP3 inflammasome is triggered by low intracellular potassium concentration. Cell Death Differ. 2007;14(9):1583–9. pmid:17599094
  19. 19. Wood JM. Bacterial osmoregulation: a paradigm for the study of cellular homeostasis. Annu Rev Microbiol. 2011;65:215–38. pmid:21663439
  20. 20. Liu Y, Ho KK, Su J, Gong H, Chang AC, Lu S. Potassium transport of Salmonella is important for type III secretion and pathogenesis. Microbiology (Reading). 2013;159(Pt 8):1705–19. pmid:23728623
  21. 21. Binepal G, Gill K, Crowley P, Cordova M, Brady LJ, Senadheera DB, et al. Trk2 Potassium Transport System in Streptococcus mutans and Its Role in Potassium Homeostasis, Biofilm Formation, and Stress Tolerance. J Bacteriol. 2016;198(7):1087–100. pmid:26811321
  22. 22. World Health Organization. Global tuberculosis report. 2024 [cited 2 May 2025. ]. https://www.who.int/publications/i/item/9789240101531
  23. 23. Vandal OH, Pierini LM, Schnappinger D, Nathan CF, Ehrt S. A membrane protein preserves intrabacterial pH in intraphagosomal Mycobacterium tuberculosis. Nat Med. 2008;14(8):849–54. pmid:18641659
  24. 24. Cholo MC, van Rensburg EJ, Anderson R. Potassium uptake systems of Mycobacterium tuberculosis: genomic and protein organisation and potential roles in microbial pathogenesis and chemotherapy. Southern African Journal of Epidemiology and Infection. 2008;23(4):13–6.
  25. 25. Chen Y, MacGilvary NJ, Tan S. Mycobacterium tuberculosis response to cholesterol is integrated with environmental pH and potassium levels via a lipid metabolism regulator. PLoS Genet. 2024;20(1):e1011143. pmid:38266039
  26. 26. Pandey AK, Sassetti CM. Mycobacterial persistence requires the utilization of host cholesterol. Proc Natl Acad Sci U S A. 2008;105(11):4376–80. pmid:18334639
  27. 27. Nesbitt NM, Yang X, Fontán P, Kolesnikova I, Smith I, Sampson NS, et al. A thiolase of Mycobacterium tuberculosis is required for virulence and production of androstenedione and androstadienedione from cholesterol. Infect Immun. 2010;78(1):275–82. pmid:19822655
  28. 28. Joshi SM, Pandey AK, Capite N, Fortune SM, Rubin EJ, Sassetti CM. Characterization of mycobacterial virulence genes through genetic interaction mapping. Proc Natl Acad Sci U S A. 2006;103(31):11760–5. pmid:16868085
  29. 29. Fieweger RA, Wilburn KM, Montague CR, Roszkowski EK, Kelly CM, Southard TL, et al. MceG stabilizes the Mce1 and Mce4 transporters in Mycobacterium tuberculosis. J Biol Chem. 2023;299(3):102910. pmid:36642182
  30. 30. Nazarova EV, Montague CR, La T, Wilburn KM, Sukumar N, Lee W, et al. Rv3723/LucA coordinates fatty acid and cholesterol uptake in Mycobacterium tuberculosis. Elife. 2017;6:e26969. pmid:28708968
  31. 31. Wipperman MF, Sampson NS, Thomas ST. Pathogen roid rage: cholesterol utilization by Mycobacterium tuberculosis. Crit Rev Biochem Mol Biol. 2014;49(4):269–93. pmid:24611808
  32. 32. Pawełczyk J, Brzostek A, Minias A, Płociński P, Rumijowska-Galewicz A, Strapagiel D, et al. Cholesterol-dependent transcriptome remodeling reveals new insight into the contribution of cholesterol to Mycobacterium tuberculosis pathogenesis. Sci Rep. 2021;11(1):12396. pmid:34117327
  33. 33. Liu R, Lu P, Chu JWK, Sharom FJ. Characterization of fluorescent sterol binding to purified human NPC1. J Biol Chem. 2009;284(3):1840–52. pmid:19029290
  34. 34. Friedland N, Liou H-L, Lobel P, Stock AM. Structure of a cholesterol-binding protein deficient in Niemann-Pick type C2 disease. Proc Natl Acad Sci U S A. 2003;100(5):2512–7. pmid:12591954
  35. 35. Schroeder F, Butko P, Nemecz G, Scallen TJ. Interaction of fluorescent delta 5,7,9(11),22-ergostatetraen-3 beta-ol with sterol carrier protein-2. J Biol Chem. 1990;265(1):151–7. pmid:2294101
  36. 36. Hayward RD, Cain RJ, McGhie EJ, Phillips N, Garner MJ, Koronakis V. Cholesterol binding by the bacterial type III translocon is essential for virulence effector delivery into mammalian cells. Mol Microbiol. 2005;56(3):590–603. pmid:15819617
  37. 37. Kayath CA, Hussey S, El hajjami N, Nagra K, Philpott D, Allaoui A. Escape of intracellular Shigella from autophagy requires binding to cholesterol through the type III effector, IcsB. Microbes Infect. 2010;12(12–13):956–66. pmid:20599519
  38. 38. Wu S-Y, Wen Y, Serre NBC, Laursen CCH, Dietz AG, Taylor BR, et al. A sensitive and specific genetically-encoded potassium ion biosensor for in vivo applications across the tree of life. PLoS Biol. 2022;20(9):e3001772. pmid:36067248
  39. 39. Gohara DW, Di Cera E. Molecular Mechanisms of Enzyme Activation by Monovalent Cations. J Biol Chem. 2016;291(40):20840–8. pmid:27462078
  40. 40. O’Brien MC, McKay DB. How potassium affects the activity of the molecular chaperone Hsc70. I. Potassium is required for optimal ATPase activity. J Biol Chem. 1995;270(5):2247–50. pmid:7836457
  41. 41. Wu Y, Qian X, He Y, Moya IA, Luo Y. Crystal structure of an ATPase-active form of Rad51 homolog from Methanococcus voltae. Insights into potassium dependence. J Biol Chem. 2005;280(1):722–8. pmid:15537659
  42. 42. Kendall SL, Withers M, Soffair CN, Moreland NJ, Gurcha S, Sidders B, et al. A highly conserved transcriptional repressor controls a large regulon involved in lipid degradation in Mycobacterium smegmatis and Mycobacterium tuberculosis. Mol Microbiol. 2007;65(3):684–99. pmid:17635188
  43. 43. Stautz J, Hellmich Y, Fuss MF, Silberberg JM, Devlin JR, Stockbridge RB, et al. Molecular Mechanisms for Bacterial Potassium Homeostasis. J Mol Biol. 2021;433(16):166968. pmid:33798529
  44. 44. Corradi V, Bukiya AN, Miranda WE, Cui M, Plant LD, Logothetis DE, et al. A molecular switch controls the impact of cholesterol on a Kir channel. Proc Natl Acad Sci U S A. 2022;119(13):e2109431119. pmid:35333652
  45. 45. Rosenhouse-Dantsker A, Noskov S, Durdagi S, Logothetis DE, Levitan I. Identification of novel cholesterol-binding regions in Kir2 channels. J Biol Chem. 2013;288(43):31154–64. pmid:24019518
  46. 46. Singh DK, Rosenhouse-Dantsker A, Nichols CG, Enkvetchakul D, Levitan I. Direct regulation of prokaryotic Kir channel by cholesterol. J Biol Chem. 2009;284(44):30727–36. pmid:19740741
  47. 47. Singh DK, Shentu T-P, Enkvetchakul D, Levitan I. Cholesterol regulates prokaryotic Kir channel by direct binding to channel protein. Biochim Biophys Acta. 2011;1808(10):2527–33. pmid:21798234
  48. 48. Balajthy A, Hajdu P, Panyi G, Varga Z. Sterol Regulation of Voltage-Gated K+ Channels. Curr Top Membr. 2017;80:255–92. pmid:28863820
  49. 49. Lee AG. Interfacial Binding Sites for Cholesterol on Kir, Kv, K2P, and Related Potassium Channels. Biophys J. 2020;119(1):35–47. pmid:32553129
  50. 50. Jiang Q-X. Cholesterol-Dependent Gating Effects on Ion Channels. Adv Exp Med Biol. 2019;1115:167–90. pmid:30649760
  51. 51. Mackenzie AB, Chirakkal H, North RA. Kv1.3 potassium channels in human alveolar macrophages. Am J Physiol Lung Cell Mol Physiol. 2003;285(4):L862-8. pmid:12909584
  52. 52. O’Grady SM, Lee SY. Chloride and potassium channel function in alveolar epithelial cells. Am J Physiol Lung Cell Mol Physiol. 2003;284(5):L689-700. pmid:12676759
  53. 53. Dechow SJ, Baker JJ, Murto M, Abramovitch RB. ppe51 Variants Enable Growth of Mycobacterium tuberculosis at Acidic pH by Selectively Promoting Glycerol Uptake. J Bacteriol. 2022;204(11):e0021222. pmid:36226966
  54. 54. Park Y, Ahn Y-M, Jonnala S, Oh S, Fisher JM, Goodwin MB, et al. Inhibition of CorA-Dependent Magnesium Homeostasis Is Cidal in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 2019;63(10):e01006-19. pmid:31383669
  55. 55. Fantini J, Barrantes FJ. How cholesterol interacts with membrane proteins: an exploration of cholesterol-binding sites including CRAC, CARC, and tilted domains. Front Physiol. 2013;4:31. pmid:23450735
  56. 56. Cané L, Guzmán F, Balatti G, Daza Millone MA, Pucci Molineris M, Maté S, et al. Biophysical Analysis to Assess the Interaction of CRAC and CARC Motif Peptides of Alpha Hemolysin of Escherichia coli with Membranes. Biochemistry. 2023;62(12):1994–2011. pmid:37224476
  57. 57. Vazquez RF, Maté SM, Bakás LS, Fernández MM, Malchiodi EL, Herlax VS. Novel evidence for the specific interaction between cholesterol and α-haemolysin of Escherichia coli. Biochem J. 2014;458(3):481–9. pmid:24351077
  58. 58. Irwin SM, Driver E, Lyon E, Schrupp C, Ryan G, Gonzalez-Juarrero M, et al. Presence of multiple lesion types with vastly different microenvironments in C3HeB/FeJ mice following aerosol infection with Mycobacterium tuberculosis. Dis Model Mech. 2015;8(6):591–602. pmid:26035867
  59. 59. Podinovskaia M, Lee W, Caldwell S, Russell DG. Infection of macrophages with Mycobacterium tuberculosis induces global modifications to phagosomal function. Cell Microbiol. 2013;15(6):843–59. pmid:23253353
  60. 60. Listenberger LL, Studer AM, Brown DA, Wolins NE. Fluorescent Detection of Lipid Droplets and Associated Proteins. Curr Protoc Cell Biol. 2016;71:4.31.1-4.31.14. pmid:27245427
  61. 61. Kunkle DE, Skaar EP. Moving metals: How microbes deliver metal cofactors to metalloproteins. Mol Microbiol. 2023;120(4):547–54. pmid:37408317
  62. 62. Waldron KJ, Robinson NJ. How do bacterial cells ensure that metalloproteins get the correct metal?. Nat Rev Microbiol. 2009;7(1):25–35. pmid:19079350
  63. 63. Vašák M, Schnabl J. Sodium and Potassium Ions in Proteins and Enzyme Catalysis. Met Ions Life Sci. 2016;16:259–90. pmid:26860304
  64. 64. Machius M, Chuang JL, Wynn RM, Tomchick DR, Chuang DT. Structure of rat BCKD kinase: nucleotide-induced domain communication in a mitochondrial protein kinase. Proc Natl Acad Sci U S A. 2001;98(20):11218–23. pmid:11562470
  65. 65. Kato M, Chuang JL, Tso S-C, Wynn RM, Chuang DT. Crystal structure of pyruvate dehydrogenase kinase 3 bound to lipoyl domain 2 of human pyruvate dehydrogenase complex. EMBO J. 2005;24(10):1763–74. pmid:15861126
  66. 66. Buch-Pedersen MJ, Rudashevskaya EL, Berner TS, Venema K, Palmgren MG. Potassium as an intrinsic uncoupler of the plasma membrane H+-ATPase. J Biol Chem. 2006;281(50):38285–92. pmid:17056603
  67. 67. Rosenhouse-Dantsker A, Leal-Pinto E, Logothetis DE, Levitan I. Comparative analysis of cholesterol sensitivity of Kir channels: role of the CD loop. Channels (Austin). 2010;4(1):63–6. pmid:19923917
  68. 68. Rhoads DB, Woo A, Epstein W. Discrimination between Rb+ and K+ by Escherichia coli. Biochim Biophys Acta. 1977;469(1):45–51. pmid:329878
  69. 69. Kobayashi H. Second system for potassium transport in Streptococcus faecalis. J Bacteriol. 1982;150(2):506–11. pmid:6279560
  70. 70. Michels M, Bakker EP. Low-affinity potassium uptake system in Bacillus acidocaldarius. J Bacteriol. 1987;169(9):4335–41. pmid:3624206
  71. 71. Li H, Papadopoulos V. Peripheral-type benzodiazepine receptor function in cholesterol transport. Identification of a putative cholesterol recognition/interaction amino acid sequence and consensus pattern. Endocrinology. 1998;139(12):4991–7. pmid:9832438
  72. 72. Baier CJ, Fantini J, Barrantes FJ. Disclosure of cholesterol recognition motifs in transmembrane domains of the human nicotinic acetylcholine receptor. Sci Rep. 2011;1:69. pmid:22355588
  73. 73. Wang Q, Cao Y, Shen L, Xiao T, Cao R, Wei S, et al. Regulation of PD-L1 through direct binding of cholesterol to CRAC motifs. Sci Adv. 2022;8(34):eabq4722. pmid:36026448
  74. 74. Singh AK, McMillan J, Bukiya AN, Burton B, Parrill AL, Dopico AM. Multiple cholesterol recognition/interaction amino acid consensus (CRAC) motifs in cytosolic C tail of Slo1 subunit determine cholesterol sensitivity of Ca2+- and voltage-gated K+ (BK) channels. J Biol Chem. 2012;287(24):20509–21. pmid:22474334
  75. 75. Brown AC, Balashova NV, Epand RM, Epand RF, Bragin A, Kachlany SC, et al. Aggregatibacter actinomycetemcomitans leukotoxin utilizes a cholesterol recognition/amino acid consensus site for membrane association. J Biol Chem. 2013;288(32):23607–21. pmid:23792963
  76. 76. Boesze-Battaglia K, Walker LP, Zekavat A, Dlakić M, Scuron MD, Nygren P, et al. The Aggregatibacter actinomycetemcomitans Cytolethal Distending Toxin Active Subunit CdtB Contains a Cholesterol Recognition Sequence Required for Toxin Binding and Subunit Internalization. Infect Immun. 2015;83(10):4042–55. pmid:26216427
  77. 77. Lai C-H, Lai C-K, Lin Y-J, Hung C-L, Chu C-H, Feng C-L, et al. Characterization of putative cholesterol recognition/interaction amino acid consensus-like motif of Campylobacter jejuni cytolethal distending toxin C. PLoS One. 2013;8(6):e66202. pmid:23762481
  78. 78. Griffin JE, Pandey AK, Gilmore SA, Mizrahi V, McKinney JD, Bertozzi CR, et al. Cholesterol catabolism by Mycobacterium tuberculosis requires transcriptional and metabolic adaptations. Chem Biol. 2012;19(2):218–27. pmid:22365605
  79. 79. García JL, Uhía I, Galán B. Catabolism and biotechnological applications of cholesterol degrading bacteria. Microb Biotechnol. 2012;5(6):679–99. pmid:22309478
  80. 80. Van der Geize R, Yam K, Heuser T, Wilbrink MH, Hara H, Anderton MC, et al. A gene cluster encoding cholesterol catabolism in a soil actinomycete provides insight into Mycobacterium tuberculosis survival in macrophages. Proc Natl Acad Sci U S A. 2007;104(6):1947–52. pmid:17264217
  81. 81. Drzyzga O, Fernández de las Heras L, Morales V, Navarro Llorens JM, Perera J. Cholesterol degradation by Gordonia cholesterolivorans. Appl Environ Microbiol. 2011;77(14):4802–10. pmid:21622796
  82. 82. Masin J, Roderova J, Osickova A, Novak P, Bumba L, Fiser R, et al. The conserved tyrosine residue 940 plays a key structural role in membrane interaction of Bordetella adenylate cyclase toxin. Sci Rep. 2017;7(1):9330. pmid:28839199
  83. 83. Huang L, Kushner NL, Theriault ME, Pisu D, Tan S, McNamara CW, et al. The Deconstructed Granuloma: A Complex High-Throughput Drug Screening Platform for the Discovery of Host-Directed Therapeutics Against Tuberculosis. Front Cell Infect Microbiol. 2018;8:275. pmid:30155446
  84. 84. Blumenthal A, Trujillo C, Ehrt S, Schnappinger D. Simultaneous analysis of multiple Mycobacterium tuberculosis knockdown mutants in vitro and in vivo. PLoS One. 2010;5(12):e15667. pmid:21203517
  85. 85. Lavin RC, Johnson C, Ahn Y-M, Kremiller KM, Sherwood M, Patel JS, et al. Targeting Mycobacterium tuberculosis response to environmental cues for the development of effective antitubercular drugs. PLoS Biol. 2021;19(7):e3001355. pmid:34319985
  86. 86. Rohde KH, Abramovitch RB, Russell DG. Mycobacterium tuberculosis invasion of macrophages: linking bacterial gene expression to environmental cues. Cell Host Microbe. 2007;2(5):352–64. pmid:18005756
  87. 87. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402–8. pmid:11846609
  88. 88. Giacalone D, Yap RE, Ecker AMV, Tan S. PrrA modulates Mycobacterium tuberculosis response to multiple environmental cues and is critically regulated by serine/threonine protein kinases. PLoS Genet. 2022;18(8):e1010331. pmid:35913986
  89. 89. Lee W, VanderVen BC, Fahey RJ, Russell DG. Intracellular Mycobacterium tuberculosis exploits host-derived fatty acids to limit metabolic stress. J Biol Chem. 2013;288(10):6788–800. pmid:23306194
  90. 90. Abramovitch RB, Rohde KH, Hsu F-F, Russell DG. aprABC: a Mycobacterium tuberculosis complex-specific locus that modulates pH-driven adaptation to the macrophage phagosome. Mol Microbiol. 2011;80(3):678–94. pmid:21401735
  91. 91. Lavin RC, Tan S. Spatial relationships of intra-lesion heterogeneity in Mycobacterium tuberculosis microenvironment, replication status, and drug efficacy. PLoS Pathog. 2022;18(3):e1010459. pmid:35344572
  92. 92. Miyoshi-Akiyama T, Matsumura K, Iwai H, Funatogawa K, Kirikae T. Complete annotated genome sequence of Mycobacterium tuberculosis Erdman. J Bacteriol. 2012;194(10):2770. pmid:22535945
  93. 93. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596(7873):583–9. pmid:34265844
  94. 94. Varadi M, Bertoni D, Magana P, Paramval U, Pidruchna I, Radhakrishnan M, et al. AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences. Nucleic Acids Res. 2024;52(D1):D368–75. pmid:37933859