• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Microbiol. Author manuscript; available in PMC Sep 1, 2011.
Published in final edited form as:
PMCID: PMC2965804

CtpV: A putative copper exporter required for full virulence of Mycobacterium tuberculosis


Copper is a required micronutrient that is also toxic at excess concentrations. Currently, little is known about the role of copper in interactions between bacterial pathogens and their human hosts. In this study, we elucidate a mechanism for copper homeostasis in the human pathogen Mycobacterium tuberculosis via characterization of a putative copper exporter, CtpV. CtpV was shown to be required by M. tuberculosis to maintain resistance to copper toxicity. Furthermore, the deletion of ctpV resulted in a 98-gene transcriptional response which elucidates the increased stress experienced by the bacteria in the absence of this detoxification mechanism. Interestingly, although the ΔctpV mutant survives close to the wild-type levels in both murine and guinea pig models of tuberculosis, animals infected with the ΔctpV mutant displayed decreased lung damage, and mutant-infected mice had a reduced immune response to the bacteria as well as a significant increase in survival time relative to mice infected with wild-type M. tuberculosis. Overall, our study provides the first evidence for a connection between bacterial copper response and the virulence of M. tuberculosis, supporting the hypothesis that copper response could be important to intracellular pathogens, in general.


Many enzymes require a metal cofactor for activity. The metals that serve as cofactors in enzymes required for life are considered biologically active metals (biometals). These metals, including iron, copper, zinc, and magnesium, are required micronutrients for many diverse cellular organisms, from humans to bacteria. For bacteria that colonize the human body, this can provide a form of environmental stress, as microbes and host cells struggle to maintain appropriate levels of the same micronutrients. The most commonly used metal cofactor, iron, has frequently been studied in the context of this struggle (Schaible et al., 2004). Required by both host cells and bacteria for a number of enzymatic activities, including respiration and detoxification, iron is kept bound by host proteins such as transferrin and lactoferrin. Successful human pathogens, including Mycobacterium tuberculosis (Mtb), have developed iron-specific uptake mechanisms and regulators which contribute to iron scavenging and survival within a host (De Voss et al., 2000; Smith, 2003). Another biologically active metal, copper, is also required by both host and bacterial enzymes, including oxidases and superoxide dismutase (MacPherson et al., 2007). The potential function of copper in host/microbe interactions has not yet been elucidated. However, studies of copper homeostasis mechanisms in pathogenic organisms have shown that copper export, in addition to acquisition, is important for virulence. For example, copper export knockout mutants of the human pathogen Pseudomonas aeruginosa displayed reduced colonization levels when tested in murine models, as well as the plant pathogens Pseudomonas fluorescens and Xanthomonas axonopodis when grown in plant hosts (Schwan et al., 2005; Zhang et al., 2007; Teixeira et al., 2008).

The previous studies of biometals suggest that although pathogenic bacteria must obtain sufficient amounts of micronutrients, they are also sensitive to metal toxicity. Therefore, an equal balance of metal import and export (homeostasis) must be obtained at levels appropriate for each metal. Presumably, the ability to sense metals in the environment and regulate the expression of homeostasis mechanisms is key to maintaining intracellular metals at appropriate concentrations. Recently, the first copper-binding transcriptional regulator in Mtb was identified (CsoR) (Liu et al., 2007), showing that this important human pathogen has the ability to respond to copper in its natural environment, the phagosome of the human macrophage. It has been shown that copper levels in primary cultured macrophages fluctuate after phagocytosis of Mtb, with detected copper levels ranging from approximately 25–426 μM over a 24 hour course of infection (Wagner et al., 2005). Additionally, it was recently shown that reactive nitrogen intermediates found in the macrophage environment cause the release of copper bound to the intracellular metallothionein protein MymT in Mtb, providing an indirect form of copper stress mediated by the host (Gold et al., 2008).

CsoR, the only known copper-binding regulator in Mtb, is encoded within a region of the Mtb genome previously associated with in vivo survival, named the iVEGI island (Talaat et al., 2004). CsoR was shown to repress the expression of its own operon, the cso operon, in the absence of copper (Liu et al., 2007). CsoR loses its ability to repress cso expression when it binds to copper, resulting in copper-induced cso gene expression occurring at levels proportional to the levels of intracellular copper. The cso operon includes the ctpV gene, which encodes a putative metal transporter previously associated with copper response in Mtb (Ward et al., 2008). In this study, we provide the first functional characterization of ctpV, and investigate its relevance to the development of tuberculosis. We show that CtpV is required for copper detoxification in Mtb and likely functions as a copper exporter. CtpV is also required for the full virulence of the bacteria in two animal models of tuberculosis, and its absence has a strong effect on host immune response to Mtb.


Construction of mutant Mtb strains

Previously, the expression of the ctpV gene was determined to be induced by copper ions via the copper-binding transcriptional regulator CsoR (Liu et al., 2007). Furthermore, ctpV was identified as a member of the Mtb transcriptional response to copper at both growth-permissive and toxic physiological levels, with highest induction occurring at toxic copper levels (Ward et al., 2008). Sequence analysis showed that ctpV has strong (~70%) protein-level similarity to previously characterized copper transporters involved in copper export and import in Escherichia coli and Enterococcus hirae, respectively (Ward et al., 2008). Due to its particularly high induction during exposure to toxic levels of copper, ctpV was hypothesized to encode a copper exporter required for detoxification in the presence of elevated copper.

To examine this hypothesis, a knockout mutant of ctpV in the virulent Mtb strain H37Rv was constructed (ΔctpV). A 2.1 kB region of ctpV was replaced by a hygromycin resistance cassette using homologous recombination (Fig. 1A) (Pelicic et al., 1997), and the mutant was confirmed using Southern blots (Fig. 1B) as well as PCR (data not shown). Because ctpV is the third gene in the 4-gene cso operon, ΔctpV was tested for possible polar effects on the downstream gene of unknown function, rv0970. Using reverse-transcriptase PCR, the transcription of rv0970 in the mutant strain was confirmed (Fig. 1C). Additionally, a complemented strain was constructed by cloning the ctpV coding region into an integrative vector (pMV361) containing the constitutive hsp60 promoter (Stover et al., 1991) and transforming this construct into the ΔctpV mutant strain. Integration of ctpV into the ΔctpV genome to construct the complemented strain ΔctpV::ctpV was confirmed with PCR (data not shown), and restored gene expression was confirmed with qRT-PCR (Fig. S1). Notably, the expression of ctpV from the hsp60 promoter in ΔctpV::ctpV versus expression from the CsoR-regulated cso promoter in the wild-type strain (H37Rv) resulted in a different expression profile of ctpV between ΔctpV::ctpV and H37Rv. Specifically, ctpV expression levels were approximately 6 fold higher in ΔctpV::ctpV relative to H37Rv in copper-free media (Fig. S1). Therefore, expression of ctpV in the complemented strain was not only restored, but also enhanced. Finally, a growth experiment in rich liquid media (7H9 with ADC) revealed that ΔctpV and its complement ΔctpV::ctpV had no generalized growth defects relative to H37Rv (Fig. 2A). The ΔctpV mutant and the complemented strain ΔctpV:ctpV were then used to experimentally characterize the role of CtpV in copper response.

Figure 1
Construction of ΔctpV mutant using the M. tuberculosis H37Rv strain
Figure 2Figure 2
Growth experiments of wild-type Mtb (H37Rv), ΔctpV, and ΔctpV::ctpV

CtpV is required for response to copper toxicity in Mtb

To examine the role of CtpV in copper transport, we compared copper sensitivity between ΔctpV and H37Rv. Growth experiments of H37Rv, ΔctpV, and ΔctpV::ctpV were measured in liquid broth cultures of minimal media (Sauton’s) supplemented with defined amounts of copper, using a range of copper concentrations previously determined to be physiologically relevant (Fig. 2B) (Wagner et al., 2005). Comparisons of colony forming units (CFUs) between the three strains revealed that ΔctpV had increased copper sensitivity relative to wild type when grown under toxic copper conditions (500 μM CuCl2). The initial inoculum (~106 cfu ml−1) for ΔctpV reached an unculturable state by 8 days, versus 14 days for H37Rv. Notably, the complemented strain displayed dramatically reduced copper toxicity and was able to withstand 500 μM copper treatment for the duration of the experiment. Presumably, this was the result of its high levels of ctpV expression relative to the wild-type strain prior to copper stress (Fig. S1), supporting the assertion that CtpV functions in copper export.

To further characterize M. tuberculosis response to copper accumulation,, we measured induction of csoR in H37Rv, ΔctpV, and ΔctpV::ctpV after 3 hours of exposure to 500 μM copper. Expression of csoR serves as a sensor of intracellular copper levels due to its autoregulation by CsoR, with increased copper leading to increased gene expression (Liu et al., 2007). Results showed that csoR was approximately twofold upregulated in ΔctpV relative to H37Rv, indicating higher intracellular copper levels (Fig. S3). Conversely, csoR was twofold down regulated in ΔctpV::ctpV relative to H37Rv, indicating lower levels of intracellular copper.

CtpV deletion results in copper accumulation

Both the survival and transcriptional profiles conducted suggested the involvement of CtpV in transporting copper outside the cells. To further address this hypothesis, we directly measured copper accumulation in cells with and without CtpV using neutron activation analysis (NAA), a technique that allows for elemental quantification via the measurement of radioactive emissions following sample irradiation in a nuclear reactor (Versieck, 1994). The threshold of detection for the element Cu using NAA is relatively high, and when used NAA to measure copper levels in M. tuberculosis, copper quantities were too close to the detection threshold to reach statistical significance. Therefore, we used the environmental strain Mycobacterium smegmatis for this experiment, as it has much higher copper tolerance than Mtb (Fig. S4) and the bacteria could therefore be exposed to higher levels of copper prior to irradiation. Briefly, the cso operon was cloned into a plasmid containing an anhydrotetracycline (ATC)-inducible promoter and expression was confirmed using qRT-PCR (data not shown). M. smegmatis containing this plasmid (M. smegmatis::cso) as well as M. smegmatis containing only the empty vector (M. smegmatis + pSE100) were grown in copper-free media in the presence of ATC. Upon reaching OD 1.0, both strains were exposed to 5 mM CuCl2 for 3 hours, at which point the bacteria were pelleted and copper levels quantified using NAA. Results were normalized to bacterial dry mass, and revealed that the strain expressing cso contained significantly lower levels of copper (Fig. 3) in two independent experiments, as determined by a Student’s t-test. The NAA experiment also tested for magnesium levels, which were not found to be different between the two strains (data not shown).

Figure 3
Bulk analysis of copper present in cell pellets of M. smegmatis containing only the empty pSE100 vector as a control, or M. smegmatis expressing the cso operon, as determined by neutron activation analysis. Values are normalized to cell mass and expressed ...

Overall, we have shown that ctpV is transcriptionally induced by high levels of copper (Liu et al., 2007), is predicted to encode a copper transporter based on sequence analysis (Ward et al., 2008), is required for resistance to copper toxicity (Fig. 2B), and that its absence results in higher levels of intracellular copper (Fig. S3, Fig. 3). Therefore, we conclude from these experiments that CtpV is necessary for copper homeostasis in Mtb and likely functions as a copper exporter.

Deletion of ctpV reveals Mtb stress response

Whole-genome microarray analyses were used to investigate what transcriptional changes occur during copper stress when ctpV is deleted from Mtb. Cultures of ΔctpV were exposed to 500 μM CuCl2 for 3 hours and transcript levels of the cells were compared to those of wild-type cultures that had been exposed to the same conditions, as published previously. This experimental setup allowed us to identify only those genes responding to copper stress in a way unique to the ΔctpV strain versus wild type. The 3 hour exposure was chosen to negate effects related to cell death that manifest after a more long-term exposure to 500 μM copper, which is known to be toxic, and was found to produce more consistent transcriptional changes relative to 15 minute and 24 hour exposure times. Using a Nimblegen-based microarray protocol, 98 genes with significantly different expression levels between ΔctpV and H37Rv after exposure to 500 μM CuCl2 were identified (Table S1). To confirm the validity of the microarray data, expression levels of nine of the genes identified in the microarray dataset were tested with qRT-PCR. The changes in gene transcript levels detected by microarray were similar to that determined by qRT-PCR for all 9 genes (Fig. S5).

The 98 genes identified are indicative of copper toxicity (examples in Table 1), which is thought to be the result of many combined effects that high copper levels have on cellular components. We summarized these effects in the following four categories. First, the redox capacity of copper, which can exist in either a Cu1+ or Cu2+ state, causes oxidative stress via the creation of reactive oxygen species within the cell (Pinto et al., 2003). The deletion of ctpV affected transcription of genes previously associated with oxidative stress, such as the regulator-encoding genes furA and sigE, as well as nuoB, predicted to encode an NADH dehydrogenase. In total, 14 of the 98 genes identified have been previously associated with oxidative stress in Mtb (Schnappinger et al., 2003). We also used qRT-PCR to test expression of a recently discovered gene associated with copper stress response, encoding the copper-chelating protein MymT (Gold et al., 2008), that was not present on the microarray slides, and found it to be significantly upregulated in ΔctpV.

Table 1
Main features of copper stress response following exposure of ΔctpV to toxic levels of copper, as measured by microarray analysis.

Second, copper stress causes protein denaturation, particularly via interactions between copper and thiol groups (Agarwal et al., 1989). A number of genes associated with protein synthesis and protein destruction were identified, including five genes predicted to encode ribosomal proteins (rpsR, rplI, rpsQ, rplF, and rpmE), proteasome subunit-encoding prcA, and genes encoding enzymes such as thiosulfate sulfurtransferase (cysA3) and thioredoxin reducatse (trxB).

Third, copper can inappropriately bind to enzymes that require other biometals for activity, resulting in enzyme inactivation (Meharg, 1994; Macomber et al., 2009). Examples of this group include genes for metalloenzymes such as Mg2+-requiring glycosyltransferase (rv1520) and polyketide synthases (papA1, papA3), along with components of the copper-requiring cytochrome c oxidase system (qcrB, ctaC). Additionally, genes associated with iron homeostasis and acquisition were also identified, including upregulation of bfrB, encoding a bacterioferritin, and downregulation of rv0247c, encoding an iron-sulfur protein, which is consistent with previously identified connections between copper homeostasis and iron (Chung et al., 2004; Kershaw et al., 2005; Teitzel et al., 2006).

Fourth, copper causes membrane stress and destabilization (Ohsumi et al., 1988; Avery et al., 1996), and the largest functional group identified (n=16 genes) was that of genes predicted to encode membrane and secreted proteins (Table S1). This includes several mce genes (mce1A, mce1B, mce1C, mce1F) as well as tatA, encoding a component of the Tat secretion system thought to be responsible for transport of the copper detoxification enzyme multicopper oxidase (Graubner et al., 2007) as well as the antigenic proteins Ag85A and Ag85C (Marrichi et al., 2008). Genes for secreted proteins such as mpt64 and esxA were also identified.

Because copper exposure has been shown to cause membrane instability and visible morphological changes (Ohsumi et al., 1988), we performed scanning electron microscopy analysis of the ΔctpV and H37Rv strains under copper-free or high-copper conditions (Fig. S6). Results confirmed that there was no gross morphological difference between the two strains, and also demonstrated that copper stress did not have a visible effect on membrane shape or size in either strain. Despite the lack of a visible difference in cell wall stability, the microarray data points to an effect of the ctpV deletion, and presumably increased intracellular copper, on the membrane and exported protein profile of ΔctpV. Notably, many of the secreted proteins identified are predicted to be immunogenic. Overall, the Mtb genes identified in this study fit into four categories associated with the effects of copper stress, suggesting that the ΔctpV bacteria experienced increased copper stress relative to wild type when exposed to the same level of copper.

Redundancy of CtpV

Interestingly, the genome of Mtb encodes a number of genes with high sequence similarity to ctpV. Specifically, ctpV is classified as encoding a metal translocation P-type ATPase protein, and there are 10 other predicted metal-translocation P-type ATPases in the H37Rv genome with significant sequence similarity to ctpV (Table S2), representing possible functional redundancy for copper export within the Mtb genome.

To precisely investigate whether any of the other P-type ATPases encoded by Mtb might be upregulated in the absence of ctpV, qRT-PCR was used to measure the transcriptional induction of the panel of 11 ATPases present in the Mtb genome. These data revealed that only one other gene encoding a P-type ATPase, ctpG, is induced in the presence of copper (20-fold), and furthermore, that it is particularly induced (60-fold) in the absence of ctpV (Fig. 4). Notably, expression levels of the ctpA and ctpB genes were unchanged, despite the presence of copper transport-associated motifs in their protein sequences. The predicted protein sequence of ctpG contains motifs common to P-type ATPase transporters in general, but lacks the motif associated with copper-specific transporters (Fig. S7), and has been predicted via sequence analysis to belong to a family of Cd, Pb, Zn, and Co transporters (Axelson, April 2005). It has also been shown to be regulated by CmtR, an SmtB-ArsR family cadmium- and lead- responsive repressor (Wang et al., 2005) that was identified via microarrays as responsive to general copper stress as well as the absence of ctpV (this study). Additionally, the lack of copper-induced transcriptional induction of the other predicted transporter-encoding genes, including ctpA and ctpB, does not exclude the possibility that they could function as copper exporters which may be constitutively expressed or regulated in a copper-independent manner. Overall, the high number of predicted metal transporters suggests the evolutionary value of metal response in Mtb, and it seems likely that other encoded genes are functionally redundant with ctpV. Genomic and transcriptional data point to ctpA, ctpB, and ctpG as possible targets of future investigation.

Figure 4
Transcriptional responses to the deletion of ctpV under toxic levels of copper. Transcript levels of all genes within the Mtb genome predicated to encode metal-transporting P-type ATPases at 500 μM vs. 0 μM copper, in both H37Rv and Δ ...

CtpV is required for full virulence in guinea pigs

The ctpV gene is part of a 29-gene genomic island previously shown to be preferentially induced in mice relative to in vitro culture (termed the in vivo expressed genomic island, iVEGI) (Talaat et al., 2004). This suggested that ctpV might play a role specific to the in vivo lifestyle of Mtb. In addition, our experiments showed that CtpV is necessary to maintain copper homeostasis, and data from other groups suggest that copper homeostasis in bacteria may play a role in pathogenesis (Francis et al., 1997; Mitrakul et al., 2004; Schwan et al., 2005), although this had never been tested in Mtb. To investigate the role of ctpV in a host infection model, groups of guinea pigs (n=12) were infected with H37Rv, ΔctpV, or ΔctpV::ctpV via an aerosol infection route. CFUs within lung tissue were determined at 1, 21, and 42 days post-infection, and lung tissues were sectioned and stained for histopathological examination. Colony counts for the ΔctpV mutant were significantly lower than those of H37Rv at 21 days post-infection (p=0.04) (Fig. 5A). However, by 42 days post-infection no colonization defect remained.

Figure 5Figure 5
Guinea pigs infection with M. tuberculosis and its isogenic mutant ΔctpV. A) Bacterial colonization of guinea pig lungs after aerosol infection with either wild-type H37Rv, ΔctpV, or the complemented strain ΔctpV::ctpV. CFUs were ...

The histology staining of infected guinea pig tissue revealed that the tissue damage seen in the guinea pigs infected with H37Rv or ΔctpV::ctpV was more extensive than the tissue damage seen in the guinea pigs infected with ΔctpV. Specifically, larger portions of lung lobes were damaged in the H37Rv-infected guinea pigs compared to those infected with ΔctpV, especially at 42 days post infection (Fig. 5B). Additionally, granulomatous responses were more severe in lungs collected from H37Rv or ΔctpV::ctpV -infected animals compared to those infected with ΔctpV (Fig. 5B). Both the colonization and pathological data in guinea pigs suggested a role played by ctpV in full virulence of Mtb. To further investigate the role of CtpV in virulence, we continued our study of pathogenesis within the murine model of tuberculosis.

CtpV is required for full virulence of Mtb in a murine model

Groups of BALB/c mice (n=30–40) were infected with H37Rv, ΔctpV, or ΔctpV::ctpV using a low-dose aerosolization protocol. Bacterial colony counts from mouse lung tissue over the course of the infection revealed that bacterial survival between the three strains over the course of infection was similar, and differences in colonization levels did not reach statistical significance at any time point (Fig. 6A). Nonetheless, mice infected with ΔctpV lived significantly longer than mice infected with wild type, with 16-week increase in time-to-death versus mice infected with H37Rv (Fig. 6B). In fact, the median survival time for mice infected with H37Rv was 31 weeks, versus 47 weeks for mice infected with ΔctpV and 42 weeks for mice infected with ΔctpV::ctpV. As determined by a log-rank statistical test (Petrie et al., 2006), survival was significantly different between the H37Rv and ΔctpV infection (p-value = 0.002), although survival was also significantly different between the H37Rv and ΔctpV::ctpV infection (p-value = 0.02).

Figure 6Figure 6
Murine infection with ΔctpV. A) Bacterial colonization of mouse lungs after aerosol infection with either wild-type H37Rv, its isogenic mutant ΔctpV, or the complemented strain ΔctpV::ctpV. CFUs were determined via homogenization ...

Despite carrying similar bacterial loads throughout the infection, histology of the infected mouse tissue revealed consistently lower levels of tissue damage in mice infected with ΔctpV versus the wild-type and complemented strains. For example, at 8 weeks post-infection, lung tissue from mice infected with ΔctpV displayed granulomatous inflammation, whereas mice infected with H37Rv displayed massive granulomatous inflammation with more lymphocytic infiltration (Fig. 7A). By 38 weeks post-infection, granulomas became more developed (presence of giant cells) and occupied almost the whole lungs of mice infected with H37Rv, compared to only 50% of tissues of mice infected with the ΔctpV mutant (Fig. 7B). Lesions observed in the complemented strain were very similar to those observed in mice infected with H37Rv strain. Overall, mouse survival and histopathological data indicated the attenuation of the ΔctpV mutant compared to other tested strains.

Figure 7
Pathology of murine infection with H37Rv, ΔctpV, and ΔctpV::ctpV

Immune response to infection is altered by CtpV deletion

Lung pathology in tuberculosis is thought to be caused mainly by the host immune response (Rook et al., 1991). Therefore, to investigate a possible mechanism for the decreased lung pathology of animals infected with ΔctpV relative to H37Rv and ΔctpV:: ctpV, lung sections from the murine infection were stained with an antibody against mouse interferon- γ, a key cytokine known to be highly expressed during tuberculosis infection (Flynn et al., 2001). As expected, no indication of IFN-γ expression was seen at 2 weeks post infection, prior to the start of the adaptive immune response (data not shown). However, by 8 weeks post infection, mice infected with H37Rv show significant IFN-γ expression, localized in areas of lung tissue damage, yet mice infected with ΔctpV showed only small amounts of IFN-γ expression (Fig. 7C). Mice infected with ΔctpV::ctpV showed an intermediate level of IFN-γ expression. Interestingly, even at the 38 week time point where mice infected with ΔctpV display large amounts of tissue damage, there was still little expression of IFN-γ relative to mice infected with H37Rv (Fig. 7D). Overall, our analysis indicates contributions of copper response genes to triggering the host immune responses.


Tuberculosis kills approximately 1.6 million people per year, making Mycobacterium tuberculosis the most deadly bacterial pathogen worldwide (Lopez et al., 2006). Understanding the response of Mtb to its intracellular environment, as well as the host response to Mtb infection, is key to learning to combat this global threat. Interplay of hosts and pathogens in response to biometals is known to play a role in bacterial infections. However, metal ion homeostasis in Mtb, as in most pathogens, has previously been studied mainly in the context of iron deprivation. On the other hand, the role of copper in bacterial infections has not yet been clearly defined. Several studies, including this one, have begun to elucidate a connection between copper response and bacterial pathogenesis. Bacterial infections in humans have long been shown to correlate with increased copper levels in the blood (Beisel et al., 1974; Fleming et al., 1991), and availability of copper ions has previously been correlated with antibacterial abilities of macrophages (Percival, 1998). Copper levels in the phagosome fluctuate via unknown mechanisms after phagocytosis of mycobacteria, as well as after stimulation with IFN-γ (Wagner et al., 2005). Interestingly, hypoxia, a condition found in mycobacterial granulomas, has recently been shown to trigger copper uptake in macrophages through upregulation of the macrophage copper transporter CTR1 (White et al., 2009). In addition, macrophage production of reactive nitrogen intermediates, known to be a major stressor in mycobacterial phagosomes, has been shown to stimulate the release of copper from mycobacterial metallothioneins (Gold et al., 2008). Together, these data suggest that copper toxicity is a form of stress experienced by Mtb and other pathogens that reside within the human macrophage.

So far, little is known about the mechanisms used by Mtb to resist copper toxicity. Experiments reported here suggest that ctpV is involved in the copper efflux response of Mtb to copper stress. Animal infection studies further suggest a relationship between copper response and host immunity and pathogenicity of the bacteria. Further, genomic studies show that ctpV is likely only one component of a larger arsenal of genes involved in copper efflux and detoxification. Interestingly, despite the large number of metal-efflux related genes encoded within the Mtb genome, Mtb is actually much more sensitive to copper than environmental strains such as M. smegmatis. The intracellular concentration of copper is kept so low that even sensitive techniques such as ICP-MS and NAA did not yield consistent copper measurements, requiring the use of a heterologous host for direct copper quantification. We are pursuing the use of a highly sensitive electron-microscopy based technique to measure individual copper atoms within Mtb.

In other pathogenic bacteria such as P. aeruginosa and L. monocytogenes, reduction of copper transport ability has been shown to cause decreased colonization levels within a host model system (Francis et al., 1997; Schwan et al., 2005). However, in Mtb the deletion of ctpV did not lead to a significant drop in colonization levels in a murine model, but rather to an effect on lung pathology of the host. Decreased lung damage was observed in both mice and guinea pigs infected with ΔctpV throughout the course of the infection. Lung damage in tuberculosis infections is hypothesized to be mainly the result of the host immune response, and in our murine experiments this decreased lung pathology correlated with decreased IFN-γ production. IFN-γ expression is associated with a protective response to Mtb in both mice and humans via activation of macrophages (Flynn et al., 2001; Casanova et al., 2002). Its expression by Th1 cells in response to IL-12 stimulation by macrophages is considered a key aspect of the adaptive immune response to Mtb. It is unclear, therefore, whether the decrease in host INF- γ expression is due to a decrease in IFN- γ production by Th1 cells, possibly via an effect of Mtb on IL-12 secretion, or whether the apparent decrease is in fact caused by reduced migration of Th1 cells to the site of infection due to unknown immunomodulatory activities of the CtpV-deficient bacteria.

Unexpectedly, the ΔctpV bacteria did not grow to higher levels or at a faster rate than the wild-type bacteria in mouse lungs, despite the lower levels of IFN- γ, which is known to be protective against Mtb infection. It is possible that the ΔctpV bacteria succumb more easily to other in vivo stimuli (e.g. copper in the phagosome), resulting in apparently similar growth rates with H37Rv. Therefore, ΔctpV might in fact share the colonization defect seen in other pathogens, but the decreased immune response obscures the ability to detect the attenuation. Alternatively, a decrease in IFN- γ production could be compensated for if other mechanisms of macrophage activation are upregulated, resulting in similar rates of bacterial growth with different rates of host tissue damage. Analyzing the full array of cytokines produced during ΔctpV or H37Rv infection, including IL-12 and TNF-α, the cytokine thought to be largely responsible for host tissue damage, will help elucidate the reason for the differences in pathology.

The basis for the different host response to ΔctpV has yet to be determined. It is possible that the ΔctpV strain doesn’t express, or expresses at altered levels, proteins that are normally immunomodulatory. This could result either from immunogenicity of CtpV itself, and/or it could stem from the difference in expression profiles between the two strains, including altered levels of secreted proteins known to be immunogenic. For example, a large number of genes (n=98) were affected by disruption of ctpV following exposure to toxic levels of copper. In addition, the host response could be affected by the levels of copper in the phagosome. Because mycobacterial intracellular copper concentration is higher in ΔctpV, it is possible that extracellular (phagosomal) levels of copper are lower, which could affect overall immune response. Overall, our data combined with that of previous studies suggest that copper homeostasis mechanisms may play a role in the outcome of bacterial infections, but the exact role for copper ions as an antibacterial response mechanism and/or immunomodulatory molecule has yet to be determined and warrants further investigation.

Experimental procedures

Mtb strains and mutant construction

Experiments were performed with Mtb strain H37Rv, its isogenic mutant ΔctpV, or the complemented strain ΔctpV::ctpV. To construct ΔctpV, 800 basepair fragments of both the upstream and downstream portion of the gene were amplified by PCR (primers listed in Table S3) and cloned into the pGEM-T Easy vector (Promega, Madison, WI). The fragments were digested with their flanking restriction enzyme sites (AflII/XbaI and HindIII/SpeI for upstream and downstream portions, respectively) and ligated into pYUB854 (Bardarov et al., 2002). After digestion by NotI and SpeI (Promega), the linearized vector was ligated into pML19, a derivative of pPR27 (Pelicic et al., 1997) where a kanamycin resistance cassette has been inserted into the PstI site. The resulting vector was named pML21. This vector was electroporated into electrocompetent Mtb using a Gene Pulser II machine (BioRad, Hercules, CA), and cells were plated onto Middlebrook 7H10 supplemented with 10% albumin-dextrose-catalase (ADC) and 50 μg ml−1 hygromycin (Invitrogen, Carlsbad, CA). After one month of growth at 32°C, transformants were grown for two weeks with shaking 32°C in Middlebrook 7H9 supplemented with 10% ADC and 50 μg ml−1 hygromycin. These cultures were plated onto Middlebrook 7H10 supplemented with 10% ADC, 2% sucrose, and 50 μg ml−1 hygromycin and incubated at 39°C for three weeks. The genomic incorporation of the plasmid was confirmed via the inability of colonies to grow on Middlebrook 7H10 supplemented with 25 μg ml−1 kanamycin.

The transformant used for experiments, termed ΔctpV, was confirmed via negative PCR for the ctpV coding region and positive PCR for the hygromycin resistance cassette with primers listed in Table S3. Additionally, Southern blot was performed on ΔctpV and wild type genomic DNA (5 μg) digested with BamHI (Promega), using probes for the remaining coding region of ctpV or the hygromycin resistance cassette. For Southerns, the Promega Prime-A-Gene kit was used as directed by the manufacturer.

For complementation of ΔctpV, the ctpV coding region was amplified and cloned into the pGEM-T easy vector for sequencing. The pGEM vector was then digested with EcoRI and HindIII (Promega), and the fragment was ligated into pMV361, which contains a kanamycin resistance cassette (Stover et al., 1991). The vector was sequenced, and then electroporated into electrocompetent ΔctpV cells and plated on 7H10 supplemented with 10%ADC with 50 μg ml−1 hygromycin and 25 μg ml−1 kanamycin. The complemented strain was confirmed using a forward primer within the pMV361 vector (hsp60) and a reverse primer within the ctpV coding region.

M. smegmatis strains and experiments

To create M. smegmatis::cso, the cso operon from Mtb was amplified using Herculase Taq DNA polymerase (Stratagene, Santa Clara, CA) and primers listed in Table S3. The fragment was digested with EcoRI (Promega) and cloned into plasmid pUC18. After verification of the fragment via sequencing, the plasmid was digested by EcoRI, blunt-ended with Klenow, and cloned into a pSE100 vector that had previously been double digested with EcoRI and HindIII and blunt-ended with Klenow. M. smegmatis mc2155 was electroporated with an integrative vector containing tetR under an intermediate promoter, and selected for on 7H10 plates containing kanamycin. The pSE100::cso vector was electroporated into this recombinant M. smegmatis strain. For copper exposure experiments, M. smegmatis::cso or M. smegmatis electroporated with an empty pSE100 vector were grown in 7H9+ADC to OD 1.0, pelleted and washed twice in Sauton’s, and then used to inoculate 50 mL cultures of Sauton’s to OD 0.1, with anhydrotetracycline (ATC) added to cultures of both strains to 40 ng mL−1. These cultures were allowed to grow to OD 1.0 prior to addition of CuCl2 to 5 mM. After 3 hours of exposure, cultures were pelleted and washed twice in Sauton’s, then dried at 100°C prior to submitting for NAA analysis.

Neutron activation analysis experiments

M. smegmatis culture pellets were irradiated and analyzed at the University of Wisconsin Nuclear Reactor (UWNR), a 1.0 MW TRIGA reactor. The detector was an Ortec High Purity Germanium detector (ORTEC, Oak Ridge, TN). Samples were irradiated with a thermal neutron flux of approximately 5E12 nv (neutrons per cm^2 per sec). Irradiation time was 10.6 seconds. After 3 minutes decay time, each vial was counted for 300 seconds, and analysis was performed by the UWNR compiled program NAACalc version 1.41.

Growth experiments

Growth experiments in rich media used Middlebrook 7H9 liquid media (Remel, Lenexa, KS), prepared as described by the manufacturer and supplemented with 10% ADC and 0.05% Tween. Minimal media growth experiments were performed in Sauton’s minimal media (which is free of copper) (Parish et al., 1998), supplemented with 0.05% Tween, prepared using water treated with Chelex (Sigma-Aldrich, St. Louis, MO). Glassware was acid-washed (1N nitric acid) to maintain metal-free conditions. Cultures were seeded to OD600 0.1 with bacterial stock washed 2x in Sauton’s, and allowed to grow for 14 days at 37°C with shaking. Colony forming units (CFUs) were determined by plating on Middlebrook 7H10 + 10% ADC solid media, with 50 μg ml−1 hygromycin added in the case of the mutant and complemented strains.

Scanning electron microscopy

After growth to OD600 0.6 in copper-free media, cultures of H37Rv, ΔctpV, or ΔctpV::ctpV were exposed to 500 μM CuCl2 for 3 hours in a 24-well tissue culture plate containing 12 mm glass coverslips. Coverslips were processed for SEM imaging as previously described (Wu et al., 2009). Briefly, coverslips were rinsed with Hanks’ Balanced Salt Solution (Lonza Walkersville, Inc., Walkersville, MD) and incubated overnight in a solution of 4% paraformaldehyde and 2.5% gluteraldehyde. After sequential alcohol dehydration, coverslips were dried in a Samdri 780-A critical point dryer (Tousimis, Rockville, MD) and splatter-coated with gold-palladium prior to imaging with a Hitachi S-570 scanning electron microscope. Bacillus size was measured using Image Pro Plus (Media Cybernetics, Inc., Bethesda, MD) and at least two images from each condition were analyzed.

Mouse infections

BALB/c mice (Harlan Laboratories, Inc., Indianapolis, IN) were infected in a Glas-Col chamber (Glas-Col, LLC, Terra Haute, IN) loaded with 10 ml of either H37Rv, ΔctpV or ΔctpV::ctpV bacteria at OD600 0.30. Infectious dose of approximately 300 CFU/animal was confirmed via a 1-day timepoint. CFUs were determined by homogenizing lung tissue in PBS buffer and plating on Middlebrook 7H10+10%ADC, followed by incubation at 37°C for one month. For the survival curve, animals were monitored daily by animal care staff not associated with the study. As specified in our animal protocol, mice were sacrificed after being identified by our animal care staff as morbidly ill, using criteria such as haunched posture, extreme weight loss, and slow or pained movements. Sections of lung, liver, and spleen tissue were taken and incubated in formalin prior to sectioning and staining with hematoxylin and eosin (H&E) and acid-fast staining (AFS). Histopathology slides were examined and scored by a pathologist not associated with the study. For immunohistochemistry, the primary antibody, rabbit anti-interferon gamma (Invitrogen), was diluted 1:1000 in Van Gogh Yellow antibody diluent (Biocare Medical, Concord, CA) and incubated for one hour. Negative control slides received only diluent in lieu of antibody. Primary antibody was detected using biotinylated goat anti-rabbit IgG secondary antibody (Biocare Medical, Concord, CA) and Streptavidin-horseradish peroxidase (Biocare Medical). Staining was visualized with DAB+ (diaminobenzidine) (Dako, Glostrup, Denmark) and counterstained with CAT hematoxylin (Biocare Medical) mixed 1:1 with distilled water.

Guinea pig infections

Female Hartley guinea pigs (250–300g, Charles River) were used for these studies. All guinea pigs were housed in a Biosafety Level-3 (BSL-3), pathogen-free animal facility and were fed water and chow ad libitum. The animals were maintained and all procedures performed according to protocols approved by the Institutional Animal Care and Use Committee at the Johns Hopkins University School of Medicine. Separate groups of guinea pigs were aerosol-infected with wild-type M. tuberculosis H37Rv, ΔctpV, or ΔctpV::ctpV using a Madison aerosol exposure chamber (College of Engineering Shops, University of Wisconsin, Madison, WI) calibrated to deliver 50–100 bacilli per animal lung. Four guinea pigs from each group were sacrificed on Days 1, 21, 42 after aerosol infection.

At the designated time points, lungs from sacrificed animals were removed aseptically, weighed, and assessed for gross pathology. The lungs were then homogenized for colony-forming unit (CFU) enumeration and processed for histological examination. Guinea pig lungs were homogenized in 10–15 ml PBS using a Kinematica Polytron Homogenizer with a 12-mm generator within a BSL-III Glovebox Cabinet (Germfree, Ormond Beach, FL). Serial tenfold dilutions of lung homogenates were plated on 7H11 selective agar (BBL) except for day 1 lung samples, which were plated neat, and plates were incubated at 37° C. The number of viable bacilli was enumerated after 4 weeks of incubation. At various time points after infection, lung samples were fixed in 10% PBS-buffered formalin, embedded in paraffin, and processed for histology. Sections were stained with hematoxylin and eosin, and Ziehl–Neelsen for microscopic evaluation.

Microarray analysis

Cultures of ΔctpV were inoculated to OD600 0.1 in Sauton’s media and allowed to grow shaking at 37°C to OD600 0.6. At this point, they were supplemented with 500 μM CuCl2 and incubated for three more hours prior to spinning down the cultures and freezing immediately at −80°C. Microarray data was obtained as described previously (Ward et al., 2008). Briefly, RNA was extracted using a Trizol-based method (Invitrogen), and treated with DNAse I (Ambion, Austin, TX) to remove contaminating DNA. cDNA was synthesized from 1 μg total RNA using an Invitrogen SuperScript ds-cDNA synthesis kit in the presence of 250 ng genome-directed primers (Talaat et al., 2002). cDNA clean up, Cy3 labeling, hybridizations, and washing steps were performed using the NimbleGen gene expression analysis protocol (Roche Nimblegen, Madison, WI). Microarray chips were purchased from NimbleGen Systems, Inc., and they contained 19 probes (60-mers) for each of the 3,989 open reading frames identified in the genome of Mtb H37Rv (Camus et al., 2002), with five replicates of the genome printed on each slide (total of 95 probes/gene). Slides were scanned using an Axon GenePix 4000B scanner (Molecular Devices Corporation, Sunnyvale, CA), and fluorescence intensity levels normalized to 1000. Significantly changed genes between H37Rv and ΔctpV were determined using the EBArrays package in R [http://www.bioconductor.org]. A cutoff value of 0.50 for the probability of differential expression, determined using a lognormal-normal (LNN) model, was used to determine statistically differentially expressed genes (Kendziorski et al., 2003). Full data are deposited in accordance with MIAME standards at Array Express [http://www.ebi.ac.uk/microarray-as/ae] accession code E-MEXP-2773.

Quantitative, real-time PCR

qRT-PCR was performed using a SYBR green-based protocol. cDNA was synthesized from DNAse-treated RNA, obtained as described above, using SuperScript III (Invitrogen) as directed by the manufacturer, in the presence of 250 ng mycobacterial genome-directed primers (Talaat et al., 2002). A 100 ng cDNA was used as template in a reaction with iTaq SYBR green Supermix with ROX passive dye (BioRad, Hercules, CA) in the presence of gene-specific primers (Table S3) at a concentration of 200 nM. Cycle conditions were 50°C for 2 min, 95°C for 3 min, and 40 cycles of 95°C for 15 s and 60°C for 30 s. Reactions were performed in triplicate on an AB7300 machine (Applied Biosystems, Foster City, CA) with fluorescence read at the 60°C step. Threshold cycle values were normalized to 16S rRNA levels.

Statistical analysis

The Student’s t-test was used to determine whether data sets were significantly different, with a cutoff value of p<0.05. The log rank test was used to determine whether survival of infected animals were significantly different, with a cutoff value of p<0.05.

Supplementary Material

Supp Figure S01-07 & Table S01-03


The authors thank Ralph Albrecht and Joseph Heintz for SEM analysis, and Thomas Pier and the University of Wisconsin TRIP lab for immunohistochemistry. We thank Kevin Austin and the University of Wisconsin Nuclear Reactor Laboratory for NAA analyses. We thank the TARGET Tuberculosis Animal Research and Gene Evaluation Taskforce at Johns Hopkins University, including Paul Converse, Petros Karakousis, and Michael Pinn. We also thank Joseph Dillard, Sarah Marcus, and Chia-wei Wu for reading the manuscript, as well as our anonymous reviewers for their helpful feedback throughout the submission process. This work was supported by grant NIH-R21AI081120 to AMT and by NIH training grant T32GM007215 to SKW.


  • Agarwal K, Sharma A, Talukder G. Effects of copper on mammalian cell components. Chem Biol Interact. 1989;69:1–16. [PubMed]
  • Avery S, Howlett N, Radice S. Copper toxicity towards Saccharomyces cerevisiae: dependence on plasma membrane fatty acid composition. Appl Environ Microbiol. 1996;62:3960–3966. [PMC free article] [PubMed]
  • Axelson KB. A P-type ATPase database. Apr, 2005.
  • Bardarov S, Bardarov SJ, Jr, Pavelka MJ, Jr, Sambandamurthy V, Larsen M, Tufariello J, et al. Specialized transduction: an efficient method for generating marked and unmarked targeted gene disruptions in Mycobacterium tuberculosis, M. bovis BCG and M. smegmatis. Microbiology. 2002;148:3007–3017. [PubMed]
  • Beisel WR, Pekarek RS, Wannemacher RW. The impact of infectious disease on trace-element metabolism of the host. In: Hoekstra WG, Suttie JW, Ganther HE, Mertz W, editors. Trace Element Metabolism in Animals-2. Baltimore: University Park Press; 1974. p. 217.
  • Camus J, Pryor M, Médigue C, Cole S. Re-annotation of the genome sequence of Mycobacterium tuberculosis H37Rv. Microbiology. 2002;148:2967–2973. [PubMed]
  • Casanova J, Abel L. Genetic dissection of immunity to mycobacteria: the human model. Annu Rev Immunol. 2002;20:581–620. [PubMed]
  • Chung J, Haile D, Wessling-Resnick M. Copper-induced ferroportin-1 expression in J774 macrophages is associated with increased iron efflux. Proc Natl Acad Sci U S A. 2004;101:2700–2705. [PMC free article] [PubMed]
  • De Voss J, Rutter K, Schroeder B, Su H, Zhu Y, Barry Cr. The salicylate-derived mycobactin siderophores of Mycobacterium tuberculosis are essential for growth in macrophages. Proc Natl Acad Sci U S A. 2000;97:1252–1257. [PMC free article] [PubMed]
  • Fleming R, Whitman I, Gitlin J. Induction of ceruloplasmin gene expression in rat lung during inflammation and hyperoxia. Am J Physiol. 1991;260:L68–74. [PubMed]
  • Flynn J, Chan J. Immunology of tuberculosis. Annu Rev Immunol. 2001;19:93–129. [PubMed]
  • Francis M, Thomas C. Mutants in the CtpA copper transporting P-type ATPase reduce virulence of Listeria monocytogenes. Microb Pathog. 1997;22:67–78. [PubMed]
  • Gold B, Deng H, Bryk R, Vargas D, Eliezer D, Roberts J, et al. Identification of a copper-binding metallothionein in pathogenic mycobacteria. Nat Chem Biol. 2008;4:609–616. [PMC free article] [PubMed]
  • Graubner W, Schierhorn A, Brüser T. DnaK plays a pivotal role in Tat targeting of CueO and functions beside SlyD as a general Tat signal binding chaperone. J Biol Chem. 2007;282:7116–7124. [PubMed]
  • Kendziorski C, Newton M, Lan H, Gould M. On parametric empirical Bayes methods for comparing multiple groups using replicated gene expression profiles. Stat Med. 2003;22:3899–3914. [PubMed]
  • Kershaw C, Brown N, Constantinidou C, Patel M, Hobman J. The expression profile of Escherichia coli K-12 in response to minimal, optimal and excess copper concentrations. Microbiology. 2005;151:1187–1198. [PubMed]
  • Liu T, Ramesh A, Ma Z, Ward S, Zhang L, George G, et al. CsoR is a novel Mycobacterium tuberculosis copper-sensing transcriptional regulator. Nat Chem Biol. 2007;3:60–68. [PubMed]
  • Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJL. Global burden of disease and risk factors. New York: Oxford University Press, The World Bank; 2006.
  • Macomber L, Imlay J. The iron-sulfur clusters of dehydratases are primary intracellular targets of copper toxicity. Proc Natl Acad Sci U S A. 2009;106:8344–8349. [PMC free article] [PubMed]
  • MacPherson I, Murphy M. Type-2 copper-containing enzymes. Cell Mol Life Sci. 2007;64:2887–2899. [PubMed]
  • Marrichi M, Camacho L, Russell D, DeLisa M. Genetic toggling of alkaline phosphatase folding reveals signal peptides for all major modes of transport across the inner membrane of bacteria. J Biol Chem. 2008;283:35223–35235. [PMC free article] [PubMed]
  • Meharg AA. Integrated tolerance mechanisms: constitutive and adaptive plant responses to elevated metal concentrations in the environment. Plant, Cell and Environment. 1994;17:989–993.
  • Mitrakul K, Loo C, Hughes C, Ganeshkumar N. Role of a Streptococcus gordonii copper-transport operon, copYAZ, in biofilm detachment. Oral Microbiol Immunol. 2004;19:395–402. [PubMed]
  • Ohsumi Y, Kitamoto K, Anraku Y. Changes induced in the permeability barrier of the yeast plasma membrane by cupric ion. J Bacteriol. 1988;170:2676–2682. [PMC free article] [PubMed]
  • Parish T, Stoker NG. Mycobacteria protocols. Totowa, N.J: Humana Press; 1998.
  • Pelicic V, Jackson M, Reyrat J, Jacobs WJ, Gicquel B, Guilhot C. Efficient allelic exchange and transposon mutagenesis in Mycobacterium tuberculosis. Proc Natl Acad Sci U S A. 1997;94:10955–10960. [PMC free article] [PubMed]
  • Percival S. Copper and immunity. Am J Clin Nutr. 1998;67:1064S–1068S. [PubMed]
  • Petrie A, Watson PF. Statistics for veterinary and animal science. Oxford; Ames, Iowa: Blackwell Pub; 2006.
  • Pinto E, Sigaud-kutner TCS, Leitao MAS, Okamoto OK, Morse D, Colepicolo P. Heavy-metal induced oxidative stress in algae. Journal of Phycology. 2003;39:1008–1018.
  • Rook G, al Attiyah R, Filley E. New insights into the immunopathology of tuberculosis. Pathobiology. 1991;59:148–152. [PubMed]
  • Schaible U, Kaufmann S. Iron and microbial infection. Nat Rev Microbiol. 2004;2:946–953. [PubMed]
  • Schnappinger D, Ehrt S, Voskuil M, Liu Y, Mangan J, Monahan I, et al. Transcriptional Adaptation of Mycobacterium tuberculosis within Macrophages: Insights into the Phagosomal Environment. J Exp Med. 2003;198:693–704. [PMC free article] [PubMed]
  • Schwan W, Warrener P, Keunz E, Stover C, Folger K. Mutations in the cueA gene encoding a copper homeostasis P-type ATPase reduce the pathogenicity of Pseudomonas aeruginosa in mice. Int J Med Microbiol. 2005;295:237–242. [PubMed]
  • Smith I. Mycobacterium tuberculosis pathogenesis and molecular determinants of virulence. Clin Microbiol Rev. 2003;16:463–496. [PMC free article] [PubMed]
  • Stover C, de la Cruz V, Fuerst T, Burlein J, Benson L, Bennett L, et al. New use of BCG for recombinant vaccines. Nature. 1991;351:456–460. [PubMed]
  • Talaat A, Howard S, Hale Wt, Lyons R, Garner H, Johnston S. Genomic DNA standards for gene expression profiling in Mycobacterium tuberculosis. Nucleic Acids Res. 2002;30:e104. [PMC free article] [PubMed]
  • Talaat A, Lyons R, Howard S, Johnston S. The temporal expression profile of Mycobacterium tuberculosis infection in mice. Proc Natl Acad Sci U S A. 2004;101:4602–4607. [PMC free article] [PubMed]
  • Teitzel G, Geddie A, De Long S, Kirisits M, Whiteley M, Parsek M. Survival and growth in the presence of elevated copper: transcriptional profiling of copper-stressed Pseudomonas aeruginosa. J Bacteriol. 2006;188:7242–7256. [PMC free article] [PubMed]
  • Teixeira E, Franco de Oliveira J, Marques Novo M, Bertolini M. The copper resistance operon copAB from Xanthomonas axonopodis pathovar citri: gene inactivation results in copper sensitivity. Microbiology. 2008;154:402–412. [PubMed]
  • Versieck J. Neutron activation analysis for the determination of trace elements in biological materials. Biol Trace Elem Res. 1994;43–45:407–413. [PubMed]
  • Wagner D, Maser J, Lai B, Cai Z, Barry Cr, Höner Zu Bentrup K, et al. Elemental analysis of Mycobacterium avium-, Mycobacterium tuberculosis-, and Mycobacterium smegmatis-containing phagosomes indicates pathogen-induced microenvironments within the host cell’s endosomal system. J Immunol. 2005;174:1491–1500. [PubMed]
  • Wang Y, Hemmingsen L, Giedroc DP. Structural and functional characterization of Mycobacterium tuberculosis CmtR, a PbII/CdII-sensing SmtB/ArsR metalloregulatory repressor. Biochemistry. 2005;44:8976–8988. [PubMed]
  • Ward S, Hoye E, Talaat A. The global responses of Mycobacterium tuberculosis to physiological levels of copper. J Bacteriol 2008 [PMC free article] [PubMed]
  • White C, Kambe T, Fulcher Y, Sachdev S, Bush A, Fritsche K, et al. Copper transport into the secretory pathway is regulated by oxygen in macrophages. J Cell Sci. 2009;122:1315–1321. [PMC free article] [PubMed]
  • Wu CW, Schmoller SK, Bannantine JP, Eckstein TM, Inamine JM, Livesey M, et al. A novel cell wall lipopeptide is important for biofilm formation and pathogenicity of Mycobacterium avium subspecies paratuberculosis. Microbial Pathogenesis. 2009;46:222–230. [PMC free article] [PubMed]
  • Zhang XX, Rainey PB. The Role of a P1-Type ATPase from Pseudomonas fluorescens SBW25 in Copper Homeostasis and Plant Colonization. Molecular Plant-Microbe Interactions. 2007;20:581–588. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...