Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2007 Nov 2;282(44):32298-310.
doi: 10.1074/jbc.M703451200. Epub 2007 Aug 28.

Mycobacterial cells have dual nickel-cobalt sensors: sequence relationships and metal sites of metal-responsive repressors are not congruent

Affiliations

Mycobacterial cells have dual nickel-cobalt sensors: sequence relationships and metal sites of metal-responsive repressors are not congruent

Duncan R Campbell et al. J Biol Chem. .

Abstract

A novel ArsR-SmtB family transcriptional repressor, KmtR, has been characterized from mycobacteria. Mutants of Mycobacterium tuberculosis lacking kmtR show elevated expression of Rv2025c encoding a deduced CDF-family metal exporter. KmtR-dependent repression of the cdf and kmtR operator-promoters was alleviated by nickel and cobalt in minimal medium. Electrophoretic mobility shift assays and fluorescence anisotropy show binding of purified KmtR to nucleotide sequences containing a region of dyad symmetry from the cdf and kmtR operator-promoters. Incubation of KmtR with cobalt inhibits DNA complex assembly and metal-protein binding was confirmed. KmtR is the second, to NmtR, characterized ArsR-SmtB sensor of nickel and cobalt from M. tuberculosis suggesting special significance for these ions in this pathogen. KmtR-dependent expression is elevated in complete medium with no increase in response to metals, whereas NmtR retains a response to nickel and cobalt under these conditions. KmtR has tighter affinities for nickel and cobalt than NmtR consistent with basal levels of these metals being sensed by KmtR but not NmtR in complete medium. More than a thousand genes encoding ArsR-SmtB-related proteins are listed in databases. KmtR has none of the previously defined metal-sensing sites. Substitution of His88, Glu101, His102, His110, or His111 with Gln generated KmtR variants that repress the cdf and kmtR operator-promoters even in elevated nickel and cobalt, revealing a new sensory site. Importantly, ArsR-SmtB sequence groupings do not correspond with the different sensory motifs revealing that only the latter should be used to predict metal sensing.

PubMed Disclaimer

Figures

FIGURE 1
FIGURE 1. Tree diagram constructed from the alignment of 554 sequences derived from the HTH_5 family of the Pfam data base
The tree diagram shows eight major groups. The main metal binding motifs occurring in proteins within each group are indicated: CmtR-like sequences possess the CXXXC metal-binding pattern in the α4 helix and potential ligands in the carboxyl-terminal region; CzrA/SmtB-like and NmtR-like possess the pattern DXHX(10)HXX(E/H) in the α5 helix, with the latter also possessing carboxyl-terminal ligands; ArsR-like possess either one or more of (i) CXCXXC, (ii) CXC, or (iii) CXXD in the α3 helix; CadC-like and ZiaR-like possess either one or more of i, ii, and iii in addition to potential ligands in the amino-terminal region, with ZiaR-like also possessing the DXHX(10)HXX(E/H) pattern in α5; the unknown lack these patterns. For each motif, the main bacterial phyla are shown in parentheses (a series of periods indicates sequences are also present from other phyla), and sequences with the motif may be from more than one branch (sub-group).
FIGURE 2
FIGURE 2. KmtR binds to the kmtR and cdf operator promoter regions
A, physical map of the kmtR gene region and position in the M. tuberculosis genome (vertical lines). B, gel-retardation assays used (from left to right) 0, 0.25, 0.5, and 1 μm KmtR with a 310-bp DNA fragment containing PkmtR (including the first 18 codons of kmtR), a 292-bp DNA fragment containing Pcdf (including the cdf start codon), or a 292-bp DNA fragment containing PRv0286 (including the first 5 codons of Rv0826) as probe and a 136-bp fragment of nonspecific competitor DNA (N). The latter contained identical sequences to the probes but lacks the operator-promoter regions. C, gel-retardation assays used (from left to right) 0, 0.5, 1, and 2 μm KmtR with probes T1 (241 bp), T2 (198 bp), or T3 (190 bp) and nonspecific competitor DNA. Predominant complexes (C) are indicated. Diagrammatic representations of the Pcdf sequences within T1, T2, and T3 are shown with the position of a degenerate 13-4-13 hyphenated inverted repeat. The sequence of the Pcdf fragment within T2 is shown in full, with arrows indicating the inverted repeat, with a similar repeat identified within PkmtR (conserved nucleotides are shown in bold).
FIGURE 3
FIGURE 3. KmtR binding to PkmtR and Pcdf as measured by fluorescence anisotropy
6-Hexachloroflourescein-labeled PkmtR 32-mer (A), Pcdf 34-mer (B), or PRv0286 32-mer (C) at 5 nm were titrated with KmtR and anisotropy, robs, measured. The curves were calculated based on a 1:1 model using SigmaPlot, KKmtR = 0.24 μm and 0.12 μm for PkmtR and Pcdf, respectively.
FIGURE 4
FIGURE 4. No metal enhances expression from a KmtR-regulated promoter in a mycobacterium grown in Middlebrook 7H9 medium
A and B, β-galactosidase activity was measured in M. smegmatis mc2155 containing kmtR and Pcdf fused to lacZ (in pJEM15kmtR-Pcdf) grown with no metal supplement and maximum permissive concentrations of Zn(II) (70 μm), Co(II) (15 μm), Ni(II) (35 μm), Cd(II) (75 nm), Cu(II) (50 μm), Pb(II) (10 μm), Ag(I) (0.75 μm), Bi(III) (0.75 μm), or Mn(II) (0.75 μm) (A), or up to inhibitory [Zn(II)] (B). Inset, growth (A595) of cultures against added [Zn(II)]. C, microtiter plate assays of Zn(II)-PAR formation (400 μm PAR), measured at 492 nm and zeroed against apo-PAR, as a function of [Zn(II)] in the absence (closed symbols) or presence (open symbols) of KmtR (2 μm).
FIGURE 5
FIGURE 5. MtSmtB binds its own operator-promoter and responds to zinc in cells grown in Chelex-treated Sauton medium
A, MtSmtB (50 μm) was incubated overnight with either 1 mm EDTA (left) or 20 μm Zn(II) (right) and increasing concentrations added to 6-hexachloroflourescein-labeled PMtSmtB 30-mer (5 nm) in the presence of 1 mm EDTA or 35 μm ZnCl2 (final [Zn(II)] = 35–35.84 μm), respectively, under anaerobic conditions and anisotropy, robs, measured. The curve represents a 1:1 binding model, KMtSmtB = 0.81 μm. B, microtiter plate assays of Zn(II)-PAR formation (400 μm PAR), measured at 492 nm and zeroed against apo-PAR, as a function of [Zn(II)] in the absence (closed symbols) or presence (open symbols) of MtSmtB (2 μm). C and D, β-galactosidase activity measured in M. smegmatis mc2155 containing MtsmtB-regulated lacZ grown up to inhibitory [Zn(II)] in Middlebrook 7H9 medium (C) or metal-depleted Sauton medium (D). Inset, β-galactosidase activity in cells grown with no metal supplement or with maximum permissive concentrations of Zn(II) (70 μm), Co(II) (15 μm), or Ni(II) (30 μm).
FIGURE 6
FIGURE 6. KmtR responds to cobalt and nickel in a mycobacterium grown in Chelex-treated Sauton medium
A–C, β-galactosidase activity measured in M. smegmatis mc2155 containing kmtR and Pcdf fused to lacZ (in pJEM15kmtR-Pcdf) or the stop codon derivative (M24*) following growth in medium with no metal supplement or maximum permissive concentrations of Zn(II), Co(II), Ni(II), Cd(II), Cu(II), Pb(II), Ag(I), Bi(III), Mn(II), or Fe(III) (A and B), or up to inhibitory concentrations of Co(II) or Ni(II) (C). D, β-galactosidase activity measured in M. bovis BCG containing pJEM15kmtR-Pcdf grown with no metal supplement or with maximum permissive concentrations of Co(II) or Ni(II), or containing the stop codon derivative (M24*) and grown with no metal supplement. E–G, β-galactosidase activity measured in M. smegmatis mc2155 containing kmtR (in pJEM15kmtR) (E), kmtR and Pcdf separated by tT4 (in pJEM15kmtR-tT4-Pcdf) (F), or kmtR and PRv0286 separated by tT4 (in pJEM15kmtR-tT4-PRv0286) (G), fused to lacZ, grown with no metal supplement or with maximum permissive concentrations of Co(II) or Ni(II), or in cells containing the stop codon derivatives of these constructs (M24*) and grown with no metal supplement.
FIGURE 7
FIGURE 7. Cobalt destabilizes KmtR·DNA complexes
A, KmtR (13.3 μm) was incubated overnight with either 1 mm EDTA (left) or 35 μm Co(II) (right), increasing concentrations were added to 6-hexachloroflourescein-labeled PkmtR 32-mer (5 nm) in the presence of 1 mm EDTA or 35 μm CoCl2 (final [Co(II)] = 35–42 μm), respectively, under anaerobic conditions, and anisotropy, robs, was measured. In the presence of EDTA it is apparent that these data do not fit a 1:1 binding model, which is consistent with a large Δrobs implying binding of multiple dimers; the curve represents a sigmoidal fit, KKmtR = 0.6 μm. B, microtiter plate assays of Ni(II)-PAR formation (400 μm PAR), measured at 492 nm and zeroed against apo-PAR, as a function of [Ni(II)] in the absence (closed symbols) or presence (open symbols) of KmtR (2 μm).
FIGURE 8
FIGURE 8. Comparison of KmtR and NmtR metal responsiveness
β-Galactosidase activity in M. smegmatis mc2155 cells containing KmtR-regulated (upper) or NmtR-regulated (lower) lacZ grown up to inhibitory [Ni(II)] or [Co(II)].
FIGURE 9
FIGURE 9. KmtR has tighter affinity for nickel and cobalt than NmtR
A, Ni(II) binding isotherm for KmtR (5 μm) monitored as tryptophan fluorescence (λex = 280 nm, λem = 440 nm). KNi is too tight to measure under these conditions. B, competition between KmtR (5 μm) and NmtR (5 μm) for Ni(II) (4 μm) monitored as tryptophan emission spectra following excitation at 295 nm. Only KmtR contains a tryptophan residue. Ni(II)-dependent difference spectra are shown for KmtR (5 μm) alone (open circles) and KmtR in the presence of equimolar NmtR (closed circles). The lower curve (triangles) shows the negligible Ni(II)-dependent difference spectrum for tyrosine residues of NmtR (5 μm) alone at this excitation wavelength. C, Co(II) binding isotherm for KmtR monitored as tryptophan fluorescence (λex = 280 nm, λem = 440 nm). The curve represents a 1:1 ligand binding model, which under these conditions (including 1 mm DTT) gives KKmtR = 6.9 μm. KNmtR under these conditions is estimated to be weaker than KKmtR and weaker than we reported previously in the absence of DTT (13). D, competition between KmtR (5 μm) and NmtR (5 μm) for Co(II) (4 μm) monitored as tryptophan emission spectra following excitation at 295 nm. Co(II)-dependent difference spectra are shown for KmtR (5 μm) alone (open circles) and KmtR (5 μm) in the presence of equimolar NmtR (closed circles). The lower curve (triangles) shows the negligible Co(II)-dependent difference spectrum for tyrosine residues of NmtR (5 μm) alone at this excitation wavelength. Conditions are 20 mm HEPES, 1 mm DTT, 50 mm NaCl (pH 7.5), 22 °C.
FIGURE 10
FIGURE 10. Metal sensing site of KmtR
β-Galactosidase activity was measured in M. smegmatis containing wild-type KmtR (WT) and various derivatives with indicated codon substitutions. Cells were grown in Chelex-treated Sauton medium with no metal supplement (black) or maximum permissive [Ni(II)] (gray) or [Co(II)] (white).
FIGURE 11
FIGURE 11. Sensory sites in ArsR-SmtB family members
A, alignment of representative sequences for the different metal-sensing sites, involving ligands at an α5 site (blue) in cyanobacterial SmtB and CzrA (24-26); α3N and α5 (dark blue) in ZiaR (16); α5C (light green) in NmtR (13); α3N (gray) in AztR (19); α4C (black) in CmtR (14); α3 (red) in ArsR (27); and α5-3 (dark green) in KmtR. The hypothetical α2α5 site (pale blue) is also shown for HlyU. The α3N site in SmtB (residues in blue) is not required for metal responsiveness. B, chart showing the abundance of the various metal-sensing sites among the 554 ArsR-SmtB family representatives. Proteins designated α3Nα5 (ZiaR-like) possess both α5 and α3N sites, although in some cases (e.g. cyanobacterial SmtB and CadC) only one site may be required for metal-mediated allostery. C, representation of the four characterized ArsR-SmtB sensors in M. tuberculosis with effector binding sites for Zn(II) at α5 in dimeric MtSmtB (blue), for Ni(II) and Co(II) at α5C (light green) in NmtR and α5-3 (dark green) in KmtR, and for Cd(II) and Pb(II) at α4C (black) in CmtR. α-helices (boxes), β-strands (arrows), DNA-binding helix-turn-helix region (open boxes), and carboxyl-terminal extension (red) are indicated, the latter being absent from MtSmtB. Binding of their effectors inhibits DNA binding and alleviates repression of their target genes (horizontal arrows). Permutations in the effector binding sites allows detection of the different metals within the cytosol, whereas differences in the sensing ligands at α5C and α5-3 in NmtR and KmtR, respectively, may alter their affinities for Co(II) and Ni(II) allowing these proteins to respond under different surplus Co(II) and Ni(II) conditions.

References

    1. Ginsberg AM, Spigelman M. Nat. Med. 2007;13:290–294. - PubMed
    1. Warner DF, Mizrahi V. Nat. Med. 2007;13:282–284. - PubMed
    1. Malik S, Abel L, Tooker H, Poon A, Simkin L, Girard M, Adams GJ, Starke JR, Smith KC, Graviss EA, Musser JM, Schurr E. Proc. Natl. Acad. Sci. U. S. A. 2005;102:12183–12188. - PMC - PubMed
    1. Hoal EG, Lewis LA, Jamieson SE, Tanzer F, Rossouw M, Victor T, Hillerman R, Beyers N, Blackwell JM, Van Helden PD. Int. J. Tuberc. Lung Dis. 2004;8:1464–1471. - PubMed
    1. Zhang W, Shao L, Weng X, Hu Z, Jin A, Chen S, Pang M, Chen ZW. Clin. Infect. Dis. 2005;40:1232–1236. - PMC - PubMed

Publication types

LinkOut - more resources