Tài liệu Báo cáo khoa học: Studies on structural and functional divergence among seven WhiB proteins of Mycobacterium tuberculosis H37Rv pdf

Studies on structural and functional divergence among
seven WhiB proteins of Mycobacterium tuberculosis
H37Rv
Md. Suhail Alam, Saurabh K. Garg* and Pushpa Agrawal
Institute of Microbial Technology, CSIR, Chandigarh, India
Mycobacterium tuberculosis has a remarkable ability to
survive under hostile conditions it encounters during
infection [1]. Despite extensive research directed
towards understanding the physiology of M. tuberculo-
sis and its molecular pathogenesis [1–3], many funda-
mental questions about the mechanisms of survival
during early infection and persistence remain poorly
understood. Among several intriguing questions, are:
(a) what are the bacterial determinants necessary for
early infection, (b) how does the bacterium counteract
or evade its host’s defenses to survive the vigorous host-
immune response, (c) what regulates the transition from
initial growth to persistence and back to active growth,
(d) are the bacteria present in a non-replicating ‘spore-
like’ state or do they replicate at all during latency, and
(e) how does the bacterium adapt to survive under the
anaerobic and nutritionally altered environment within
the granuloma? The answers to these questions are
likely to provide insight into the mechanisms by which
M. tuberculosis establishes infection and persists within
Keywords
iron–sulfur cluster; Mycobacterium
tuberculosis; protein disulfide reductase;
redox system; WhiB
Correspondence
P. Agrawal, Institute of Microbial

Technology, Sector-39A, Chandigarh
160 036, India
Fax: +91 172 269 0585
Tel: +91 172 263 6680 ⁄ 263 6681; Ext 3264
E-mail:
*Present address
Department of Environmental and Biomolec-
ular Systems, Oregon Health and Science
University, Beaverton, OR, USA
(Received 16 September 2008, revised 22
October 2008, accepted 23 October 2008)
doi:10.1111/j.1742-4658.2008.06755.x
The whiB-like genes (1-7) of Mycobacterium tuberculosis are involved in cell
division, nutrient starvation, pathogenesis, antibiotic resistance and stress
sensing. Although the biochemical properties of WhiB1, WhiB3 and WhiB4
are known, there is no information about the other proteins. Here, we
elucidate in detail the biochemical and biophysical properties of WhiB2,
WhiB5, WhiB6 and WhiB7 of M. tuberculosis and present a comprehensive
comparative study on the molecular properties of all WhiB proteins. UV–
Vis spectroscopy has suggested the presence of a redox-sensitive [2Fe–2S]
cluster in each of the WhiB proteins, which remains stably bound to the
proteins in the presence of 8 m urea. The [2Fe–2S] cluster of each protein
was oxidation labile but the rate of cluster loss decreased under reducing
environments. The [2Fe–2S] cluster of each WhiB protein responded differ-
ently to the oxidative effect of air and oxidized glutathione. In all cases,
disassembly of the [2Fe–2S] cluster was coupled with the oxidation of
cysteine-thiols and the formation of two intramolecular disulfide bonds.
Both CD and fluorescence spectroscopy revealed that WhiB proteins are
structurally divergent members of the same family. Similar to WhiB1,
WhiB3 and WhiB4, apo WhiB5, WhiB6 and WhiB7 also reduced the disul-

fide of insulin, a model substrate. However, the reduction efficiency varied
significantly. Surprisingly, WhiB2 did not reduce the insulin disulfide, even
though its basic properties were similar to those of others. The structural
and functional divergence among WhiB proteins indicated that each WhiB
protein is a distinguished member of the same family and together they
may represent a novel redox system for M. tuberculosis.
Abbreviations
ANS, 8-anilinonapthalene-1-sulfonate; GSH, reduced glutathione; GSSG, oxidized glutathione; IAA, iodoacetamide; ThT, thioflavin T;
Trx, thioredoxin.
76 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
the host and the means to eliminate latent infection, a
phase of the disease that poses the most significant
obstacle to the eradication of tuberculosis. To survive
and establish successful infection, M. tuberculosis
appears to have acquired a strong network of genes to
sense and respond to stress conditions; the properties of
many of these are poorly understood.
A family of genes, whiB, has received attention
because of their involvement in cell division (whiB2),
fatty acid metabolism and pathogenesis (whiB3), antibi-
otic resistance (whiB7) and in sensing a variety of stress
conditions [4–9]. Seven genes, whiB1 ⁄ Rv3219, whiB2 ⁄
Rv3260c, whiB3 ⁄ Rv3416, whiB4 ⁄ Rv3681c, whiB5 ⁄
Rv0022c, whiB6 ⁄ Rv3862c and whiB7 ⁄ Rv3197A, have
been identified in M. tuberculosis [10,11] as orthologs of
the whiB gene of Streptomyces coelicolor A3(2), which
has been shown to be involved in sporulation [12].
Although, WhiB proteins are annotated as putative
transcription factors [12], to date it has not been
shown directly that these proteins work as transcrip-

tion factors. We have previously reported that WhiB1 ⁄
Rv3219 [13], WhiB3 ⁄ Rv3416 [14] and WhiB4 ⁄ Rv3681c
[15] are protein disulfide reductases. WhiB4 has been
postulated to act as a sensor of oxidative stress, wherein
the inactive holo protein (containing a [4Fe–4S] clus-
ter) transformed into an active apo protein (without
an iron–sulfur cluster) in oxidizing environments and
gained protein disulfide reductase activity [15]. How-
ever, to date the biochemical features of WhiB2, WhiB5,
WhiB6 and WhiB7 from M. tuberculosis have not been
reported. The observations that different whiB muta-
tions impart distinct phenotypes and respond differently
to stress conditions indicate importance of each member
separately in mycobacterial physiology. The available
information on WhiB proteins demands careful investi-
gation of the biochemical and biophysical properties of
each.
Mycobacterial WhiB proteins have 22–67% identity
with WhiB protein of S. coelicolor A3(2). Sequence
analysis of M. tuberculosis WhiB proteins shows the
presence of four conserved cysteines arranged as ‘C-X
19-
36
-C-X-X-C-X
5-7
-C’ [16]. Notably, two cysteines are
present in a conserved CXXC motif, except in WhiB5 ⁄
Rv0022c where it is CXXXC (CLRRC). Proteins with
the CXXC motif have been implicated in diverse func-
tions, for example, protein disulfide oxidoreductase

activity [17], redox sensing [18] and the coordination of
metal cofactors [19]. The functional importance of the
conserved cysteine residues in iron–sulfur cluster coor-
dination and protein disulfide reductase has been dem-
onstrated in WhiB4 [15]. Recently, cysteines of WhiB3
have also been shown to act as a ligand for the O
2
- and
NO-responsive [4Fe–4S] cluster [9].
The presence of four conserved cysteines and a
CXXC motif in WhiB proteins from M. tuberculosis
raises several questions: are all WhiB proteins coordi-
nated with an iron–sulfur cluster? If yes, then what are
their basic properties? Are the iron–sulfur clusters
equally oxidation labile? Does removal of the iron–
sulfur cluster lead to disulfide bond formation? Are the
structural features of mycobacterial WhiB proteins
similar? Do all WhiB proteins behave like protein
disulfide reductase? The objective of this study is to
answer several of the questions raised above.
This is the first study to report the biochemical and
biophysical properties of WhiB2, WhiB5, WhiB6 and
WhiB7 of M. tuberculosis and also compare the prop-
erties of all seven WhiB proteins. We show that, simi-
lar to WhiB3 and WhiB4, other freshly purified WhiB
proteins also coordinate a [2Fe–2S] cluster which
respond differently to the oxidizing environment.
Except WhiB2, apo WhiB5, WhiB6 and WhiB7 also
reduce insulin in vitro, but the efficiency of the reduc-
tion varies. An extensive biophysical study suggested

that the WhiB proteins of M. tuberculosis are struc-
turally different. The functional relevance of their
divergent molecular properties is discussed.
Results
All seven whiB genes of M. tuberculosis encode
iron–sulfur proteins
Previous work on WhiB3 [9] and WhiB4 [15] identified
the presence of cysteine-bound iron–sulfur cluster in
these proteins. We speculated that all seven WhiB pro-
teins may also coordinate an iron–sulfur cluster. There-
fore, we overexpressed the recombinant WhiB proteins
(with an N-terminal S-tag and C-terminal 6 · His tag)
in Escherichia coli BL21 (DE3). Overexpression at
37 °C for 3 h led to the formation of light brown inclu-
sion bodies. However, induction at 16 °C for 20 h
resulted in the expression of  10–20% of each WhiB
protein in the soluble form. On SDS ⁄ PAGE, the mass
of Ni
2+
-NTA-purified WhiB proteins corresponded to
their theoretically calculated molecular mass (predicted
molecular mass +  5 kDa tags) (Fig. S1). Proteins
purified from a soluble fraction or after denaturation or
by in-column refolding were  98% pure and were
brownish (Figs S1 and S2).
The presence of four conserved cysteines and the
brownish appearance of purified WhiB1, WhiB2,
WhiB5, WhiB6 and WhiB7 indicated the presence of
an iron–sulfur cluster. To identify and confirm the
presence of the iron–sulfur cluster, the absorption

spectra of the purified proteins were recorded in the
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 77
range 200–700 nm. In addition to a peak at 280 nm,
two additional peaks at  333–340 and  420–
424 nm, along with two broad shoulders at  460 and
 560–580 nm were observed (Fig. 1). The peaks were
characteristic of an [2Fe–2S] cluster [20], therefore, it
was assumed that freshly purified WhiB1, WhiB2,
WhiB5, WhiB6 and WhiB7 also coordinated the [2Fe–
2S] cluster. The absorption spectra of different WhiB
proteins were largely indistinguishable, however, in
WhiB6 and WhiB7, the shoulder at  460 nm was
more prominent than in others. This subtle change in
the peak pattern may be because of their differential
electronic environment. The nature and type of amino
acids and their side-chain orientations around iron–
sulfur cluster coordination sites are the likely cause of
minor variations in the electronic properties, which
were reflected in their absorption spectra.
The brownish appearance of the protein purified in
the presence of 8 m urea indicated that the iron–sulfur
cluster of WhiB proteins had survived treatment by a
denaturant, a feature very similar to WhiB3 [14] and
WhiB4 [15]. Unlike proteins purified from the soluble
fraction, which had a spectral feature typical of the
[2Fe–2S] cluster, proteins in 8 m urea showed a single
peak at  400–415 nm (Fig. S3). The differential peak
features may be due to the solvent-induced confor-
mational change, which is possibly because of changes

in the chemical environment around the iron–sulfur
cluster, the partial destruction of the cluster or its con-
version to other forms. In order to investigate the
probable reason(s) for the observed difference, the pro-
teins were processed for in-column refolding. The
absorption spectra of the in-column refolded proteins
were similar to those of their native counterparts
(Fig. S3). Interestingly, iron–sulfur cluster-specific peak
intensities were similar in both conditions. These data
suggest that the coordination of iron–sulfur clusters to
the WhiB proteins was unaffected by 8 m urea and the
differences in peak patterns were due to the presence
of urea. In order to acquire firm evidence for this
observation, the total iron content of proteins purified
under different conditions was measured.
The total iron content of the native and in-column
refolded protein varied between 0.14 and 0.20 atoms
per monomer (Table 1). The sub-stoichiometric iron
content of iron–sulfur proteins is generally due to the
impaired incorporation of the cluster into the protein
during overexpression in E. coli and ⁄ or loss during
purification when conditions are not strictly anaerobic
[21]. We attempted to reconstitute the iron–sulfur clus-
ter in WhiB proteins in vitro using FeCl
3
and Na
2
S,
but did not succeed. Therefore, incorporated l-cys-
teine as a sulfur source in the reconstitution assay.

IscS ⁄ Rv3025c, a cysteine desulfurase [9] of M. tuber-
culosis was cloned, expressed in E. coli and purified by
metal-affinity chromatography (data not shown). The
WhiB proteins were incubated in the reaction mixture
along with FeCl
3
, IscS and
35
S-cysteine. We observed
an IscS-dependent mobilization of sulfur from l-cyste-
ine to the iron–sulfur cluster of WhiB proteins
(Fig. 2A). In the control reactions, where IscS was
excluded or the iron concentration was limited (10-fold
less), we did not observe any signal (Fig. 2A). Further
characterization of the iron–sulfur cluster of the recon-
stituted samples could not be carried out because none
of the samples gave an EPR signal at 120K using
liquid nitrogen (data not shown). It is possible that a
further decrease in temperature (using liquid helium)
would be required in order to detect the EPR signal.
Nevertheless, the absorption spectra of the reconsti-
tuted proteins showed a single peak at  420 nm indi-
cating the presence of a [4Fe–4S] cluster (Fig. 2B). The
presence of a similar cluster has been reported in
WhiB3 and WhiB4.
The iron content of proteins purified from the solu-
ble fraction, from inclusion bodies, under denaturing
conditions and after refolding was similar (Table 1).
The data clearly suggested that the protein fold
responsible for holding the iron–sulfur cluster was

resistant to the denaturing effect of 8 m urea. The abil-
ity of the iron–sulfur cluster to survive the effects of
protein denaturants is a feature of high potential iron–
sulfur proteins [22]. It is possible that WhiB proteins
also fall into the same category. However, detailed
analysis would be required to establish this.
Iron–sulfur clusters of M. tuberculosis WhiB
proteins are redox sensitive
Previously, we reported that the iron–sulfur cluster of
WhiB4 disintegrates under an oxidizing environment,
but not under reducing conditions [15]. The rate of dis-
integration was directly correlated with the duration
and strength of the oxidizing environment. Similarly,
in this study, the intensity of the brown color and the
iron–sulfur cluster-specific peaks decreased gradually
as the time of exposure to air increased (Fig. S4). The
results suggest that the iron–sulfur clusters were sus-
ceptible to oxidative degradation, although the rate of
degradation varied significantly. WhiB1 lost  65% of
its iron–sulfur clusters in the initial 6 h, whereas in
WhiB6 and WhiB7 the loss was  8–10%. After 48 h
of air exposure, losses were as follows:  80% in
WhiB1,  75% in WhiB2 and WhiB5,  65% in
WhiB3,  60% in WhiB4, and  35–40% in WhiB6
and WhiB7 (Fig. 3A). It was evident that the iron–sulfur
Molecular properties of M. tuberculosis WhiB proteins Md. S. Alam et al.
78 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
300 350 400 450 500 550 600 650 700
0.000
0.015

0.030
0.045
0.060
0.075
0.090
~560–580 nm
420 nm
340 nm
WhiB1, 50 µM
WhiB1, 50 µM, alkylated
WhiB2, 50 µM
WhiB2, 50 µM, alkylated
WhiB3, 50 µM
WhiB3, 50 µM, alkylated
WhiB4, 50 µM
WhiB4, 50 µM, alkylated
WhiB5, 50 µM
WhiB5, 50 µM, alkylated
WhiB6, 50 µM
WhiB6, 50 µM, alkylated
WhiB7, 50 µM
WhiB7, 50 µM, alkylated
Absorbance
λ
(nm)
λ
(nm)
λ
(nm)
λ

(nm)
λ
(nm)
λ
(nm)
λ
(nm)
300 350 400 450 500 550 600 650 700
0.000
0.015
0.030
0.045
0.060
0.075
0.090
~560–580 nm
424 nm
340 nm
Absorbance
300 350 400 450 500 550 600 650 700
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
~560–580 nm
422 nm

Absorbance
337 nm
300 350 400 450 500 550 600 650 700
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
~560– 580 nm
420 nm
337 nm
Absorbance
300 350 400 450 500 550 600 650 700
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
~560–580 nm
424 nm
333 nm

Absorbance
300 350 400 450 500 550 600 650 700
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
Absorbance
~560–580 nm
~460 nm
420 nm
340 nm
300 350 400 450 500 550 600 650 700
0.00
0.03
0.06
0.09
0.12
0.15
0.18
~560–580 nm
~460 nm
424 nm
340 nm
Absorbance
Fig. 1. UV–Vis absorption spectra of WhiB proteins. The absorption spectra of purified proteins (50 lM, thick line) show the presence of a
[2Fe–2S] cluster in WhiB proteins. Numbers (in nm) indicate the peak at the specified wavelength. Alkylation was carried out by incubating

the purified proteins (50 l
M) with 20 mM IAA for 1 h at 25 °C in the dark and the spectra were recorded (thin line) after the baseline correc-
tion. The spectra for WhiB3 and WhiB4 are taken from Alam & Agrawal [14] and Alam et al. [15], respectievely.
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 79
clusters were most stable against air oxidation in
WhiB6 and WhiB7, and most labile in WhiB1.
To study the sensitivity towards oxidized glutathione
(GSSG), reduced glutathione (GSH) and dithiothreitol,
proteins were incubated with 10 mm of each agent and
the absorbance at 424 nm (A
424
) was recorded at dif-
ferent time intervals up to 42 h. All WhiB proteins
showed differential sensitivity towards oxidation by
GSSG, and similar to air oxidation, the iron–sulfur
clusters of WhiB6 and WhiB7 were comparatively
more stable (Fig. 3B). A reducing environment (in the
presence of GSH or dithiothreitol) significantly low-
ered the rate of disintegration of the iron–sulfur cluster
in each of the WhiB proteins (Fig. 3C,D). Therefore,
disassembly of the iron–sulfur cluster under oxidizing
conditions and its stability under reducing conditions
suggested that the iron–sulfur clusters of M. tuberculo-
sis WhiB proteins are redox sensitive. We assume that
the iron–sulfur clusters of different WhiB proteins
would respond differently to the oxidative stress
encountered by M. tuberculosis in vivo.
Iron–sulfur clusters of WhiB proteins are
differentially exposed to the external environment

The differential sensitivity of the iron–sulfur cluster
towards different oxidizing agents could be attributed
to their relative surface accessibility. We hypothesized
that the iron–sulfur cluster of WhiB6 and WhiB7 may
be sequestered in the interior of the holo protein,
thereby shielding it from oxidative degradation. In
Table 1. Total iron content in WhiB proteins. Proteins under differ-
ent conditions were purified as described in Experimental proce-
dures. Data for each protein sample are expressed as means ± SD
(three independent protein preparations).
Samples
Atoms of iron
per monomer
WhiB1 (native) 0.131 ± 0.014
WhiB1 (in 8
M urea) 0.141 ± 0.052
WhiB1 (refolded) 0.138 ± 0.028
WhiB1 (alkylated) 0.008 ± 0.005
WhiB2 (native) 0.145 ± 0.022
WhiB2 (in 8
M urea) 0.142 ± 0.020
WhiB2 (refolded) 0.142 ± 0.035
WhiB2 (alkylated) 0.006 ± 0.005
WhiB5 (native) 0.185 ± 0.028
WhiB5 (in 8
M urea) 0.186 ± 0.036
WhiB5 (refolded) 0.188 ± 0.045
WhiB5 (alkylated) 0.010 ± 0.007
WhiB6 (native) 0.212 ± 0.065
WhiB6 (in 8

M urea) 0.198 ± 0.050
WhiB6 (refolded) 0.208 ± 0.072
WhiB6 (alkylated) 0.007 ± 0.003
WhiB7 (native) 0.182 ± 0.035
WhiB7 (in 8
M urea) 0.189 ± 0.066
WhiB7 (refolded) 0.175 ± 0.020
WhiB7 (alkylated) 0.007 ± 0.005
+ + – + + +
10X 10X 10X 10X 10X 0.1X
WhiB
35
S-Cys
FeCl
3
IscS
– + + + + +
+ – + + + +
WhiB1
WhiB2
WhiB3
WhiB4
WhiB5
WhiB6
WhiB7
Auto radiogram
Ponceau S
stained
300 350 400 450 500 550 600
0.0

0.1
0.2
0.3
0.4
0.5
IscS (2 µM)
WhiB (30 µ
M)
WhiB (30 µM) + IscS (2 µM)
~420 nm
Absorbance
λ
(
nm
)

A
B
Fig. 2. In vitro assembly of the iron–sulfur cluster in WhiB proteins.
(A) The upper panel is an autoradiogram showing IscS-dependent
incorporation of radioactive sulfur in the iron–sulfur cluster of all
seven WhiB proteins. The lower panel is a representative ponceaue
S-stained blot showing the status of protein spotting. Reconstitu-
tion was carried out in a 500-lL reaction volume as described in
Experimental procedures. The concentration of FeCl
3
was either a
10-fold molar excess (10·) or 10-fold lower (0.1·) than that of WhiB
proteins. After the reaction, unwanted and free components were
dialyzed. The poly(vinylidene difluoride) membrane was first treated

with methanol for 5 s and washed thoroughly. The membrane was
equilibrated with a buffer containing 50 m
M Tris ⁄ HCl, pH 9.0,
150 m
M NaCl and 10 mM dithiothreitol for 5 min. In total, 30 lL
(10 lL at one time) of the indicated samples were spotted, air dried
and developed using Phosphorimager (Bio-Rad, Hercules, CA,
USA). (B) Absorption spectra of a representative in vitro reconsti-
tuted WhiB protein. All seven WhiB proteins showed similar
features.
Molecular properties of M. tuberculosis WhiB proteins Md. S. Alam et al.
80 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
order to study the surface accessibility of the iron–
sulfur cluster of WhiB proteins, freshly purified proteins
were incubated with 20 mm EDTA and the degree of
iron chelation was monitored by recording the A
424
at
different intervals up to 20 h. Immediate chelation of
iron was not observed in any WhiB protein. However,
as time increased, the degree of iron chelation
increased and the extent of chelation varied (Table 2).
Almost 30% of the iron was chelated within 2 h in
WhiB1 and WhiB2, whereas in WhiB6 and WhiB7 it
was negligible over the same period. Even after 20 h of
incubation, iron chelation was  15–20% (minimum)
in WhiB6 and WhiB7, whereas it was  60% (maxi-
mum) in WhiB1 and WhiB2; in the other proteins it
varied between 20% and 60% (Table 2). From the
data, it appears that the surface accessibility of the

iron–sulfur cluster is significantly different in different
WhiB proteins.
Cysteine residues form two intramolecular
disulfide bonds after removal of the
iron–sulfur cluster
It has been shown in WhiB4 that the cysteine-thiols,
which are ligands of the iron–sulfur cluster, undergo
oxidation and form two intramolecular disulfide bonds
after disassembly of the iron–sulfur cluster [15]. The
presence of two intramolecular disulfide bonds has also
been demonstrated in apo WhiB1 [13] and apo WhiB3
[14]. Proteins containing intramolecular disulfide
bond(s) often show retarded mobility on SDS⁄ PAGE
under reducing conditions [15,23]. Both apo WhiB2
and apo WhiB5 showed significant retarded mobility
on SDS ⁄ PAGE under reducing conditions, indicating
the presence of intramolecular disulfide bond(s)
(Fig. 4). Alkylation of cysteine by iodoacetamide
(IAA) coupled with MS was used to determine the sta-
tus of cysteines after removal of the iron–sulfur cluster.
Reaction of IAA with a cysteine-thiol causes an
increase in molecular mass of 57 Da, therefore, the
0
15
30
45
60
7575
90
105

120
Effect of air
0 h
6 h
12 h
24 h
48 h
WhiB7
WhiB6
WhiB5
WhiB4
WhiB3
WhiB2
WhiB1
% Change in A
420
0
15
0303
45
0606
7575
0909
105
021021
Effect of GSSG (10 mM)
0 h
2 h
6 h
20 h

30 h
42 h
WhiB7
WhiB6
WhiB5
WhiB4
WhiB3WhiB2
WhiB1
% Change in A
420
0
15
0303
45
0606
75
75
0909
105
021021
Effect of GSH (10 mM)
0 h
2 h
6 h
20 h
30 h
42 h
WhiB7
WhiB6WhiB5
WhiB4

WhiB3
WhiB2
WhiB1
% Change in A
420
15
0
0303
45
0606
7575
0909
105
021021
Effect of dithiothreitol (10 mM)
0 h
2 h
6 h
20 h
30 h
42 h
WhiB7
WhiB6
WhiB5
WhiB4
WhiB3
WhiB2
WhiB1
% Change in
A

420
AB
CD
Fig. 3. Bar diagram showing the kinetics of [2Fe–2S] cluster loss upon treatment with various oxidizing and reducing agents. Effect of (A)
air, (B) GSSG, (C) GSH and (D) dithiothreitol. Freshly purified protein (50 l
M) was incubated with 10 mM of the indicated agents at 25 °C.
The loss of the [2Fe–2S] cluster was measured by recording A
420
at different time intervals. The reading at t = 0 was set to 100% and the
change in A
424
(residual) is expressed relative to the first reading. A suitable baseline correction was made before recording each spectrum.
Results are an average of three independent protein preparations. Values for WhiB4 were taken from Alam et al. [15].
Table 2. Stability of the iron–sulfur cluster from various WhiB
proteins against EDTA. The initial reading was set to 100% and
the change in A
420
(residual) is expressed relative to the reading
at t = 0. Data are expressed as means ± SD (three independent
protein preparations).
Proteins
Change in A
420
(%)
0h 2h 6h 20h
WhiB1 100 62 ± 4 53 ± 4 44 ± 3
WhiB2 100 72 ± 5 44 ± 3 38 ± 6
WhiB3 100 82 ± 3 70 ± 5 58 ± 3
WhiB4 100 92 ± 2 59 ± 5 49 ± 6
WhiB5 100 80 ± 2 65 ± 3 58 ± 3

WhiB6 100 96 ± 3 89 ± 2 84 ± 2
WhiB7 100 98 ± 2 96 ± 4 80 ± 4
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 81
total increase in the mass after reduction of the disul-
fide bond reflects the total number of cysteines present
in the thiol and disulfide forms. In the oxidized state,
both WhiB2 and WhiB5 showed a major peak corre-
sponding to the theoretical molecular mass of the
recombinant protein. However, the reduced proteins
had increased molecular masses, representing alkyl-
ation of four cysteine residues in each case (Fig. 5).
Although, WhiB5 and WhiB6 did not show any mobil-
ity differences under reducing conditions, a similar
increase in mass was found after reduction (Fig. 5).
The difference in mass between the oxidized and
reduced forms suggested the presence of four cysteine-
thiols in the reduced apo WhiB proteins. Because none
of the cysteines was present in a thiol form in the oxi-
dized protein (except for one in WhiB6 which has five
cysteines), it was concluded that the apo form of all
WhiB proteins contained two intramolecular disulfide
bonds.
All apo WhiB proteins, except WhiB2, reduce the
insulin disulfide
Previously, we reported that apo WhiB1 [13] WhiB3
[14] and WhiB4 [15] are protein disulfide reductases.
The enzymatic activity of WhiB4 was shown to be gov-
erned by the CXXC motif [15]. Because the CXXC
motif is present in all WhiB family members of

M. tuberculosis, except WhiB5 ⁄ Rv0022c (CXXXC), we
tested the protein disulfide reductase activity of WhiB2,
WhiB5, WhiB6 and WhiB7 by insulin disulfide reduc-
tion assay. This is a standard assay to asses the disulfide
reductase activity of any protein in which reduction of
the insulin disulfide by dithiothreitol in the presence of
a test protein is monitored [24]. Reductase activity was
calculated by dividing the maximal slope of the curve
(DA
650
Æmin
)1
) by the onset time of precipitation (time
when A
650
reached 0.05) [25]. Except WhiB2, all WhiB
proteins catalyzed the reduction of insulin disulfide
(Table 3, Fig. S5). However, for WhiB2, the possibility
of the presence of a natural in vivo substrate protein(s)
still remains. Because M. tuberculosis RshA also has
aC
86
XXC
89
motif, and purified recombinant RshA
(lab preparation) was therefore used as a control, but it
did not catalyze insulin reduction.
Formation of a reversible intramolecular disulfide
bond between the cysteines of the CXXC motif is
essential for protein disulfide reductase activity [16,26].

Therefore, we assume that in WhiB proteins, one disul-
fide bond is formed between the two cysteines of
the CXXC motif (CXXXC in the case of WhiB5) and
the other between the remaining two cysteines. The
assumption is supported by our earlier data, in which
a similar arrangement of intramolecular disulfide
bonds in WhiB4 was established [15]. In WhiB6, one
of the intramolecular disulfide bonds appeared to have
formed between Cys53 and Cys56 but the involvement
of cysteines for the second bond is little hard to pre-
dict, as it contains five cysteines (Cys12, Cys34, Cys53,
Cys56 and Cys62).
Divergence in the secondary structure
composition of WhiB proteins of M. tuberculosis
The multiple sequence alignment of WhiB proteins of
M. tuberculosis showed 49–66% sequence homology
and 31–50% identity with respect to each other
(Table S1). However, because of the variation in amino
acid composition, it is possible that structural variations
may be an important determinant of their functional
properties in vivo. Therefore, the structural organization
of each M. tuberculosis WhiB protein was studied using
biophysical tools. The secondary structure of each
WhiB5
26.9
20.0
36.5
WhiB6
18.4
14.4

25.0
WhiB7
18.4
14.4
25.0
Marker
Oxidized
1 m
M dithiothreitol
2%
β
-ME
26.9
20.0
36.5
WhiB2
kDa
Fig. 4. Mobility shift of apo WhiB2, WhiB5, WhiB6 and WhiB7 on
15% SDS ⁄ PAGE. Oxidized protein (5 lg) was incubated in the
absence or presence of different reducing agents, as indicated, for
1 h at 25 °Cin50m
M Tris ⁄ HCl, pH 8.0, 200 mM NaCl. After reduc-
tion, free thiols were alkylated with 20 m
M IAA for 1 h at 25 °Cin
the dark. Finally the samples were resolved by 15% SDS ⁄ PAGE
and proteins were visualized by Coomassie Brilliant Blue staining.
Molecular properties of M. tuberculosis WhiB proteins Md. S. Alam et al.
82 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
WhiB protein was analyzed by CD spectroscopy. The
far-UV CD spectra of WhiB proteins were dissimilar

because their molar ellipticities varied significantly
(Fig. 6). The spectra showed a-helix, b-strand and
random coil features. However, the proportion of each
feature varied among WhiB proteins, as evident from
the difference in negative molar ellipticity at specific
wavelengths, i.e. 208 and 222 nm (a helix signature),
218 nm (b strand signature), 202–204 nm (random coil
signature). In WhiB5 and WhiB6, the structure was
dominated by a helices and b strands and the propor-
tion of these structural elements was higher in WhiB6.
WhiB1, WhiB2 and WhiB4 showed relatively increased
molar ellipticity at 202–204 nm, indicating the presence
of a significant proportion of random coils (Fig. 6). The
Fig. 5. MALDI-TOF spectroscopic analysis of oxidized and reduced form of apo WhiB proteins. ‘Oxd’ represents the ‘oxidized and alkylated’
protein, whereas ‘Red’ represents ‘reduced and alkylated’ protein.
Table 3. Protein disulfide reductase activity of apo WhiB proteins.
The data for each protein sample (3 l
M) are expressed as
means ± SD (three independent protein preparations).
Samples
Reductase activity
(· 10
)3
DA
650
nmÆmin
)2
)
WhiB1 4.78 ± 0.25
WhiB2 0.56 ± 0.16

WhiB3
a
4.19 ± 0.62
WhiB4
b
42.2 ± 0.86
WhiB5 10.96 ± 0.55
WhiB6 2.79 ± 0.22
WhiB7 5.44 ± 0.82
RshA 0.58 ± 0.10
Buffer control 0.62 ± 0.15
a
Taken from Alam & Agrawal [14].
b
Taken from Alam et al. [15].
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 83
–14 000
–12 000
–10 000
–8000
–6000
–4000
–2000
0
2000
WhiB2, Oxd
WhiB2, Red
190 200 210 220 230 240 250
–6000

–5000
–4000
–3000
–2000
–1000
0
1000
2000
WhiB5, Oxd
WhiB5, Red
λ
(nm)
190 200 210 220 230 240 250
λ
(nm)
190 200 210 220 230 240 250
λ
(nm)
190 200 210 220 230 240 250
λ
(nm)
190 200 210 220 230 240 250
λ
(nm)
λ
(nm)
190 200 210180 220 230
240 250
–6000
–5000

–4000
–3000
–2000
–1000
0
1000
2000
3000
4000
WhiB6, Oxd
WhiB6, Red
–5000
–4000
–3000
–2000
–1000
0
1000
WhiB3, Oxd
WhiB3, Red
–10 000
–8000
–6000
–4000
–2000
0
WhiB4, Oxd
WhiB4, Red
–14 000
–12 000

–10 000
–8000
–6000
–4000
–2000
0
2000
[
θ]
MRW
(deg·cm
2
·
dmol
–1
)
[
θ]
MRW
(deg·cm
2
·
dmol
–1
)
[
θ]
MRW
(deg·cm
2

·
dmol
–1
)
[
θ]
MRW
(deg·cm
2
·
dmol
–1
)
190 200 210 220 230 240 250
λ
(nm)
–4000
–3500
–3000
–2500
–2000
–1500
–1000
–500
0
500
WhiB7, Oxd
WhiB7, Red
[θ]
MRW

(deg·cm
2
·
dmol
–1
)
[
θ]
MRW
(deg·cm
2
·
dmol
–1
)
[
θ]
MRW
(deg·cm
2
·
dmol
–1
)
WhiB1, Oxd
WhiB1, Red
Fig. 6. Secondary structure analyses of apo WhiB proteins. Far-UV
CD spectra of apo proteins (0.2 mgÆmL
)1
) were recorded at 25 °C.

Reduction was carried out by incubating the protein in buffer C
supplemented with 1 m
M dithiothreitol for 1 h at 25 °C. The far-UV
CD spectrum of WhiB1 (0.2 mgÆmL
)1
) was recorded as described
in Garg et al. [13], whereas the data of WhiB3 and WhiB4 were
taken from Alam & Agrawal [14] and Alam et al. [15] respectively.
Molecular properties of M. tuberculosis WhiB proteins Md. S. Alam et al.
84 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
proportion of all three secondary structural elements in
WhiB3 and WhiB7 appeared similar. Disulfide bond
formation did not affect the secondary structure, except
for WhiB5 and WhiB6, and the effect was more
pronounced in WhiB6 (Fig. 6).
It was observed that the secondary structure of
WhiB1 [13], WhiB3 [14] and WhiB4 [15] resists thermal
denaturation. Thus, we asked would other WhiB pro-
teins also show similar features? Surprisingly, the sec-
ondary structures of WhiB5 and WhiB6 started to
melt at 50 and 77 °C respectively (Fig. 7). At 70 °C,
WhiB5 lost almost all its secondary structure, whereas
in WhiB6 some secondary structure was maintained
even at 95 °C. Neither WhiB2 nor WhiB7 showed
thermal denaturation. Together, the data suggest that
considerable structural differences exist among the
WhiB proteins of M. tuberculosis, with WhiB5 and
WhiB6 appearing to be the most structurally divergent
family members.
CD spectroscopy is considered more sensitive for the

recognition of a helices and less reliable for b strands,
thus the data obtained from CD spectroscopy are an
approximate assessment. To estimate the level of
b-structures in different WhiB proteins, a thioflavin T
(ThT)-binding assay was performed. ThT shows strong
fluorescence in the presence of crossed b-sheet struc-
tures [27,28]. The binding of ThT to each apo WhiB
protein was measured by fluorescence spectroscopy.
20 30 40 50 60 70 80 90 100
–8000
–6000
–4000
–2000
0
Temp (°C)
222 (deg·cm
2
·dmol
–1
)

WhiB1, Oxd
WhiB1, Red
–2400
–1600
–800
0
WhiB3, Oxd
WhiB3, Red
222 (deg·cm

2
·dmol
–1
)

20 30 40 50 60 70 80 90 100
Temp (°C)
–2000
–1600
–1200
–800
–400
0
WhiB7, Oxd
WhiB7, Red
222 (deg·cm
2
·dmol
–1
)

20 30 40 50 60 70 80 90 100
Temp (°C)
–6000
–5000
–4000
–3000
–2000
–1000
0

WhiB4, Oxd
WhiB4, Red
222 (deg·cm
2
·dmol
–1
)

20 30 40 50 60 70 80 90 100
Temp (°C)
–4000
–3000
–2000
–1000
0
WhiB5, Oxd
WhiB5, Red
222 (deg·cm
2
·dmol
–1
)

20 30 40 50 60 70 80 90 100
Temp (°C)
–5000
–4000
–3000
–2000
–1000

0
WhiB6, Oxd
WhiB6, Red
222 (deg·cm
2
·dmol
–1
)

30 40 50 60 70 80 90 100
Temp (°C)
WhiB2, Oxd
WhiB2, Red
–6000
–5000
–4000
–3000
–2000
–1000
0
222 (deg·cm
2
·dmol
–1
)

20 30 40 50 60 70 80 90 100
Temp (°C)
Fig. 7. Thermal denaturation kinetics of oxidized and reduceed apo WhiB proteins. Far-UV CD spectra of each protein showed pronounced
ellipticity at 222 nm therefore, the thermal stability of the secondary structure was tested by recording the change in ellipticity at 222 nm

with increasing temperature from 25 to 90 °C. The thermal denaturation kinetics of WhiB1 (0.2 mgÆmL
)1
) was studied as described in Garg
et al. [13], whereas the data of WhiB3 and WhiB4 were taken from Alam & Agrawal [14] and Alam et al. [15] respectively.
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 85
The fluorescence intensities of each protein varied with
respect to each other (Fig. 8). In WhiB5 and WhiB6 it
was several fold higher than in the others, whereas for
the rest it was within a similar range (Fig. 8). From
the data, it appears that b-sheet structures are a major
contributors in WhiB5 and WhiB6, but the proportion
is relatively low and similar in other WhiB proteins. It
should be noted that WhiB5 aligned only with WhiB3
and WhiB4, whereas the WhiB6 sequence aligned only
with WhiB4 (NCBI; />blast/bl2seq/wblast2.cgi) (Table S1), suggesting that at
the amino acid sequence level both WhiB5 and WhiB6
differ from the others, and the difference is clearly
reflected in their secondary structure. The other impor-
tant point is that although the WhiB proteins were
predicted to attain a a-helical structure [16], WhiB5
and WhiB6 appeared to have significant amounts of
b strand. In WhiB5 and WhiB6 the reduction of intra-
molecular disulfide bonds resulted in a slight decrease
in b-sheet structure (decreased fluorescence intensity),
whereas this was reversed in the other proteins.
Structural diversification at the tertiary structure
level among different WhiB proteins
Both CD spectroscopy and the ThT-binding assay pro-
vide information only about the secondary structure.

Therefore, to investigate differences at a tertiary level,
the surface topology of each of the WhiB proteins was
studied. The presence of total exposed hydrophobic
patches was probed using 8-anilinonapthalene-1-sulfo-
nate (ANS) binding by fluorimetry. The ANS
fluorescence of various WhiB proteins was significantly
different (Table 4). The fluorescence intensity of
WhiB3 was maximal and threefold higher than that of
WhiB6 (minimum). Based on ANS fluorescence, the sur-
face hydrophobicity of various WhiB proteins decreases
in the following order: WhiB3 > WhiB1 > WhiB4 >
WhiB2 > WhiB5 > WhiB7 > WhiB6 (Table 4). WhiB4
showed a significant increase in ANS fluorescence after
reduction of the disulfide bonds, whereas WhiB1,
WhiB2, WhiB5 and WhiB7 showed a low but repro-
ducible increase in ANS fluorescence after reduction of
the disulfide bonds, indicating a minor conformational
change followed by the exposure of certain buried
hydrophobic patches.
CD and fluorescence data clearly suggested struc-
tural variations among seven WhiB proteins of
M. tuberculosis. Furthermore, we observed that in wes-
tern blot analysis polyclonal antibodies raised against
WhiB1 did not cross-react with WhiB4 and vice versa
(data not shown). This may be because of variations in
the antigenic epitopes of WhiB1 and WhiB4, which is
likely to be due to conformational differences between
the two proteins. The relative cross-reactivity of poly-
clonal antibodies was used to probe the conforma-
tional variations at the tertiary level among WhiB

proteins. Polyclonal antibodies against all WhiB pro-
teins were raised in rabbit and their cross-reactivity
was tested by ELISA. Dilutions of primary antibodies
which gave A
450
= 0.5 ± 0.05 were selected for the
cross-reactivity assay and were as follows: 1 : 35 000
for WhiB1; 1 : 25 000 for WhiB2; 1 : 20 000 for
WhiB3, WhiB4 and WhiB5; 1 : 15 000 for WhiB6 and
1 : 50 000 for WhiB7. The activity of the positive con-
trol (antibody raised against that particular antigen)
was taken to be 100% and cross-reactivity with other
WhiB proteins was represented relative to the positive
0
20
40
60
150
200
250
300
WhiB7
WhiB6
WhiB5
WhiB4
WhiB3
WhiB2
WhiB1
Fluorescence itensity (at 482 nm)
Oxidized

Reduced
Fig. 8. ThT-binding assay. Bar diagram showing the maximum fluo-
rescence intensity of WhiB proteins (5 l
M) after binding with ThT
(50 l
M). The samples were excited at 340 nm and the emission
spectra were recorded between 420 and 600 nm. The maximum
fluorescence intensity for each of the proteins is represented in the
form of a bar. Values obtained in blanks (without protein) were sub-
stracted from the sample readings. Results are an average of three
independent experiments.
Table 4. ANS binding to WhiB proteins. Surface hydrophobicities
were calculated as described in Experimental procedures. The val-
ues obtained in blanks (without protein) were subtracted form the
sample readings. Data are expressed as means ± SD (three inde-
pendent protein preparations).
Proteins
Surface hydrophobicity (AU)
Oxidized Reduced
WhiB1 5.68 ± 0.33 6.40 ± 0.38
WhiB2 4.85 ± 0.24 5.62 ± 0.21
WhiB3 7.44 ± 0.47 7.71 ± 0.46
WhiB4 4.44 ± 0.22 6.00 ± 0.42
WhiB5 4.21 ± 0.21 4.61 ± 0.70
WhiB6 2.30 ± 0.18 2.38 ± 0.21
WhiB7 3.25 ± 0.26 3.58 ± 0.61
Molecular properties of M. tuberculosis WhiB proteins Md. S. Alam et al.
86 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
control. Figure 9 clearly shows considerable differences
in antibody cross-reactivity. The differential cross-reac-

tivity suggested that the antigenic sites (mostly confor-
mational epitopes) on different WhiB proteins are
heterogeneous. We assume that the conformational dif-
ference among various WhiB proteins is the possible
cause of this heterogeneity. Because WhiB proteins
share significant sequence homology minor cross-reac-
tivity may be due to the presence of antibodies which
are reacting to the linear epitopes or to conserved con-
formational epitopes. The data showed that there are
significant structural differences among WhiB proteins
of M. tuberculosis.
Discussion
The purified recombinant WhiB proteins had spectral
resemblance to proteins coordinating the [2Fe–2S] clus-
ter. Earlier studies also showed that recombinant
WhiB3 [9] and WhiB4 [15] of M. tuberculosis, purified
under normal conditions, coordinate a [2Fe–2S] clus-
ter. However, based upon the in vitro reconstitution
assay, both were found to be [4Fe–4S] cluster coordi-
nating proteins. Iron–sulfur clusters are one of the
most ancient and versatile cofactors of several impor-
tant class of proteins and have been implicated in vari-
ety of functions [29,30]. Iron and sulfur can be
combined in several ways to produce different cluster
types, e.g. [2Fe–2S], [3Fe–4S], [4Fe–4S] and more com-
plex structures [31]. The susceptibility of iron–sulfur
clusters to oxidation makes them a good sensor of
redox conditions within a cell [32,33]. It has been com-
monly observed that [4Fe–4S] clusters are highly oxi-
dation labile and rapidly transform into the relatively

stable [2Fe–2S] cluster [34,35], therefore, it is possible
that during purification under normal conditions (as
strict anaerobic conditions could not be maintained),
the [4Fe–4S] cluster undergoes oxidation and forms
the [2Fe–2S] cluster which is relatively stable. How-
0
20
40
60
80
100
120
Anti-WhiB1 Ab
Reactivity (%)Reactivity (%)Reactivity (%)Reactivity (%)Reactivity (%)Reactivity (%)Reactivity (%)
0
20
40
60
80
100
120
Anti-WhiB2 Ab
120
0
20
40
60
80
100
Anti-WhiB3 Ab

120
AntiWhiB4 Ab
0
20
40
60
80
100
100
120
Anti-WhiB5 Ab
0
20
40
60
80
100
100
120
Anti-WhiB6 Ab
0
20
40
60
80
80
100
120
Anti-WhiB7 Ab
0

20
40
60
80
WhiB7
W
h
iB6
WhiB5
WhiB4
WhiB3
WhiB2
WhiB1
Fig. 9. Antibody cross-reactivity assay by ELISA. Bar diagram
showing the percentage cross-reactivity of all seven polyclonal anti-
bodies raised against each of the WhiB proteins. The absorbance
obtained in the positive control (where the antibody used was
raised against that particular antigen) was set to 100% and the
cross-reactivity obtained with other WhiB proteins is expressed rel-
ative to positive control. The readings obtained in negative controls
(BSA coated) were substracted from test readings. The other con-
trols were pre-immune sera, wells incubated with primary antibod-
ies only and with secondary antibodies only. The labels mentioned
on the x-axis denote the respective WhiB proteins used as antigen
in each case. All experiments were performed in triplicate and the
results are an average of three independent experiments.
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 87
ever, in vivo, the possibility of the presence of [4Fe–4S]
in WhiB proteins remains very high.

Protein ligands for the canonical clusters are typi-
cally sulfide ions of cysteines. The involvement of four
conserved cysteines in the coordination of the [4Fe–4S]
cluster has been confirmed in WhiB3 [9] and WhiB4
[15]. The brown color of the purified protein and the
iron–sulfur cluster-specific peaks also disappeared
when proteins were treated with IAA (Figs 1 and S2),
supporting the notion that the cysteines are the proba-
ble iron–sulfur cluster coordination sites in WhiB
proteins of M. tuberculosis. Because WhiB6 has five
cysteine residues, a detailed investigation is needed to
determine the exact coordination sites. Although the
arrangement of cysteines in the primary structure is
not identical, the ‘C-X
2
-C-X
5
-C’ arrangement (except
for WhiB5 where it is ‘C-X
3
-C-X
7
-C’) is conserved and
the position of the N-terminal cysteine appears to be
flexible. Therefore, in the primary structure, although
the cysteines lie far apart with respect to each other, in
a 3D form they are likely to come closer to hold the
iron–sulfur cluster. The formation of two intramolecu-
lar disulfide bonds upon cluster removal strengthened
the argument that in the 3D form, the cysteines are in

close proximity.
Conditional expression of whiB genes in response to
environmental stress is well documented [8,36,37]. The
differential expression of whiB genes under different
growth phases [8] indicated that the WhiB proteins of
M. tuberculosis are segregated temporally. The differ-
ential response showed by whiB genes under different
stress conditions [8] and our observation that all seven
WhiB proteins responded differently to oxidizing envi-
ronments in vitro, propose that the WhiB proteins
of M. tuberculosis would respond to the redox signal
differently.
M. tuberculosis has evolved a remarkable ability to
adapt to the changing environment during infection.
In the process, it has evolved a highly efficient network
of specific gene products which assures survival and
multiplication under unfavorable nutritional, pH and
redox conditions [1]. Among several mechanisms for
resistance to intracellular killing, the scavenging of free
radicals and the reactivation of degenerated proteins
during infection by a family of proteins named ‘thio-
redoxins’ is one of the elegant mechanisms of defense
and self-sustenance seen in several organisms [38–40].
Oxidized thioredoxin (Trx) with a disulfide at its active
site (CXXC) is reduced by NADPH and thioredoxin
reductase (TrxR), and then functions as a general
protein disulfide reductase [41,42].
Antioxidant defense in M. tuberculosis always
appeared unusual in many aspects. Similar to other
actinomycetes, M. tuberculosis lacks a glutathione-

dependent detoxification system. Instead, it contains
mycothiol [43] and a mycothiol reductase [44]. In the
absence of functional oxyR, fnr, etc. [10,45] it appears
that M. tuberculosis has evolved several accessory
signaling molecules and a protective network which
remain unexplored. The presence of the WhiB family
in the form of disulfide reductase is likely to be an
important components of this unexplored system.
Although insulin is a non-natural substrate, the dif-
ference in activity indicates that the efficiency of reduc-
tion of target disulfides by WhiB proteins may vary
in vivo. Moreover, the involvement of a particular
WhiB protein in vivo would depend upon their struc-
tural features and redox potential. The redox potential
of disulfide oxidoreductases has been shown to be
largely determined by the amino acids present in and
around the active site (CXXC motif) [46,47]. WhiB
proteins of M. tuberculosis did not show any conserva-
tion in the dipeptide present between the two cysteines
of CXXC motif (Fig. S6). Therefore, we argue that
each protein is likely to have a different redox poten-
tial. The crystal structures of several thioredoxins have
shown that the active site is located at the N-terminus
of a-2 helix of thioredoxin fold and is separated by a
kink from rest of the helix due to presence of a proline
[25,48–50]. The kink provides proper positioning of
the active site to facilitate electron flow during the
thiol–disulfide exchange reactions [51]. Interestingly, in
all WhiB proteins except WhiB2, the amino acid imme-
diately after the C-terminal cysteine of the CXXC

motif is proline (Fig. S6). In WhiB2 the absence of
proline and therefore of the kink might not allow the
protein to attain an optimal geometry, even though
there is a CXXC motif, it lacked reductase activity.
Gel-filtration chromatography suggested that apo
WhiB2 exists as a homodimer (data not shown)
indicating that the overall surface topology is well con-
structed and the non-covalent interactions are properly
established. However, a lack of reductase activity due
to the absence of a native-like structure could not be
ruled out. Interestingly, in WhiB5 despite the insertion
of an extra amino acid between the two cysteines of
the CXXC motif, the protein retained disulfide reduc-
tase activity. It is possible that during evolution, or
due to some mutational event, an extra amino acid
was inserted, however WhiB5 retained disulfide reduc-
tase activity due to its functional importance (yet to be
discovered) in M. tuberculosis.
E. coli Trx has been implicated in at least 26 distinct
cellular processes [52]. The cell division protein FtsZ,
rod-determining protein MreB, enzymes of fatty acid
metabolism, etc. have been shown to interact with Trx
Molecular properties of M. tuberculosis WhiB proteins Md. S. Alam et al.
88 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
[52,53]. Thioredoxin-mediated modulation in the activ-
ity of transcription factors NF-jB and AP-1 is well-
documented [54,55]. Therefore, a change in gene
expression after the deletion of whiB7 [6] could be seen
as the absence of a component which possibly exerts
its function via a protein–protein interaction rather

than a DNA–protein interaction. To date, there has
not been a single report to show that the WhiB pro-
teins work as a DNA-binding protein. It is possible
that they act as an essential structural component of a
multiprotein complex and ⁄ or are directly or indirectly
involved in mediating the assembly of transcription
factors that regulate the expression of downstream
gene(s).
Iron–sulfur clusters are known to modulate the
structural and functional properties of several pro-
teins [29,56]. Because the cysteine-thiols would be
ligated to iron–sulfur cluster in holo WhiB proteins,
they are not free for electron flow and disulfide
exchange. Therefore, there is a high possibility that
removal of the cluster is essential for the reductase
activity of WhiB proteins. Indeed, the iron–sulfur
cluster of M. tuberculosis WhiB4 has a regulatory
role [15]. Removal of the iron–sulfur cluster under
oxidizing conditions is associated with conforma-
tional change followed by a gain of disulfide reduc-
tase activity.
A fundamental question is that if one thioredoxin-
related protein ‘can do it all’, why was evolution direc-
ted to choose multiple Trx-like proteins. We do not
have clear answer to this question, but in general, mul-
tigene families are thought to have evolved to address
questions of substrate specificity. In a scenario in
which there is a difference in structural organization
among WhiB proteins and because different whiB
genes are involved in different cellular processes, the

existence of target ⁄ substrate specificity appears con-
ceivable. Identification of the cellular targets of WhiB
proteins is our next objective. From the data described
here, we assume that each of the WhiB proteins in
M. tuberculosis, although a part of single family, is
an entity in itself. Furthermore, because of the seven
proteins, six are disulfide reductase, we propose that
WhiB proteins represent a novel redox system in
M. tuberculosis.
Experimental procedures
Bacterial strains and reagents
Genomic DNA from M. tuberculosis H37Rv was prepared
as described previously [57]. E. coli DH5a was used for
general cloning procedures, whereas expression of recombi-
nant proteins was carried out in BL21(DE3). Plasmid pET-
29a and E. coli BL21(DE3) were purchased from Novagen
(Darmstadt, Germany). T4 DNA ligase was obtained from
Promega (Madison, WI, USA) and restriction ⁄ DNA-modi-
fying enzymes were from New England Biolabs (Ipswich,
MA, USA). Oligonucleotide primers were supplied by Bio-
basic Inc. (Markham, Canada). Ni
2+
-NTA agarose and
other PCR ⁄ plasmid purification kits were from Qiagen
(Hilden, Germany). The EDTA-free protease inhibitor
cocktail was from Roche (Mannheim, Germany). All ana-
lytical grade chemicals were either from Sigma-Aldrich
(Bangalore, India) or as indicated. Throughout the study,
all buffers were thoroughly degassed and purged with nitro-
gen just before use. When required, protein samples were

purged with nitrogen before storage or transfer. Standard
recombinant DNA techniques were followed as described
elsewhere [58].
Protein overexpression and purification
Gene-specific primers were used to amplify the complete
ORFs (without stop codon) of whiB2 ⁄ Rv3260c,
whiB5 ⁄ Rv0022c, whiB6 ⁄ Rv3862c, whiB7 ⁄ Rv3197A and
iscS ⁄ Rv3025c using genomic DNA of M. tuberculosis
H37Rv as a template. The nucleotide sequence of primers
is listed in Table S2. The authenticity of PCR products
was determined by sequencing both DNA strands in an
ABI Prism automated DNA Sequencer 310. PCR-ampli-
fied ORFs were cloned at EcoRI ⁄ XhoI(whiB 2 and
whiB6), EcoRI ⁄ SalI(whiB5 and whiB7) or SacI ⁄ HindIII
(iscS) sites of pET-29a to get pET-O (O defines the
respective ORFs) under isopropyl thio-b-d-galactoside-
inducible promoter. The pET-O transformed E. coli BL21
(DE3) cells were grown at 37 °C in Luria–Bertani broth
containing 30 lgÆmL
)1
kanamycin until D
600
was  0.5–
0.6 and expression was induced by the addition of
0.3 mm isopropyl thio-b-d-galactoside for 18–20 h at
16 °C (for WhiB proteins) or for 3 h at 30 °C (for IscS).
Cells from 1000 mL culture were re-suspended in 10 mL
buffer A (50 mm NaH
2
PO

4,
300 mm NaCl, 30 mm imid-
azole, protease inhibitor cocktail, pH 8.0). After one
freeze–thaw cycle, cells were lyzed by sonication. The
proteins were purified from the soluble fraction using
Ni
2+
-NTA affinity column as described by the manufac-
turer. WhiB1, WhiB3 and WhiB4 were expressed and
purified using previously made constructs as described
previously [13–15]. Proteins were purified under denatur-
ing conditions as described by the manufacturers (Qia-
gen), whereas for refolding the ‘matrix-assisted in-column
refolding’ method was used as described previously [14].
Purified proteins were either used directly for spectro-
scopic ⁄ biochemical analysis or were dialyzed as required.
Throughout the study, unless otherwise mentioned,
proteins purified from the soluble fraction of the lysate
were used.
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 89
Effect of EDTA on iron chelation
Freshly purified proteins (50 lm) were incubated with
20 mm EDTA at 25 °C. Samples were aliquoted at different
time intervals, and the extent of chelation by EDTA was
studied by measuring absorption at 420 nm. In parallel,
samples without EDTA (as a control) were also incubated
similarly and A
420
was measured. To exclude the effect of

air oxidation, readings from control samples were sub-
stracted from the test samples and the values thus obtained
are represented. The initial reading was set to 100% and
the change in A
420
(residual) is expressed relative to the
reading at t = 0. Suitable baseline corrections were made
before recording the absorbance.
In vitro assembly of iron–sulfur cluster by IscS
using radiolabeled cysteine
The purified proteins were dialyzed against buffer B
(50 mm Tris ⁄ HCl, pH 9.0, 150 mm NaCl, 10 mm dith-
iothreitol). Purified IscS from M. tuberculosis was used as
a cysteine desulfurase. The reaction (500 lL) was set in
buffer B and contained 30 lm WhiB protein, 2 lm IscS,
300 lm FeCl
3
, 300 lml-cysteine and 15 lCi
35
S-labeled
l-cysteine (BARC, Mumbai, India). The reaction was car-
ried out at 10 °C for 16–18 h in darkness. After incubation,
the samples were extensively dialyzed against buffer B and
absorption spectra were recorded. In order to study the
incorporation of
35
S into the iron–sulfur cluster and there-
fore in the WhiB proteins, 30 lL samples were spotted onto
a poly(vinylidene difluoride) membrane, air dried and visu-
alized by phosphorimaging.

Mass spectrometry
Purified apo WhiB proteins (10 lm) were reduced with
1mm dithiothreitol for 1 h at 25 °C in buffer C (50 mm
Tris ⁄ HCl, pH 8.0, 200 mm NaCl). Free thiols were alkyl-
ated using 20 mm IAA for 1 h at 37 °C in darkness. The
protein without dithiothreitol treatment was alkylated with
IAA as a control. Protein samples were precipitated with
10% trichloroacetic acid for 15 min on ice, washed twice
with chilled acetone, air dried and re-suspended in 0.1% tri-
fluoroacetic acid. Protein (2 lL) was mixed with an equal
volume of the matrix (sinnapinic acid) and the mass was
measured in MALDI-TOF-MS, Voyager 4402 (Applied
Biosystems, Foster City, CA, USA).
CD spectroscopy
Far-UV CD spectra of apo proteins were recorded in a
JASCO J810 CD spectropolarimeter (Tokyo, Japan) at a
protein concentration of 0.2 mgÆmL
)1
in buffer C at 25 °C,
with a scan speed of 20 nmÆmin
)1
and 2 nm bandwidth.
Reduction was carried out by incubating the protein with
1mm dithiothreitol for 1 h at 25 °C in the same buffer.
Thermal denaturation of the oxidized and reduced apo pro-
teins (0.2 mgÆmL
)1
) in buffer C were monitored by measur-
ing the change in ellipticity at 222 nm with increasing
temperature from 25 to 95 °C at a speed of 2 °CÆmin

)1
.
Data were recorded at every 0.5 °C, with an 8 s averaging
time and at a 1.5 nm bandwidth. For all spectral measure-
ments, a path length of 1 mm was used. Ten spectra were
averaged for each sample.
Fluorescence spectroscopy
All fluorescence measurements were recorded at 25 °Cina
Perkin–Elmer Luminescence Spectrometer LS50B (Wal-
tham, MA, USA). To measure the surface hydrophobicities,
10 lm of oxidized apo proteins in buffer C was mixed with
ANS (250 lm) and incubated for 2 min at 25 °C. There-
after, the samples were excited at 388 nm and fluorescence
emission spectra were recorded in the range 400–600 nm.
The surface hydrophobicity (SH) was calculated using a
formula:
SH(AU) ¼
Maximum emission fluorescence intensity (AU)
X
where X ¼
total number of hydrophobic amino acids
total number of amino acids in the protein
:
To study the binding of ThT to different WhiB
proteins, 50 lm ThT was added to 5 lm apo proteins in
buffer C. The excitation was set at 430 nm and the emis-
sion was recorded between 420 and 600 nm. All spectra
were recorded at a scan speed of 50 nmÆmin
)1
. The

excitation and emission slit width was 3 nm for ANS flu-
orescence, whereas for ThT binding it was 5 and 10 nm
respectively. Ten spectra were averaged for each sample.
Throughout the study, the readings of solvent blanks
were subtracted from the sample spectra before plotting
the graphs.
Insulin disulfide reduction assay
WhiB proteins catalyzed reduction of insulin disulfides by
dithiothreitol was analyzed as described previously [23],
with some modifications. Three different concentrations 1,
2 and 3 lm of apo proteins were pre-incubated for 1 h in
reaction buffer D (0.1 m phosphate buffer pH 7.5, 200 mm
NaCl and 1 mm dithiothreitol). The reaction was started
by the addition of insulin to a final concentration of
0.13 mm. Precipitation of the reduced insulin chains was
monitored at 650 nm in a Perkin–Elmer (k35) spectropho-
tometer, every 0.1 min until A
650
reached to 0.8 or for
50 min.
Molecular properties of M. tuberculosis WhiB proteins Md. S. Alam et al.
90 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
Immunization of rabbit, collection of sera and
ELISA
New Zealand White rabbits were immunized with 250 lgof
purified WhiB proteins. Before immunization 5 mL blood
was collected for pre-immune sera. Equal volumes (500 lL)
of protein and incomplete Freund’s adjuvants were mixed
and an emulsion was prepared. The rabbits were immu-
nized subcutaneously. After 21 days, the first booster was

given and blood was collected after 5 days. The serum was
isolated and the titer was determined by ELISA. Use of
animals for the generation of polyclonal antibodies was
carried out with prior approval from the Institutional Ani-
mal Ethics Committee (institutional registration number
55 ⁄ 1999 ⁄ CPCSEA, dated 11 November 1999). Animal
handling and experimental design was performed in accor-
dance with the approved guidelines.
Each well of a flat-bottom microtitre ELISA plate
was coated with 0.5 lg of purified WhiB protein (100
llÆwell
)1
) in 0.05 m bicarbonate buffer, pH 9.6, and incu-
bated for overnight at 4 °C. After blocking, different dilu-
tions of test sera (1 : 5000 to 1 : 80 000) prepared in
NaCl ⁄ P
i
containing 0.1% skimmed milk were added
(100 lL) to each well and incubated at 37 °C for 3 h. Goat
anti-rabbit IgG (100 lL) conjugated to horseradish peroxi-
dase were diluted 1 : 5000 in NaC ⁄ P
i
containing 0.1%
skimmed milk, added to wells and incubated at 37 °C for
1 h. For the activity assay, 100 lL of peroxidase substrate,
3,3¢,5,5¢-tetramethylbenzidine (Bangalore Genei, Bangalore,
India) (diluted 1 : 20 in water), was added to the wells and
incubated at 37 °C for 30 min. The reaction was stopped
by adding 25 lLof1m H
2

SO
4
and A
450
was measured in
an ELISA reader (PowerWave XS Bio-Tek Instruments,
Inc, Winooski, VT, USA). The antibody dilution which
gave A
450
= 0.5 ± 0.05 was used for the cross-reactivity
assay.
Miscellaneous
UV–Vis absorption spectra of WhiB proteins (50 lm) puri-
fied under native, denaturing and in-column refolding
conditions were recorded in the range 200–700 nm, immedi-
ately after purification using a Perkin-Elmer (k35) spectro-
photometer at 25 °C where the elution buffer was used for
the baseline correction. The total amount of non-heme iron
was estimated by the o-bathophenanthroline disulfonate
method [59,60] as described elsewhere [15].
To remove the iron–sulfur cluster, purified proteins were
incubated with EDTA and potassium ferricyanide at a
molar ratio of 1 : 50 : 20 (protein : EDTA : ferricyanide) at
25 °C for 30 min. The chelated iron was removed by
dialysis against buffer D and was used as apo protein for
spectral studies and activity assays.
The molar extinction coefficient of various WhiB proteins
was determined using vector-nti software. The values were
as follows: 17 550 m
)1

Æcm
)1
for WhiB1, 13 140 m
)1
Æcm
)1
for
WhiB2, 18 830 m
)1
Æcm
)1
for WhiB3 and WhiB4,
16 980 m
)1
Æcm
)1
for WhiB5, 27 200 m
)1
Æcm
)1
for WhiB6 and
17 550 m
)1
Æcm
)1
for WhiB7. Throughout, the study protein
concentrations were calculated using A
280
reading.
Acknowledgements

MSA and SKG are thankful to the ‘Council of Scien-
tific and Industrial Research’ (CSIR), Govt of India,
for ‘Senior Research Fellowship’. Financial assistance
by CSIR and ‘Department of Biotechnology’ (DBT),
Govt of India, is duly acknowledged. This report is
IMTECH communication no. 08 ⁄ 2008. MSA, SKG
and PA designed the experiments. SKG carried out
some of the work on WhiB1. MSA carried out the rest
of the work. MSA and PA wrote the manuscript.
References
1 Smith I (2003) Mycobacterium tuberculosis pathogenesis
and molecular determinants of virulence. Clin Microbiol
Rev 16, 463–496.
2 Mehrotra J & Bishai WR (2001) Regulation of viru-
lence genes in Mycobacterium tuberculosis. Int J Med
Microbiol 291, 171–182.
3 Sacchettini JC, Rubin EJ & Freundlich JS (2008) Drugs
versus bugs: in pursuit of the persistent predator Myco-
bacterium tuberculosis. Nat Rev Microbiol 6, 41–52.
4 Gomez JE & Bishai WR (2000) whmD is an essential
mycobacterial gene required for proper septation and
cell division. Proc Natl Acad Sci USA 97, 8554–8559.
5 Steyn AJ, Collins DM, Hondalus MK, Jacobs WR Jr,
Kawakami RP & Bloom BR (2002) Mycobacte-
rium tuberculosis WhiB3 interacts with RpoV to affect
host survival but is dispensable for in vivo growth. Proc
Natl Acad Sci USA 99, 3147–3152.
6 Morris RP, Nguyen L, Gatfield J, Visconti K, Nguyen
K, Schnappinger D, Ehrt S, Liu Y, Heifets L, Pieters J
et al. (2005) Ancestral antibiotic resistance in Mycobac-

terium tuberculosis. Proc Natl Acad Sci USA 102 ,
12200–12205.
7 Raghunand TR & Bishai WR (2006) Mapping essential
domains of Mycobacterium smegmatis WhmD: insights
into WhiB structure and function. J Bacteriol 188,
6966–6976.
8 Geiman DE, Raghunand TR, Agarwal N & Bishai WR
(2006) Differential gene expression in response to expo-
sure to antimycobacterial agents and other stress condi-
tions among seven Mycobacterium tuberculosis whiB-like
genes. Antimicrob Agents Chemother 50, 2836–2841.
9 Singh A, Guidry L, Narasimhulu KV, Mai D, Tromb-
ley J, Redding KE, Giles GI, Lancaster JR Jr & Steyn
AJ (2007) Mycobacterium tuberculosis WhiB3 responds
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 91
to O
2
and nitric oxide via its [4Fe–4S] cluster and is
essential for nutrient starvation survival. Proc Natl
Acad Sci USA 104, 11562–11567.
10 Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C,
Harris D, Gordon SV, Eiglmeier K, Gas B, Barry CE
3rd et al. (1998) Deciphering the biology of Mycobac-
terium tuberculosis from the complete genome sequence.
Nature 393, 537–544.
11 Camus JC, Pryor MJ, Medigue C & Cole ST (2002)
Re-annotation of the genome sequence of Mycobacte-
rium tuberculosis H37Rv. Microbiology 148, 2967–
2973.

12 Davis NK & Chater KF (1992) The Streptomyces coeli-
color whiB gene encodes a small transcription factor-like
protein dispensable for growth but essential for sporula-
tion. Mol Gen Genet 232, 351–358.
13 Garg SK, Alam MS, Soni V, Radha Kishan KV &
Agrawal P (2007) Characterization of Mycobacte-
rium tuberculosis WhiB1 ⁄ Rv3219 as a protein disulfide
reductase. Protein Expr Purif 52, 422–432.
14 Alam MS & Agrawal P (2008) Matrix-assisted refold-
ing and redox properties of WhiB3 ⁄ Rv3416 of Myco-
bacterium tuberculosis H37Rv. Protein Expr Purif 61,
83–91.
15 Alam MS, Garg SK & Agrawal P (2007) Molecular
function of WhiB4 ⁄ Rv3681c of Mycobacterium tubercu-
losis H37Rv: a [4Fe–4S] cluster co-ordinating protein
disulphide reductase. Mol Microbiol 63, 1414–1431.
16 Soliveri JA, Gomez J, Bishai WR & Chater KF (2000)
Multiple paralogous genes related to the Streptomy-
ces coelicolor developmental regulatory gene whiB are
present in Streptomyces and other actinomycetes.
Microbiology 146, 333–343.
17 Ritz D & Beckwith J (2001) Roles of thiol-redox path-
ways in bacteria. Annu Rev Microbiol 55, 21–48.
18 Paget MS & Buttner MJ (2003) Thiol-based regulatory
switches. Annu Rev Genet 37, 91–121.
19 Romero-Isart N & Vasak M (2002) Advances in the
structure and chemistry of metallothioneins. J Inorg
Biochem 88, 388–396.
20 Orme-Johnson WH & Orme-Johnson NR (1982) The
problem of determining cluster type. In Iron–Sulfur Pro-

teins (Spiro TG, ed.), pp. 67–96. Wiley-Interscience,
New York, NY.
21 Paraskevopoulou C, Fairhurst SA, Lowe DJ, Brick P &
Onesti S (2006) The Elongator subunit Elp3 contains a
Fe4S4 cluster and binds S-adenosylmethionine. Mol
Microbiol 59, 795–806.
22 Bertini I, Cowan JA, Luchinat C, Natarajan K &
Piccioli M (1997) Characterization of a partially
unfolded high potential iron protein. Biochemistry 36,
9332–9339.
23 Loferer H, Wunderlich M, Hennecke H & Glockshuber
R (1995) A bacterial thioredoxin-like protein that is
exposed to the periplasm has redox properties compara-
ble with those of cytoplasmic thioredoxins. J Biol Chem
270, 26178–26183.
24 Holmgren A (1979) Thioredoxin catalyzes the reduction
of insulin disulfides by dithiothreitol and dihydrolipoa-
mide. J Biol Chem 254, 9627–9632.
25 Shi Y, Tang W, Hao SF & Wang CC (2005) Contribu-
tions of cysteine residues in Zn
2
to zinc fingers and
thiol-disulfide oxidoreductase activities of chaperone
DnaJ. Biochemistry 44, 1683–1689.
26 Holmgren A (1985) Thioredoxin. Annu Rev Biochem 54,
237–271.
27 LeVine H III (1993) Thioflavine T interaction with syn-
thetic Alzheimer’s disease beta-amyloid peptides: detec-
tion of amyloid aggregation in solution. Protein Sci 2,
404–410.

28 Lewandowska A, Matuszewska M & Liberek K (2007)
Conformational properties of aggregated polypeptides
determine ClpB-dependence in the disaggregation pro-
cess. J Mol Biol 371, 800–811.
29 Beinert H (2000) Iron-sulfur proteins: ancient struc-
tures, still full of surprises. J Biol Inorg Chem 5, 2–15.
30 Johnson DC, Dean DR, Smith AD & Johnson MK
(2005) Structure, function, and formation of biological
iron–sulfur clusters. Annu Rev Biochem 74, 247–281.
31 Beinert H, Holm RH & Munck E (1997) Iron-sulfur
clusters: nature’s modular, multipurpose structures.
Science 277, 653–659.
32 Hidalgo E, Ding H & Demple B (1997) Redox signal
transduction via iron–sulfur clusters in the SoxR tran-
scription activator. Trends Biochem Sci 22, 207–210.
33 Kiley PJ & Beinert H (1998) Oxygen sensing by the
global regulator, FNR: the role of the iron–sulfur
cluster. FEMS Microbiol Rev 22, 341–352.
34 Berndt C, Lillig CH, Wollenberg M, Bill E, Mansilla
MC, de Mendoza D, Seidler A & Schwenn JD (2004)
Characterization and reconstitution of a 4Fe–4S ade-
nylyl sulfate ⁄ phosphoadenylyl sulfate reductase from
Bacillus subtilis. J Biol Chem 279, 7850–7855.
35 Jakimowicz P, Cheesman MR, Bishai WR, Chater KF,
Thomson AJ & Buttner MJ (2005) Evidence that the
Streptomyces developmental protein WhiD, a member
of the WhiB family, binds a [4Fe–4S] cluster. J Biol
Chem 280, 8309–8315.
36 Betts JC, Lukey PT, Robb LC, McAdam RA & Dun-
can K (2002) Evaluation of a nutrient starvation model

of Mycobacterium tuberculosis persistence by gene and
protein expression profiling. Mol Microbiol 43, 717–731.
37 Ramakrishnan L, Federspiel NA & Falkow S (2000)
Granuloma-specific expression of Mycobacterium viru-
lence proteins from the glycine-rich PE-PGRS family.
Science 288, 1436–1439.
38 Fernando MR, Nanri H, Yoshitake S, Nagata-Kuno K
& Minakami S (1992) Thioredoxin regenerates proteins
inactivated by oxidative stress in endothelial cells. Eur J
Biochem 209, 917–922.
Molecular properties of M. tuberculosis WhiB proteins Md. S. Alam et al.
92 FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS
39 Nakamura H, Nakamura K & Yodoi J (1997) Redox
regulation of cellular activation. Annu Rev Immunol 15,
351–369.
40 Nishinaka Y, Masutani H, Nakamura H & Yodoi J
(2001) Regulatory roles of thioredoxin in oxidative
stress-induced cellular responses. Redox Rep 6, 289–295.
41 Holmgren A (1989) Thioredoxin and glutaredoxin sys-
tems. J Biol Chem 264, 13963–13966.
42 Arner ES & Holmgren A (2000) Physiological functions
of thioredoxin and thioredoxin reductase. Eur J Bio-
chem 267, 6102–6109.
43 Newton GL, Arnold K, Price MS, Sherrill C, Delcard-
ayre SB, Aharonowitz Y, Cohen G, Davies J, Fahey
RC & Davis C (1996) Distribution of thiols in microor-
ganisms: mycothiol is a major thiol in most actinomyce-
tes. J Bacteriol 178, 1990–1995.
44 Patel MP & Blanchard JS (1999) Expression, purifica-
tion, and characterization of Mycobacterium tuber-

culosis mycothione reductase. Biochemistry 38,
11827–11833.
45 Zahrt TC & Deretic V (2002) Reactive nitrogen and
oxygen intermediates and bacterial defenses: unusual
adaptations in Mycobacterium tuberculosis. Antioxid
Redox Signal 4, 141–159.
46 Chivers PT, Prehoda KE & Raines RT (1997) The
CXXC motif: a rheostat in the active site. Biochemistry
36, 4061–4066.
47 Quan S, Schneider I, Pan J, Von Hacht A & Bardwell
JC (2007) The CXXC motif is more than a redox rheo-
stat. J Biol Chem 282, 28823–28833.
48 Soderberg BO, Sjoberg BM, Sonnerstam U & Branden
CI (1978) Three-dimensional structure of thioredoxin
induced by bacteriophage T4. Proc Natl Acad Sci USA
75, 5827–5830.
49 Hall G, Shah M, McEwan PA, Laughton C, Stevens
M, Westwell A & Emsley J (2006) Structure of Myco-
bacterium tuberculosis thioredoxin C. Acta Crystallogr
D Biol Crystallogr 62, 1453–1457.
50 Bao R, Chen Y, Tang YJ, Janin J & Zhou CZ (2007)
Crystal structure of the yeast cytoplasmic thioredoxin
Trx2. Proteins 66, 246–249.
51 Eklund H, Gleason FK & Holmgren A (1991) Struc-
tural and functional relations among thioredoxins of
different species. Proteins 11 , 13–28.
52 Kumar JK, Tabor S & Richardson CC (2004) Proteo-
mic analysis of thioredoxin-targeted proteins in
Escherichia coli. Proc Natl Acad Sci USA 101, 3759–
3764.

53 Balmer Y, Koller A, del Val G, Manieri W, Schurmann
P & Buchanan BB (2003) Proteomics gives insight into
the regulatory function of chloroplast thioredoxins.
Proc Natl Acad Sci USA 100, 370–375.
54 Hayashi T, Ueno Y & Okamoto T (1993) Oxidoreduc-
tive regulation of nuclear factor kappa B. Involvement
of a cellular reducing catalyst thioredoxin. J Biol Chem
268, 11380–11388.
55 Schenk H, Klein M, Erdbrugger W, Droge W & Schu-
lze-Osthoff K (1994) Distinct effects of thioredoxin and
antioxidants on the activation of transcription factors
NF-kappa B and AP-1. Proc Natl Acad Sci USA 91,
1672–1676.
56 Kiley PJ & Beinert H (2003) The role of Fe–S proteins
in sensing and regulation in bacteria. Curr Opin Micro-
biol 6, 181–185.
57 Raghava GP, Solanki RJ, Soni V & Agrawal P (2000)
Fingerprinting method for phylogenetic classification
and identification of microorganisms based on variation
in 16S rRNA gene sequences. Biotechniques 29, 108–
112.
58 Sambrook J & Russell DW (2001) Molecular Cloning:
A Laboratory Manual, 3rd edn. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY.
59 Plank DW, Kennedy MC, Beinert H & Howard JB
(1989) Cysteine labeling studies of beef heart aconitase
containing a 4Fe, a cubane 3Fe, or a linear 3Fe cluster.
J Biol Chem 264, 20385–20393.
60 Carroll KS, Gao H, Chen H, Leary JA & Bertozzi CR
(2005) Investigation of the iron–sulfur cluster in Myco-

bacterium tuberculosis APS reductase: implications for
substrate binding and catalysis. Biochemistry 44, 14647–
14657.
Supporting information
The following supplementary material is available:
Fig. S1. Purification of WhiB proteins.
Fig. S2. Brownish color of WhiB proteins purified
under different conditions.
Fig. S3. UV–Vis spectra of WhiB proteins purified
under denaturing or in-column refolding conditions.
Fig. S4. Effect of air oxidation on the stability of
[2Fe–2S] cluster of various WhiB proteins.
Fig. S5. Insulin disulfide reduction assay of apoWhiB5,
WhiB6 and WhiB7.
Fig. S6. Comparison of the amino acid sequence at
CXXC motif.
Table S1. Percentage identity and similarity among
WhiB proteins of M. tuberculosis.
Table S2. List of primers used in this study.
This supplementary material can be found in the
online version of this article.
Please note: Wiley-Blackwell is not responsible for
the content or functionality of any supplementary
materials supplied by the authors. Any queries (other
than missing material) should be directed to the corre-
sponding author for the article.
Md. S. Alam et al. Molecular properties of M. tuberculosis WhiB proteins
FEBS Journal 276 (2009) 76–93 ª 2008 The Authors Journal compilation ª 2008 FEBS 93