Content uploaded by Min-Kyu Kwak
Author content
All content in this area was uploaded by Min-Kyu Kwak on Jul 09, 2018
Content may be subject to copyright.
Research Article
Anti-σfactor YlaD regulates transcriptional activity
of σfactor YlaC and sporulation via manganese-
dependent redox-sensing molecular switch in
Bacillus subtilis
Min-Kyu Kwak
1,
*
,†
, Han-Bong Ryu
1,
*, Sung-Hyun Song
1,
*, Jin-Won Lee
2
and Sa-Ouk Kang
1,‡
1
Laboratory of Biophysics, School of Biological Sciences, and Institute of Microbiology, Seoul National University, Seoul 151-742, Republic of Korea;
2
Department of Life Science
and Research Institute for Natural Sciences, Hanyang University, Seoul 04763, Republic of Korea
Correspondence: Jin-Won Lee ( jwl@hanyang.ac.kr) or Sa-Ouk Kang (kangsaou@snu.ac.kr)
YlaD, a membrane-anchored anti-sigma (σ) factor of Bacillus subtilis, contains a
HX
3
CXXC motif that functions as a redox-sensing domain and belongs to one of the zinc
(Zn)-co-ordinated anti-σfactor families. Despite previously showing that the YlaC tran-
scription is controlled by YlaD, experimental evidence of how the YlaC–YlaD interaction
is affected by active cysteines and/or metal ions is lacking. Here, we showed that the P
yla
promoter is autoregulated solely by YlaC. Moreover, reduced YlaD contained Zn and iron,
while oxidized YlaD did not. Cysteine substitution in YlaD led to changes in its secondary
structure; Cys3 had important structural functions in YlaD, and its mutation caused dis-
sociation from YlaC, indicating the essential requirement of a HX
3
CXXC motif for regulat-
ing interactions of YlaC with YlaD. Analyses of the far-UV CD spectrum and metal
content revealed that the addition of Mn ions to Zn–YlaD changed its secondary structure
and that iron was substituted for manganese (Mn). The ylaC gene expression using βGlu
activity from P
yla
:gusA was observed at the late-exponential and early-stationary phase,
and the ylaC-overexpressing mutant constitutively expressed gene transcripts of clpP
and sigH, an important alternative σfactor regulated by ClpXP. Collectively, our data
demonstrated that YlaD senses redox changes and elicits increase in Mn ion concentra-
tions and that, in turn, YlaD-mediated transcriptional activity of YlaC regulates sporulation
initiation under oxidative stress and Mn-substituted conditions by regulating clpP gene
transcripts. This is the first report of the involvement of oxidative stress-responsive
B. subtilis extracytoplasmic function σfactors during sporulation via a Mn-dependent
redox-sensing molecular switch.
Introduction
Bacillus subtilis is an outstanding model organism for understanding sigma (σ) factor-associated
Gram-positive bacterial physiology, including growth/division, redox control, sporulation, and/or
metal ion regulation [1–7]. These interests have partially converged around the well-characterized
process of inherent transcriptional gene expression required for various stress responses, with diverse
mechanisms of six and four types of vegetative- and sporulation-specificσfactor activities [8]. In this
context, the sensing mechanisms of reactive oxygen species (ROS) in bacteria are concomitantly
related to cellular redox regulation via oxidative stress-responsive transcription factors, such as OxyR,
OhrR, and Hsp33 proteins, having redox-active cysteines with or without zinc (Zn)-co-ordinated
ligands [9–11]. B. subtilis also possesses well-established oxidative stress-response sensors, including
some σand anti-σfactors, for adapting to peroxide and/or disulfide stress conditions, which trigger
*These authors equally
contributed to this work.
†
Present address: Division of
Paideia, Sungkyul University,
Gyeonggi-do 14097, Republic
of Korea.
‡
Present address: Irwee
Institute, Research Park 940-
521, Gwanak-ro 1, Gwanak-gu,
Seoul National University, Seoul
151-742, Republic of Korea.
Accepted Manuscript online:
14 May 2018
Version of Record published:
5 July 2018
Received: 1 December 2017
Revised: 29 April 2018
Accepted: 14 May 2018
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2127
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
redox-sensitive regulators [i.e. proteins with specific sensory functions, such as the transcriptional repressor
PerR, which senses hydrogen peroxide (H
2
O
2
) by metal-catalyzed histidine oxidation [12], the transcriptional
repressor OhrR, involved in organic hydroperoxide resistance [13], and proteins that become active under
diverse stress conditions, such as a redox controlled RNA polymerase (RNAP)-binding Spx under thiol stress
and general stress RNAP σsubunit σ
B
[14]]. One of the most noticeable responses to a wide range of oxidative
stresses of the global regulators in this strain is their systematic linking to each other and the drastic induction
of both stress-responsive detoxification enzymes and sporulation [3,15], which are necessarily associated with
and co-regulated by sequential expression of growth phase-specificσfactors [16]. For instance, for peroxide
stresses, the adaptive responses are mainly under the control of three proteins, σ
B
, PerR, and OhrR, in the vege-
tative and/or stationary phase [13].
Importantly, the expression of these regulators during oxidative stress is known to affect metal ion homeosta-
sis [5,7,16–18]. Many metalloproteins are regulated by intracellular metal levels, best exemplified by Fur in
Escherichia coli, which has been described as an iron-responsive repressor of iron-transport systems [19,20], and
PerR from B.subtilis, which senses oxidative stress [21]. Despite our understanding of B. subtilis Fur, MntR,
PerR, TnrA, and σB in co-ordinating iron and manganese (Mn) homeostasis [7], the effect of metal imbalances
is another necessary but poorly understood mechanism, especially of bacterial σand anti-σfactor/physiology.
Zn, iron, and Mn depravation stops growth, but the effect on this on σand anti-σfactor activity is elusive. Both
predominantly present divalent cations, Mn and iron, donate the cytoplasmic reducing equivalent at ∼0.5 mM
[22–24]. Additionally, because Zn-binding enhances the reactivity of cysteine by stabilizing the negative charge
on the thiolate anion and lowering its pK
a
value, this leads to a more active cysteine against H
2
O
2
at a neutral
pH [5,25]. The Zn released by redox changes might have a significant biological relevance as a cellular antioxi-
dant. For instance, the intracellular accumulation as a result of elevated Mn levels commonly inhibits the activity
of metal-dependent proteins, such as iron-co-ordinated E. coli ferrochelatase [26] and Zn-containing anti-σ
factor RsrA, and significantly induces a generalized stress response by the release of σ
B
from its anti-σfactor
RsbW with limited cofactor availability of σ
B
-associated RsbU activity during σ
B
-dependent general stress
response [17]. To our knowledge, if upon exposure to environmental oxidants such as ROS or diamide, the
active cysteines are quickly oxidized and Zn is also released [27,28], this causes the dissociation of the structural
Zn site with all its conformational consequences, which explains how the oxidative modification can lead to
such dramatic functional changes in an entire protein. The well-conserved RsrA–σ
R
interaction prevents binding
of σ
R
to core RNA polymerase. However, during disulfide stress, Zn is released. These phenomena cause con-
formational rearrangements in RsrA, which lead to the dissociation of σ
R
[27,28]. Thus, the effect of metal sub-
stitution by environmental stresses in these regulators, especially σand anti-σfactors, related to specific metal
efflux (i.e. Mn
2+
), is significantly important to assess bacterial physiology.
RNAP transcriptional activity can also be altered by promoter specificity and environmental stresses, which
define whether RNAP binds to the primary σsubunit or alternative σfactors [29]. The extracytoplasmic func-
tion (ECF) family of σfactors are a group of alternative σfactors from various bacteria [30]. Similar to σ
factors interacting with RNAP, they also regulate gene expression in response to various extracytoplasmic
stimuli [30], which are sensed by ECF σfactors consisting of a structurally unique group of proteins against
oxidative stress, high salt, and heat [31]. Whereas 19 σfactors have been reported in the B. subtilis genome,
several kinds of ECF σfactors have been identified, among them seven ECF σfactors [32,33]. The cellular roles
of these seven ECF σfactors encoded by B. subtilis show distinctly different characteristics of ECF function in a
Gram-positive model organism [33]. The σ
M
[34], σ
W
,σ
X
[31], and σ
V
factors [35] are known to mainly par-
ticipate in cell envelope stress maintenance [36]. The activity of σ
W
, a ZAS-family member, known to contain a
conserved HX
3
CXXC motif forming disulfide bonds [27,37], is induced by alkaline stress [38] and salt stress
[39], rather than by oxidants [6]. In contrast, the functions of the other three ECF σfactors, including σ
Y
,σ
Z
,
and σ
YlaC
, are not clear, similar to the ECF σfactors of other microbial species, including Corynebacterium,
Mycobacterium,Nocardia, and Rhodococcus [33]. The poorly studied ECF σfactor, σ
Z
, encoded by the sigZ
gene, is monocistronic, but its anti-σfactor is unknown. The σ
Z
factor does not seem to regulate many genes,
but participates in regulating yrpG, which is activated by this σfactor [40].
InthecaseofYlaCinB. subtilis,thisσfactor is encoded by ylaC, which is located upstream of the ylaD gene,
and its product is believed to be the anti-YlaC factor in the ylaABCD operon by transcriptional analysis [33,41,42].
However, the ylaABCD operon is also poorly understood and, moreover, the function of YlaA and YlaB is elusive.
Although the ECF σfactor YlaC, similar to σ
Y
and σ
Z
, has not been sufficiently investigated [33], B. subtilis resist-
ance to oxidative stress through this σfactor has been partially reported elsewhere [41,42]. This clue comes from
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2128
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
the fact that, from a structural point of view, YlaD is very similar to the anti-σ
R
factor of Streptomyces coelicolor
RsrA, with shared cysteine residues that can sense the cytoplasmic thiol–disulfide state [33].
Along with its structural and functional information, by transcriptional analysis of the ylaABCD operon
encoding an σfactor of extracytoplasmic function family [41], we previously reported that YlaC activity
enhances peroxidase activity and contributes to oxidative stress resistance in the ylaC-overexpressing mutant,
which is significantly less sensitive to oxidative stresses (i.e. diamide and H
2
O
2
)[42]. This was supported by
experiments using the GUS reporter gene expressed by the yla operon promoter, which was simultaneously
induced by exogenous H
2
O
2
treatment. Immunoblotting analyses also showed the increased ylaC transcript and
YlaC protein expression via the treatment of exogenous H
2
O
2
; the ylaC-overexpressing and ylaD-disrupted
mutants exhibited higher sporulation rates, in contrast with those of ylaC-disrupted and ylaD-overexpressing
mutants. Importantly, the interaction between YlaC and YlaD was shown using native-polyacrylamide gel elec-
trophoresis (PAGE), by the recombinant proteins YlaC and controlled redox state of YlaD, hypothesizing YlaC
transcription regulation by the YlaD redox state and its possible contribution to oxidative stress resistance in
B. subtilis. However, even though a new ECF σfactor–anti-σfactor pair was suggested, little is known about
the molecular mechanism of regulation of this ECF σfactor by its anti-σfactor. This is particularly true in
terms of how anti-σfactor YlaD receives and senses signals and what type of stress response they may induce,
but also whether the substitution effect of metal ions and/or active cysteines alters the redox homeostasis based
on proposed roles of HX3CXXC motif at the N-terminal domain of YlaD, which is expected to serve as a
Zn-binding site to participate in the formation of disulfide bonds.
Based on the transcriptional analysis and oxidative stress response from previous studies [41,42], we
hypothesized that YlaC and YlaD are ECF σfactor and cognate anti-σfactor of YlaC, respectively, and that
YlaD contains HX3CXXC motif that can sense the redox state as well as the metal-binding site at N-terminal
region. YlaC was expected to be regulated by YlaD and the YlaC–YlaD interaction expected to be related to
redox regulation under oxidative stress conditions and under metal ion control, in contrast with other B. subti-
lis ECF σfactors. Therefore, we intend to explain in detail how YlaD regulates YlaC by recognizing changes in
redox state and metal ions (i.e. Zn, iron, and Mn). Following this, the release of YlaC from YlaD contributes to
sporulation by the induction of clpP in B.subtilis.
Here, we biophysically evidence the YlaD–YlaC interaction by observing that the YlaD HX3CXXC motif acts
as an oxidative stress-sensing domain, using H
2
O
2
and controlled redox state of YlaD, involved in regulating the
transcriptional activity of YlaC. We revealed for the first time that oxidized YlaD can be released from the
YlaC–YlaD complex; free YlaC auto-regulates its own promoter Pyla. Importantly, we showed that only reduced
YlaD contains Zn and iron, cysteine substitution in YlaD changes the secondary structure, and Cys3 mutation
causes dissociation from YlaC. Most importantly, Mn ion addition to Zn–YlaD significantly changed its second-
ary structure and a YlaC-overexpressing mutant constitutively expressed sigH and clpP gene transcripts. Based on
our data, YlaD senses both redox changes and increasing Mn ion concentrations and that, in turn, YlaC, whose
transcriptional activity is regulated by YlaD, contributes to regulation of sporulation initiation, both under
oxidative stress conditions and under metal ion substitution during development by modulating of clpP gene
expression. Collectively, we demonstrated that the redox state of YlaD determines the transcriptional activity of
YlaC, evidenced by the insertion of Mn into its Zn- and/or iron-binding site. We thus present new information
regarding the initiation of sporulation from the redox changes of Zn-co-ordinated YlaD in B. subtilis.
Experimental
Strains and plasmids
The bacterial strains and plasmids [42] used in this study are listed in Table 1. For DNA manipulation and
YlaC and YlaD overproduction, E.coli strains DH5αand BL21 were used as hosts, using pET (Novagen, U.S.
A.) and glutathione S-transferase (GST) gene fusion systems (GE Healthcare), respectively.
Overproduction and purification of YlaD
N
and mutated YlaD
N
YlaC and a YlaD variant with the C-terminal membrane helix domain (GLLIMKAACWFGAAVAMMLIIKLLI)
removed were expressed and purified from E.coli BL21 as described previously [42]. Primers were used to
change the cysteine residues to serine (the mutated codons are shown in bold and restriction enzyme sites are
underlined. (Table 1)). Additionally, the ylaDE:pGEM T easy vector was used as a template for site-directed
mutagenesis.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2129
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
Table 1 Strains and primers used in the present study.
Strain, primer, or
plasmid Genotype, sequence, or description Source or reference
B. subtilis
PS832 Wild type (Trp
+
revertant of B. subtilis 168) Peter Setlow
HB001 PS832 derivative (control strain) containing pRB374 The present study
HBC01 PS832 derivative overexpressing the ylaC gene The present study
HBC02 PS832 derivative expressing a nonfunctional ylaCD
gene
The present study
HBD01 PS832 derivative overexpressing the ylaD gene Ryu et al.[42]
HBD02 PS832 derivative expressing a nonfunctional ylaD gene Ryu et al.[42]
HBP01 PS832 derivative (control strain) containing pMLK83 for
β-Gul activity
The present study
HBP02 PS832 derivative containing the pMLK83:Pyla
promoter for β-Gul activity
The present study
E. coli
DH5αF
−
ΔlacU169(
f
80lacZΔM15)endA1 rec1hsdR17 deoR
supE44 thi-1 λ
-
gyrA96 relA1
[76]
BL21(DE3)pLysS F
-
ompTr
B
-m
B
-(DE)/pLysS [77]
JM105 supE endA sbcB15 hsdR4 rpsL thi Δ(lac-proAB)[78]
HBC1 Strain overproducing YlaC The present study
HBDN1 Strain overproducing YlaD
N
The present study
HBDN13S Strain overproducing mutated YlaD
N
(Cys
3
→Ser) The present study
HBDN118S Strain overproducing mutated YlaD
N
(Cys
18
→Ser) The present study
HBDN133S Strain overproducing mutated YlaD
N
(Cys
33
→Ser) The present study
HBDN136S Strain overproducing mutated YlaD
N
(Cys
36
→Ser) The present study
Primers
ylaCEOF 50-GTGGTTTTATTTCATATGAAGCATAGG-30NdeI, for overproducing YlaC in E. coli
ylaCEOR 50-GCAGGTCTCTCTCGAGAAAGCTCATCATATCCG-30XhoI, for overproducing YlaC in E. coli
ylaCBOF 50-GGCCATTTGCATGCCGGTGTGG-30SphI, for overproducing YlaC in B. subtilis
ylaCBOR 50-TTTCTAGAAAGCAGGTCATATC-30XbaI, for overproducing YlaC in B. subtilis
ylaCBDF 50-TATTCAGAAGGATCCATCAT-30BamHI, for expression of nonfunctional ylaC
gene in B. subtilis
ylaCBDR 50-TTAACTTCAGGAGCTCCTGC-30SacI, for expression of nonfunctional ylaC
gene in B. subtilis
ylaDEOF 50-GAGGAGGATCCGATATGACCTGCTTTC-30BamHI, for overproducing YlaD
N
in E. coli
ylaDEOR 50-GCAGGCAGCTCTCGAGATCAGTCATCAATAGTA-30XhoI, for overproducing YlaD
N
in E. coli
ylaDBOF 50-TAAAGCTTAAGAAAAACATGAC-30HindIII, for overproducing YlaD in B. subtilis
ylaDBOR 50-CTTGCCGGATCCGGACAAGCGC-30BamHI, for overproducing YlaD in B. subtilis
DC3SF 50-TCCGATATGACCAGCTTTCTAGTAAGAGAC-30The present study
DC3SR 50-GTCTCTTACTAGAAAGCTGGTCATATCGGA-30The present study
DC18SF 50-TATCTTGAAGGTGATAGCAAAAGAGAAACG-30The present study
DC18SR 50-CGTTTCTCTTTTGCTATCACCTTCAAGATA-30The present study
DC33SF 50-GAGCATTTAAAAATGAGCAGCAGCTGCAGA-30The present study
DC33SR 50-TCTGCAGCTGCTGCTCATTTTTAAATGCTC-30The present study
DC36SF 50-AAAATGTGCAGCAGCAGCAGAGACATGTAT-30The present study
DC36SR 50-ATACATGTCTCTGCTGCTGCTGCACATTTT-30The present study
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2130
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
To overexpress the ylaD gene in E. coli, the gene was amplified by PCR with ylaDEOF and ylaDEOR
primers. The resulting 260 bp fragment was cloned into a pGEM-T easy vector for DNA sequencing. After
BamHI/XhoI digestion, the fragment was cloned into a pGEX-4T-1 vector.
To overproduce YlaD
N
,E. coli BL21 cells were transformed with a gene product obtained by BamHI/XhoI
digestion and grown on an LB agar plate containing 50 mg/ml of ampicillin and 34 mg/ml of chlorampenicol at
37°C for 16 h. Cells were induced by adding 1 mM isopropylthio-β-D-galactoside (IPTG) and were harvested after
further incubation at 22°C for 18 h. The resulting cell lysate was applied to a Glutathione Sepharose 4 fast flow
column. To remove the GST tag, the eluted proteins were dialyzed in PBS buffer and treated with thrombin
according to the manufacturer’s instructions. The cleaved YlaD
N
was further purified through a Glutathione
Sepharose 4 fast flow column, a Superdex 75 column, and a Superose 12 column. The purified YlaD
N
was dia-
lyzed in storage buffer (20 mM Tris–HCl, pH 8.0, 20% glycerol, and 200 mM NaCl) and stored at −70°C.
To exchange cysteine to serine in YlaD
N
, a pGEM-T easy:ylaD
N
plasmid was used as a template. The
primers used for site-directed mutagenesis are listed in Table 1.
For site-directed mutagenesis, the QuikChange site-directed mutagenesis kit (Stratagene) was used to switch
amino acid. The oligonucleotide primers, each complementary to opposite strands of template, were extended
during 15 cycles of amplification using PfuTurbo DNA polymerase. Each 25 ml reaction mixture contained the
following: 2.5 ml of 10× reaction buffer, 25 ng of template, each 50 ng of mutagenic primers, 1 ml of dNTP
mixture, and 0.5 mlofPfuTurbo DNA polymerase (2.5 U/ml). Following the PCR, the product was treated with
the DpnI endonuclease for 1 h. The DpnI endonuclease is specific for methylated and hemimethylated DNA
and was used to digest the parental DNA template and to select for mutation-containing synthesized DNA.
The PCR-amplified DNA containing the desired mutations was then transformed into E. coli DH5αand
transformants were selected on an LB agar containing 50 mg/ml ampicillin. Following colony selection, over-
production and purification of mutated YlaD
N
was carried out as the previously described method of
overproduction and purification of YlaD
N
.
In vitro transcription assay
RNA polymerase was purified from B. subtilis for use in in vitro transcription assays. All purification steps were
performed at 4°C and the in vitro transcription assays were performed as previously described [43]. Specifically,
an in vitro transcription assay of the P
yla
promoter with RNAP in complex with YlaC was performed as
follows. YlaD was preincubated with 1 mM DTT and 100 mM ZnCl
2
at 37°C for 30 min. Subsequently, YlaC
(30 nM) was added, and the solution was incubated at 37°C for 10 min, after which RNAP was added to the
reaction mixture. YlaD was analyzed at molar ratios of 0, 2, 4, 8, and 16 versus YlaC. The P
yla
promoter was
used as a template for the in vitro transcription assay, and RNAP and the template were added to the reaction
mixture at the indicated time.
Modification of YlaD with 4-acetamido-40-maleimidylstilbene-2,20-disulfonic
acid and iodoacetamide, followed by matrix-assisted laser desorption/
ionization time-of-flight mass spectroscopy
The purified YlaD protein was dialyzed against TNDG
20
and then incubated with H
2
O
2
and diamide (YlaD
concentration, 34 mM). After incubation at 25°C for 20 min, 20% trichloroacetic acid (final concentration) was
added to end the reaction. The protein was recovered by precipitation with 20% TCA and modified by incuba-
tion with buffer containing 15 mM 4-acetamido-40-maleimidylstilbene-2,20-disulfonic acid (AMS) or 50 mM
iodoacetamide (IAA). The mass of protein treated with IAA was analyzed by matrix-assisted laser desorption/
ionization time-of-flight (MALDI-TOF) mass spectroscopy.
Two-dimensional polyacrylamide gel electrophoresis
Crude cell extracts from the wild type and mutants, including PS832, HBC01, and HBC02 cells (350 mg), were
separated using Immobiline
™
DryStrip gels (pH 4–7, 13 cm; GE Healthcare) for first-dimension isoelectric
focusing, as recommended by the manufacturer. Gel strips were then equilibrated in SDS equilibration buffer
(50 mM Tris–HCl [ pH 8.8], 6 M urea, 30% glycerol, and 2% SDS) and subjected to second-dimension electro-
phoresis on a 12% SDS–polyacrylamide gel. Following electrophoresis, gels were stained with Coomasie
Brilliant Blue R-250 (CBBR-250).
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2131
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
Native PAGE analysis of YlaC–YlaD interaction
YlaC and YlaD proteins in TNDG
20
buffer were incubated at 37°C in the presence of 1 mM DTT or 2.5 mM
H
2
O
2
for 30 min. The samples were electrophoretically separated on a native 12% polyacrylamide gel at 4 mA
for 12–13 h. The running gel and buffer included 2 and 10 mM DTT, respectively [28,44].
A mobility shift assay by native PAGE using YlaD (2 mg) after the modification with AMS was performed as
follows. To obtain the reduced form of YlaD, it was incubated with 1 mM DTT at room temperature for the
indicated time. To generate the oxidized form of YlaD, its reduced form was allowed to react with 1 mM H
2
O
2
and 1 mM diamide at room temperature for the indicated time. To stop the reactions, a final concentration of
20% TCA was added to the reaction mixtures, and the samples were precipitated. Then, 15 mM AMS was
added to the reaction mixtures. Each sample was resolved by electrophoresis on a 16% non-reducing Tricine
polyacrylamide gel; C, purified YlaD (control) + 15 mM AMS; Oxi, oxidized form; Red, reduced form. The P
yla
promoter was used as a template for the in vitro transcription assay, and RNAP and the template were added
to the reaction mixture at the indicated time. C, the reduced form of YlaD obtained by treatment with 1 mM
DTT at room temperature for 30 min.
To monitor the effect of metal ions in the YlaC–YlaD complex, after YlaD (4 or 8 mM) was preincubated
with TNDG
20
buffer containing 16 mM ZnCl
2
, YlaC was added. Then, the samples were treated with 16 mM
MnCl
2
. The samples were resolved by electrophoresis on a 12% native PAGE gel and a running gel, whose
buffers contained 2 and 10 mM DTT, respectively. The samples were visualized by CBBR-250 staining.
Northern blot analysis
Total RNA was isolated from B. subtilis using Modified Kirby Mix, as proposed previously [45]. Before RNA
isolation, B. subtilis cells were grown in 2× YT or SMM (Spizizen minimal medium).
Northern blot analysis was performed as described previously [46]. The probe was amplified by PCR and
labeled with [α-
32
P]-dATP at 37°C for 16 h. For the northern blot analysis, ∼25 mg of total RNA was loaded
and separated on a 1% agarose gel containing 0.22 M formaldehyde and transferred to a Hybond-N+ mem-
brane filter (GE Healthcare).
Western blot analysis
The crude B. subtilis cell extract (100 mg) was resolved by SDS–PAGE using a 12% gel. The resolved proteins
were transferred to a nitrocellulose membrane (GE Healthcare), and western blot analysis was performed as
proposed previously [47]. The resulting western blot signals were visualized using a colorimetric detection kit
(GE Healthcare), according to the manufacturer’s instructions.
Construction and analysis of a gusA fusion reporter strain
To determine the β-glucuronidase (β-Glu) activity of the P
yla
:gusA fusion reporter in B. subtilis PS832, the
pMLK83 vector was used as described previously [48]. The Pyla promoter region was amplified by PCR using
PylaF (50-ATACTGAAAGCTTTATATTG-30) and PylaR (50-ACGAACAAGGATCCTTTACT-30) primers. β-Glu
activity was measured as previously proposed [43]. β-Glu activity units were calculated using the formula
1000 × A
420
/reaction time (min) × OD
595
of culture.
The sample was obtained after the addition of various metal ions to the cells in the exponential growth
phase which were then incubated at 37°C for 30 min. The β-Glu activity of P
yla
:gusA was measured as follows.
After the addition of the metal ions, the cells were harvested every 15 min for northern blot analysis. Total
RNA was extracted from the harvested cells, and 25 mg of total RNA was used. C, before treatment; Zn,
100 mM ZnCl
2
; Mn, 100 mM MnCl
2
; Ca, 100 mM CaCl
2
; Mg, 100 mM MgCl
2
; Fe, 100 mM FeCl
2
.
High-resolution S1 mapping
For high-resolution S1 mapping of the clpP promoter, the corresponding probe was prepared by PCR using
clpP primers (forward: 50-TATTTCGAGAGGCCGTTTTTTAAA-30and reverse: 50-AACGTTGTCATCAATCG-
CAGATCC-30), which were cloned into the clpP:pGEM T-easy vector. The hybridization reaction contained
25 mg of RNA and the S1 nuclease mapping analysis was carried out as previously described [43]. The products
were analyzed on a sequencing gel with the sequencing ladder generated from the forward primer of the clpP
gene using the clpP:pGEM T-easy vector as a template.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2132
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
Determination of growth, survival, and sporulation rate
Cell growth was observed every hour, and viable B. subtilis mutant cells and their spores were counted every
2 h for 24 h in 2× YT medium to obtain the spore count. To show effect of the activities of YlaC and YlaD on
the growth of B. subtilis, each strain was cultured on LB agar and LN media [LB agar medium containing neo-
mycin (10 mg/ml) at 37°C for 24 h]. PS832, B. subtilis wild type; HB001, control; HBC01, +ylaC; HBC02,
ΔylaC; HBD01, +ylaD; HBD02, ΔylaD.
To determine the effect of H
2
O
2
on sporulation, the survival rate of the PS832 cells and mutants was
observed as follows. After treating the cells in exponential growth with 1 mM H
2
O
2
, the cells were withdrawn
every 30 min. The collected cells were spread on an LB agar medium containing neomycin (10 mg/ml) and
incubated at 37°C for 16–18 h. The survival rate was calculated as the ratio of the total number of cells before
and after H
2
O
2
treatment. The sporulation ratio (right) was calculated as the ratio of the number of spores to
the total number of surviving cells. The ratio of spores to total viable cells, indicating the sporulation rate, was
determined by the resistance of spores against heat treatment [49,50]. At least three independent repetitions
were performed for all experiments.
Far-ultraviolet circular dichroism spectrum
The far-UV CD (ultraviolet circular dichroism) spectrum was recorded on a Jasco J715 spectropolarimeter at
ambient temperature. The protein concentration was ∼6mM. The measured ellipticity, θ,wasconvertedtomolar
ellipticity, [θ], by the relation: [θ]=θ/10 lc
MRW
,wherelis the path length and c
MRW
is the protein molar concen-
tration per residue. Quantification of the secondary structure was performed using the self-consistent method
[51]. All the proteins prepared were dialyzed against TNG
10
buffer (10 mM Tris–HCl, 150 mM NaCl, and 10%
glycerol [ pH 8.0]). YlaD and all the mutants were treated with 1 mM DTT to obtain their reduced forms.
To analyze the conformational changes in YlaD in its oxidized state, YlaD was treated with 1 mM H
2
O
2
.
Each spectrum represents the average of 5–7 measurements for all proteins (6 mM). The far-UV CD spectra of
apo YlaD
N
were observed as follows. Apo YlaD
N
and reduced YlaD
N
were dialyzed in the buffer (10 mM Tris–
HCl, 150 mM NaCl, 10% glycerol, pH 8.0) containing 1 mM DTT. YlaD
N
Zn
and YlaD
N
Mn
were generated from
YlaD
N
Red
and 1 mM EDTA.
2D LC–MS/MS
Mass analysis and protein identification were carried out using a ProteomeX LTQ 2D LC–MS/MS spectrometer
(Sinco). Amino acid sequences were aligned using Clustal 2.1.
Statistical analysis
The results are presented as the mean ± standard deviation (SD). The statistical significance of the differences
was evaluated by Student’st-test in Microsoft Office Excel 2015. For all comparisons, differences with P< 0.05
(**), P< 0.01(**), and P< 0.001(***) were considered statistically significant.
Results
Ylac transcriptional activity is sensed by YlaD redox state
We found that YlaC serves as an ECF σfactor, which might allow B. subtilis to respond to oxidative stress, and
that YlaC-deficient cells show defects in sporulation-specificphenotype,suchasylad-overexpressing mutants
[41,42]. Despite the predicted role of YlaC in sensing oxidative stress in our previous mutant data [42], it
contains no cysteine resides and therefore does not sense oxidants in the same way as σ
R
from S.coelicolor [27].
Our hypothesis prompted us to study the process of transcriptional gene expression in the stress response of
B. subtilis to changes in Mn availability by YlaC and YlaD, owing to the fact that many genes induced by Mn
2+
are regarded as being activated by σ
B
, which is controlled by Mn
2+
-dependent enzymes. Also, the induction of
Hsp33 [52], RsrA [27], and protein kinase C [53], with a cysteine-containing Zn center as the redox switch, high-
lighted the possible alterations of the oxidation-induced Zn-binding ability of these regulators.
To address this, we preferentially hypothesized that this role should be performed by YlaD, a membrane-
bound, putative anti-σfactor for YlaC consisting of 97 amino acids, including five cysteines, with a calculated
molecular mass of 11.26 kDa (http://web.expasy.org/compute_pi/). Because the predicted amino acid sequence
of YlaDs contained a conserved HX
3
CXXC motif which can act as a stress-sensing domain, YlaD was expected
to sense oxidative stress, similar to OrfH in Mycobacterium tuberculosis, OrfE in Myxococcus xanthus, and
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2133
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
RsrA in S. coelicolor (Supplementary Figure S1). The B. subtilis yla operon contains four ORFs (i.e. ylaA,ylaB,
ylaC, and ylaD) and the Pyla promoter has been predicted to be autoregulated by YlaC [54](Figure 1A).
Whereas the cellular roles of membrane-anchored YlaA and YlaB are unknown, YlaB has been reported to be a
regulatory protein for YlaD [41].
Next, to prove our prediction that YlaD is an anti-σfactor for YlaC, we determined the transcription
patterns from the P
yla
promoter driven by the B. subtilis YlaC-containing RNAP and observed whether this
promoter in the yla operon might be autoregulated by YlaC (Figure 1B). YlaC recognized the P
yla
promoter,
and a gradual increase in the concentration of the reduced form of YlaD proportionally inhibited YlaC-driven
transcription. The oxidized YlaD did not show transcriptional activity by itself. To show the exact mechanism
by which redox changes in YlaD regulate YlaC, we used a gel-mobility shift assay using YlaD with modification
of AMS and in vitro transcription from the P
yla
promoter (Figure 1C), which indicated that YlaD senses redox
changes by binding to YlaC and regulates YlaC transcriptional activity. Interestingly, our experimental results
were different from previous reports that no interaction occurs between YlaC and YlaD, but showing direct
interactions of YlaB with YlaD through a yeast two-hybrid assay [41,55]. These phenomena support direct
interaction between YlaB and YlaD, because ylaB overexpression in addition to ylaC exhibits a mitigated
attenuation of the expression of whole ylaABCD operon by YlaD and thereby YlaB might interact with YlaD.
Additionally, the yeast two-hybrid system has steadily developed from a means to observe interactions between
a few proteins to a tool that can demonstrate protein–protein interactions. We carefully thought that yeast
system might not be always suitable to analyze plasmid DNA directly. When considering data that ylaD disrup-
tion causes significant enhanced transcription of the whole ylaABCD operon and YlaC-mediated transcriptional
activity of ylaABCD operon is significantly attenuated in ylaD-overexpressing mutant [41]; thus, we also postu-
lated that YlaC interacts with YlaD solely under the reduced state in our experimental conditions (Figure 1D).
This result inferred that YlaC is inactivated by interaction with the reduced form of YlaD, and in this way,
YlaC transcriptional activity can be regulated by redox changes in YlaD.
Transcriptional activity of YlaC increases by oxidative stresses
To confirm our data (Figure 1) using a different experimental approach, we observed the increase in the β-Glu
activity of the Pyla:gusA reporter gene by treatment with oxidants (Supplementary Figure S2A). Northern ana-
lysis also showed that ylaC gene transcription is highly induced after treatment under the same experimental
conditions (Supplementary Figure S2B), and that YlaC protein is correspondingly induced after H
2
O
2
treatment
(Supplementary Figure S2C). Here, we interested in how ylaC gene transcripts increased relative to other ylaA
transcripts given that they are all encoded within the same operon, a result that was counter-intuitive. A
possible explanation is that the yla operon might have a second promoter in addition to P
yla
. In this regard, a
previous report demonstrated that the ylaABCD operon contains two promoters, the Pyla promoter and an
Spx-dependent σ
A
consensus promoter. The ylaC gene is expressed using the Spx-dependent σ
A
consensus
promoter when diamide was added to exponentially growing cells; however, the YlaC protein was not detected
[41]. For this reason, to test whether YlaC might recognize the Spx-dependent σ
A
promoter (Supplementary
Figure S2D), in addition to Pyla,anin vitro transcription assay was performed using the Spx-dependent σ
A
consensus promoter as the template. Transcripts obtained by in vitro transcription only appeared from the Pyla
promoter (Supplementary Figure S2E); moreover, an S1 nuclease protection analysis also demonstrated that
YlaC recognized the Pyla promoter (Supplementary Figure S2F). However, transcripts from the Spx-dependent
σ
A
consensus promoter were not detected. Hence, YlaC solely recognizes P
yla
within the yla operon and this
promoter is autoregulated by YlaC, indicating that functional YlaC is solely derived from the Pyla promoter.
Based on our data, it can be clearly hypothesized that oxidative stresses activate the transcriptional activity of
YlaC and that oxidized YlaD does not interact with YlaC. To prove this assumption under various oxidative
stress conditions, the ylaC gene transcript was monitored by the measurements of in vivo β-Glu activity derived
from Pyla:gusA, and YlaC protein expression was observed by western blot analysis. The β-Glu activity of the
Pyla:gusA promoter was increased by treating with 0.1 mM H
2
O
2
and 1 M NaCl, while it was not significantly
induced by heat treatment at 50°C and 5% ethanol (Supplementary Figure S3A). These data were also cross-
confirmed by western blot analysis using an anti-YlaC antibody (Supplementary Figure S3B). Hence, YlaC has
the ability to respond to external stress, particularly oxidative stress, and this is regulated in vivo by the redox
state of YlaD.
To determine the effect of gene expression changes in ylaC and ylaD mutants under various oxidative stress
conditions (i.e. H
2
O
2
, diamide, paraquat, and menadione), the wild-type PS832 and mutants, including HB001,
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2134
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
Figure 1. Autoregulation of YlaC and redox sensing by YlaD.
(A)Configuration of the yla operon of B. subtilis and the Pyla promoter autoregulated by YlaC are shown. The coding regions of
the ylaA,ylaB,ylaC, and ylaD genes are indicated by thick arrows, and the promoter region (Pyla), which is autoregulated by
YlaC, is indicated by a thin arrow. (B)In vitro transcription assay of the P
yla
promoter with RNAP in complex with YlaC and
(C) Mobility shift of YlaD (2 mg) after the modification with AMS. All experiments here were performed as described in detail in
the Experimental section. Each sample was resolved by electrophoresis on a 16% non-reducing Tricine polyacrylamide gel.
(Top) C, purified YlaD (control) + 15 mM AMS; Oxi, oxidized form; Red, reduced form. (Down) C, the reduced form of YlaD by
treatment with 1 mM DTT as described. (D) Formation of the YlaC–YlaD complex under reducing conditions using native PAGE.
YlaC and YlaD were incubated with TNDG
20
buffer at 37°C for 30 min.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2135
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
HBC01, HBC02, HBD01, and HBD02, were used to assess their resistance against toxic oxidants
(Supplementary Figure S4). When observing the sensitivity to external oxidants in exponentially growing cells,
the disk diffusion analysis showed that the HBC01 (+ylaC) was shown to be significantly less sensitive than
PS832 by the treatment of H
2
O
2
and diamide. Meanwhile, HBC02 (ΔylaC) was more sensitive than PS832 and
other mutants under the same experimental conditions. Under the specific oxidative stress condition such as the
superoxide generator paraquat treatment, in the case of B. subtilis σ
M
, this ECF σfactor is known to be induced,
but not by H
2
O
2
, even though it is very difficult to know whether these responsive pathways are mainly due to
from superoxide or subsequently produced H
2
O
2
[56]. However, as another convincing evidence of the fact that
the ylaC gene as well as YlaC were highly expressed by the treatment with H
2
O
2
based on our finding and previ-
ous work, our data implied that YlaC responds to oxidative stress by peroxides, rather than by superoxides.
Ylad is one of the seven anti-σfactors in the ZAS family found in B. subtilis
Because metal ions (i.e. Zn, iron, and Mn) are commonly found in complexes with sulfur in redox-sensing
systems, the effect of oxidative stresses on the metal ions co-ordinated in YlaD was investigated (Table 2). The
Zn ion can interact with reduced SH moiety in free cysteines and senses the redox state. With respect to the
catalytic, structural, and regulatory functions of many proteins, Zn also plays an important role in maintaining
biological processes, such as the redox switch [5,57]. This is because, although YlaD has been classified as a
member of the ZAS family [27,33], it is elusive whether YlaD binds to Zn or not. To prove this assumption, we
determined the metal ion content of a truncated YlaD variant lacking the C-terminal transmembrane domain
(Table 1; designated as YlaD
N
); this variant was overproduced and successfully purified from E.coli in a previ-
ous study [42]. To calculate the metal ion content of YlaD
N
under different redox conditions, YlaD
N
was
treated with 10 mM H
2
O
2
and diamide, respectively, and the metal ion content was measured using ICP-AES.
The molarity of the subunit was calculated based on a molecular mass of 8.5 kDa. Interestingly, the ICP-AES
data demonstrated that YlaD
N
contains one iron and one Zn ion per subunit of YlaD
N
, and that iron and Zn
ions are not detected by treatment with H
2
O
2
or diamide (Table 2). Based on this, we newly confirmed that
YlaD is a ZAS-family member, one of the seven anti-σfactors found in B.subtilis. ICP-AES data also indicated
that one Zn ion can bind per reduced YlaD protein in vivo and the reduced form of YlaD also bound one iron
ion. Despite the ICP-AES data, we failed to identify a conserved iron-binding site in YlaD. These data
(Figure 1, Supplementary Figures S2 and S3, and Table 2) strongly indicated that redox changes in YlaD might
be a critical determinant for the transcriptional activity of YlaC.
The YlaC–YlaD interaction is decreased by the increase in Mn concentrations
and critically affects redox state and binding affinity of YlaD to YlaC
Inspired by the ICP-AES data (Table 2) and studies explaining that many metallo-regulators regulate their
targets as metal-dependent sensors [58], we again hypothesized that the negative effect of YlaD on YlaC activity
might be shown through metal ion control in vivo, concomitant with the redox changes in YlaD, as previously
demonstrated in S.coelicolor RsrA [27,28,59]. To show the effect of metal ion substitution on the interaction of
YlaC with YlaD, which reflects growth and/or sporulation in B. subtilis, we again observed the β-Glu activity of
Table 2 Metal ion contents in YlaD.
To calculate the metal ion contents of YlaD under redox conditions,
YlaD was treated with 10 mM H
2
O
2
and diamide, respectively. The
metal ion contents were determined by ICP-AES. The molarity of the
subunit was calculated based on its molecular mass of 8973.15 Da.
Sample
Metal contents (mol of metal ion/mol of
YlaD
N
)
Zn Ni Fe Mn Cu
YlaD
N
1.2 n.d. 0.8 n.d. n.d.
10 mM H
2
O
2
n.d. n.d. n.d. n.d. n.d.
10 mM diamide n.d. n.d. n.d. n.d. n.d.
Abbreviations: nd, not detected.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2136
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
the Pyla:gusA gene in vivo. The addition of metals to the culture media for metal substitution, particularly Mn,
increased the β-Glu activity of the Pyla:gusA gene (Figure 2A, top panel). The increase in ylaC gene transcrip-
tion was further re-examined and confirmed using a northern blot analysis (Figure 2A, bottom panel). These
observations explained that the transcriptional activity of YlaC is metal-dependent, indicating that the anti-σ
activity of YlaD might be altered by the intracellular Mn concentration. Hence, we examined whether Mn ions
alter the YlaC–YlaD interaction in detail. Using native PAGE, a protein–protein complex was formed between
YlaC and YlaD under reducing conditions in the presence of Zn, but not after adding Mn to the complex
mixture (Figure 2B), indicating that YlaD controls the transcriptional activity of YlaC in response to the Mn
concentrations, and that the reduced YlaD
N
containing Zn ions can only interact with YlaC. Additionally, in
the absence of Mn supplementation, we previously showed that the formation of the YlaC–YlaD
N
complex is
affected by redox changes caused by the presence of external oxidants in vitro [42], but it is not known how
this affects activity in vivo. Under these conditions, the YlaC–YlaD
N
complex is detected as a new protein band
by native PAGE solely under reduced conditions; however, when the purified YlaC and YlaD
N
were incubated
in the presence of 2.5 mM H
2
O
2
, no complex was detected [42]. These in vitro data significantly coincided
with the data showing the ability of the reduced form of YlaD to bind to YlaC through both in vitro transcrip-
tion assays using the P
yla
promoter engaged in RNAP in complex with YlaC (Figure 1B) and using the mobility
shift of YlaD after modification using AMS (Figure 1C).
Both Mn concentration and redox changes in YlaD affect in vivo YlaC–YlaD
interaction
To understand our data regarding the YlaC–YlaD interaction more clearly, we showed the effect of Mn ions in
vivo on the binding activity of YlaD to YlaC and the corresponding transcriptional activity of YlaC affecting
growth and/or sporulation. Here, we expected that the increase in YlaC expression reflects the increase in the
extent of oxidized YlaD, as elucidated by our previous results (Figures 1 and 2, Supplementary Figures S2 and
S3, and Table 2). To clarify this further, the effect of ylaC gene expression during sporulation in the presence
or absence of Mn ions was monitored in ylaC mutants grown in SMM. In the presence of Mn ions, the sporu-
lation of ylaC-overexpressing HBC01 was significantly more rapid and effective than in wild-type PS832
(Supplementary Figure S5). Importantly, the nonfunctional ylaC gene mutant HBC02, which does not interact
with ylaD, showed defects in spore formation (Supplementary Figure S5A). To support these data, the wild-
type PS832 and mutants were then cultured in SMM supplemented with 100 mM MnCl
2
. Even though all the
PS832 and mutants increased in cell density, cell growth and spore formation increased significantly solely in
HBC01 in the presence of MnCl
2
(Supplementary Figure S5B,C). This activation of sporulation in HBC01 cells
showed a highly similar pattern to that seen in the absence of MnCl
2
. However, we failed to test the behavior
of different ylaD mutants in the presence of Mn ions, because the different ylaD mutants did not display a
Figure 2. Effect of metal ions on the expression of ylaC and the metal ion-dependent formation of the YlaC–YlaD complex.
(A) Measurement of β-Glu activity using the P
yla
:gusA fusion reporter and (B) metal ion-dependent formation of the YlaC–YlaD
complex. The arrows indicate the positions of YlaC, YlaD (YlaD
Zn
,YlaD
Mn
), and the YlaC–YlaD complex.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2137
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
reproducible profile in response to the addition of exogenous Mn ions, despite numerous attempts. Regardless
of this, our data indicated that the respective decrease and increase in the sporulation rate of ylaC-deficient
HBC02 and ylaC-overexpressing HBC01 mutants are partially thought to lack the ylaC gene and to lose the
binding activity of YlaD, respectively, under the presence or absence of Mn ion, respectively. These results were
very similar to the YlaD inactivation seen under oxidized states in vitro (Figures 1 and 2).
Considering these data, we tested the sporulation rate in both ylaC and ylaD mutants to verify whether ylaD
gene expression affects growth and development under the oxidative stress conditions associated with ylaD
expression and its binding activity to YlaC (Supplementary Figure S6). In the absence of oxidants, the
ylaC-overexpressing mutant HBC01 exhibited growth defects compared with other strains. However, this
mutant and ylaD-deficient HBD02 showed remarkably higher sporulation efficiencies compared with HBC02
and HBD01. Simultaneously, ylaC-deficient HBC02 displayed a lower sporulation rate than that of wild-type
PS832, although spore formation of this strain was normal (Supplementary Figure S6A). These phenomena
support our previous findings that ylaC gene expression triggers sporulation [42], and that the transcriptional
activity of YlaC is activated under oxidative stress (Supplementary Figures S2 and S3). This indicates that tran-
scriptional activity by YlaC is more intense when the ylaD gene is disrupted. Additionally, our mutant pheno-
types were also observed on agar plates, where HBC02 and HBD01 mutants showed clear growth defects and
formed small colonies, in contrast with the HBC01 and HBD02 mutants (Supplementary Figure S6B). These
growth defects, observed in both liquid culture and agar plates for HBC02 and HBD01, were thought to be
closely associated with the enhanced sporulation rates due to the increased transcriptional activity of YlaC,
inversely proportional to YlaD activity. Similar results in ylaC-overexpressing HBC01 and ylaD-overexpressing
HBD01 mutants, which reflect the increase in growth and sporulation, were also shown in cells grown in the
presence of exogenous H
2
O
2
(Supplementary Figure S6C). These data also demonstrated that ylaC expression
accelerates B. subtilis sporulation and the reduced form of YlaD acts as a negative regulator of this mechanism
by interacting with YlaC. We also observed that for ylaC-overexpressing HBC01 and ylaD-deficient mutants,
the entry into sporulation was similar in cells treated with H
2
O
2
.
The structural stability of YlaD requires Cys3
Next, site-directed mutagenesis was performed to determine their importance in the HX
3
CXXC motif-
containing YlaD structure and function in vitro, as YlaD contains five cysteines, including four N-terminal
cysteines and one C-terminal cysteine that resides in the transmembrane domain. Specifically, each of the four
N-terminal cysteines in YlaD
N
was individually mutated to a serine residue (DC3S, DC18S, DC33S, and
DC36S; Figure 3), and, after confirmation by sequencing, each mutant protein was expressed in E.coli using
pGEX-4T-1 and purified. Structural changes in these mutants were shown using far-UV CD and whether
mutations influenced the YlaC–YlaD interaction (Figure 3). The far-UV CD analysis using reduced YlaD
N
revealed a typical α-helical conformation with two negative bands at 208 and 222 nm, and a positive band at
190 nm. However, the α-helical and β-sheet conformations of oxidized YlaD
N
, supplemented with 1 mM
H
2
O
2
, had a small, but discernable, shift in helical conformation (Figure 3A,B). This result inferred that
reduced YlaD
N
had a tighter α-helical structure, due to the bound metal ions, and that oxidation, which causes
loss of the metal ions, weakens the α-helical structure; this is also supported by the ICP-AES data (Table 2). In
the DC3S mutant, the α-helical and β-sheet structures disappeared, and a random coil conformation was
adopted regardless of the oxidation/reduction state (Figure 3C). In the DC18S mutant, there was a slight shift
towards an increased α-helical content under oxidized conditions (Figure 3D). The DC33S mutant did not
have an α-helical structure (Figure 3E), while the behavior of the DC36S mutant was similar to that of the
DC18S mutant (Figure 3F).
While DC36S showed a similar pattern of negative and positive bands, the DC18S and DC33S mutants lost
their signature negative bands characteristic of α-helical conformations, and DC3S showed a random coil struc-
ture under reduced conditions (Figure 3). To further understand these structural changes, the reduced and oxi-
dized forms of each YlaD mutant were compared. In the case of wild-type Zn–YlaD, we clearly observed a
small, but discernable, shift in the band toward 200 nm in the helical conformation in the oxidized state, com-
pared with that in the reduced state. Unlike wild-type Zn–YlaD, the structure of three of the cysteine-
substitution mutants (i.e. DC3S, DC18S, and DC36S) was not changed by oxidation state, particularly for
DC3S (Figure 3C), which was a random coil (Figure 5A). In contrast, DC33S demonstrated changes in its
random coil structure under oxidized conditions (Figure 3E). These results confirmed that Cys3 and Cys33
might play important roles in the structural stability of YlaD.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2138
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
Next, we studied the YlaC- and Zn-binding properties of the cysteine-substituted YlaD mutants. All cysteine-
substituted mutants displayed a lack of YlaC-binding activity measured under native PAGE conditions, which
are conducive to YlaC–YlaD binding, and the Zn content appeared reduced in each case, especially for DC3S
(data not shown). We conclude that Cys3 is the most essential cysteine for the structural stability of YlaC and
for the YlaC-binding activity of YlaD, as three of these are needed. To map the disulfide bond-forming cysteine
residues that we performed MALDI-TOF mass spectrometric analysis of tryptic peptides generated from the
reduced and oxidized forms of YlaD protein (Figure 4). For this experiment, free thiol groups in YlaD were
modified by IAA, which is a widely used alkylating agent that reacts with reduced thiols. YlaD was oxidized by
H
2
O
2
treatment and reduced by DTT treatment, determined by examining the mobility shift following treat-
ment with IAA (Figure 1C). MALDI-TOF analysis showed that YlaD
N
was modified by IAA. The average
non-IAA-modified mass was 8973.15 Da (http://expasy.org/cgi-bin/peptide-mass.pl) and the mass modified by
Figure 3. Far-UV CD spectra of the E. coli-derived YlaD protein and cysteine-substitution mutants, following redox changes.
All experimental procedures required for far-UV CD spectra were described in the Experimental section. The indicated lines of
far-UV CD spectra are as follows. Red, the reduced form of YlaD; Oxi, the oxidized form of YlaD. (A) The reduced form of YlaD and
its related cysteine-substitution mutants, (B)YlaD,(C) DC3S (Cys3 to Ser), (D) DC18S (Cys18 to Ser), (E) DC33S (Cys33 to Ser), and
(F) DC36S (Cys36 to Ser).
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2139
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
four IAAs was 8230.7565 Da, with modifications occurring at Cys
3
(T1 site), Cys
18
(T2 site), and Cys
33
and
Cys
36
(T4 site). In the presence of 1 mM H
2
O
2
, T1 and T2 were observed as major peaks, and T2 and T4 were
observed as minor peaks (Figure 4). These data inferred that the Cys
3
–Cys
18
and Cys
33
–Cys
36
pairs formed
disulfide bonds under oxidizing conditions.
Biophysical analysis reveals YlaC binding to YlaD to be inhibited by Mn
Based on earlier experimental data showing the β-Glu activity of P
yla
:gusA and northern blot analysis, transcrip-
tion of the ylaC gene from the Pyla promoter was increased by MnCl
2
treatment in cell culture. These data
indicated the loss of YlaD binding to YlaC in the presence of Mn, regardless of the presence of Zn, which acts
to promote YlaC–YlaD (Figure 2 and Supplementary Figure S2). Based on the data, YlaD appears to regulate
the interaction with YlaC, depending on the Mn ion concentration in the cell.
To understand the ion metal-sensing mechanism more clearly, metals were removed from YlaD
N
by treatment
with 1 mM EDTA before far-UV CD spectrum analysis (Figure 5). Additionally, the apo YlaD was saturated with
Zn to prepare a functional Zn–YlaD protein together with the Mn–YlaD by pre-saturating with Zn, removing the
unbound Zn, and then treating with Mn. Interestingly, YlaD
N
treated with 1 mM EDTA had changes in its
α-helix structure at 222 and 190 nm and its structural confirmation also shifted to a β-sheet (Figure 5A). These
data indicate that YlaD needs to contain metal ions, which play an important role in metal ion sensing as well as
being required for structural stability. To discover which metal ion was sensed by YlaD, a far-UV CD spectrum
analysis was carried out with apo YlaD
N
in the presence of Zn or Mn ions. YlaD
N
containing the Zn ion com-
pletely reformed the α-helical structure, whereas YlaD
N
containing the Mn ion did not reform a functional YlaD
N
(Figure 5B). These data explain that Mn ions can be inserted into the metal-binding site in YlaD
N
, and the effect
of this ion on structural stability is important. However, the protein retains Zn. This would be most simply
explained by the presence of a separate, though currently poorly defined, metal site that can bind Fe or Mn.
We therefore asked whether the structural change in YlaD is reversible after a redox switch. To assess how
YlaD can recover its original structure, Mn -saturated YlaD was prepared as described above for Zn–YlaD. In
Figure 4. Analysis of disulfide bond formation in YlaD.
Five micrograms of YlaD was incubated with 1 mM DTT and 1 mM H
2
O
2
for 30 min as indicated. After electrophoresis, the
proteins were treated with trypsin and analyzed by MALDI-TOF mass spectrometry. The site cleaved by trypsin is indicated by
an arrow as well as the table provided above.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2140
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
contrast with our expectation that the structural change was potentially reversible after the metal switch, Mn
-saturated YlaD did not recover, unlike Zn–YlaD, after the redox switch (Figure 6A). As previously concluded
(Figure 1 and Table 2), YlaD loses both its Zn and iron ions in the oxidized state, and the resulting structural
change causes the release of YlaC from the YlaC/YlaD complex, thereby activating its transcriptional activity.
To investigate how structural changes in YlaD, following its metal ion switch, control the transcriptional activ-
ity of YlaC, we analyzed the Mn content in Mn-saturated YlaD obtained from Zn–YlaD. Surprisingly,
Mn-saturated YlaD contained both Zn and Mn (Figure 6B), implying that either Zn/Fe–or Zn/Mn–YlaD had
a different YlaC-binding affinity, as has been previously seen with PerR and its target DNA.
Ylad senses sporulation–initiation signaling and regulates the transcriptional
activity of YlaC, which stimulates spore formation by promoting clpP gene
expression
Based on previous works and our previous data of the YlaC induction by oxidants such as diamide and H
2
O
2
,
contributing to H
2
O
2
resistance [42], we compared the target protein expression levels following either ylaC
Figure 5. Far-UV CD spectra of apo YlaD
N
.
(A) Far-UV CD spectrum of reduced YlaD
N
and apo YlaD
N
.(B) Far-UV CD spectrum of YlaD
N
on the effect of metal ions.
Far-UV CD spectrum of the metal-dependent YlaD
N
protein. Each spectrum (Aand B) shown is the average of five
measurements of YlaD
N
(6 mM). YlaD
N
Zn
, YlaD
N
containing Zn ion; YlaD
N
Mn
, YlaD
N
containing Mn ion. All experimental details
were described in the Experimental section.
Figure 6. Far-UV CD spectra of YlaD and its metal ion components, determined after the addition of Mn to Zn–YlaD in
solution.
(A) Far-UV CD spectrum of the purified YlaD protein of E. coli BL21. Each spectrum shown represents the average of five
measurements of each sample (6 mM). YlaD
Zn
and YlaD
Mn
were generated by treating YlaD
Red
with 1 mM EDTA. YlaD
Zn
→Mn was
generated by adding MnCl
2
to YlaD
Zn
,andYlaD
Mn
→Zn was generated by adding ZnCl
2
to YlaD
Zn
.(B) The concentrations of the
metal ion components were determined by ICP-Atomic Emission Spectrometer after the treatment with 100 mMMnCl
2
. Experiments
were performed independently at least three times. All experimental procedures were described in the Experimental section.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2141
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
disruption or overexpression using 2D-PAGE (two-dimensional polyacrylamide gel electrophoresis;
Supplementary Figure S7). The ylaC-deficient mutant had a diminution in the levels of thioredoxin, peroxire-
doxin (ytgI), GroES, pyruvate decarboxylase E-1αsubunit, and ClpP expression. Most of these proteins are
related to redox-dependent or developmentally regulated processes (Supplementary Table S1). Based on these
findings, we re-examined B. subtilis development because earlier experiments showed deficient spore formation
in ylaC-deficient HBC02 and ylaD-overexpressing HBD01 cells (Supplementary Figures S5 and S6). The data
showed that the YlaC activity affects spore formation through its redox- or metal-dependent transcriptional activ-
ity since it is regulated by the YlaD functions, which are capable of sensing both redox changes and metals by
changing its structural conformations. When observing the initiation of sporulation during growth, the β-Glu
activity of the P
yla
:gusA fusion reporter increased markedly during the early-stationary phase and then subse-
quently decreased, as did the activity of many ECF σfactors (Figure 7A,B). Thus, we hypothesized that YlaC
might function as an ECF σfactor capable of causing the expression of sporulation-related genes when YlaD
senses sporulation-inducing signals, such as H
2
O
2
or Mn. To determine whether YlaC facilitates sporulation, we
monitored the expression of genes related to the initiation of sporulation by northern blot analysis. Use of the
wild-type strain, a YlaC-overproduction mutant, and a YlaC-deletion mutant showed that both the kinA and
spo0A genes were induced by the intracellular GTP concentration (caused by nutritional limitation) and were
similarly expressed in both strains. However, expression of the sigH and clpP genes was modulated by various
stress factors and displayed distinct differences. These two genes were constitutively overexpressed in the
YlaC-overproducing mutants, whereas their expression levels were low in the ylaC-deficient HBC02 mutant
(Figure 7C).
These data demonstrate that YlaC induces the expression of the clpP gene, whose mutation seriously impairs
growth and development [60], and that ClpP and ClpX may regulate σ
H
activity. To test this possibility, we
monitored clpP gene expression by northern blot analysis under YlaC-inducing conditions, such as exposure to
H
2
O
2
and Mn ions. Transcripts of the clpP gene increased in the wild-type strain, as well as the
YlaC-overproduction mutant, under both conditions, whereas the YlaC-deletion mutant maintained a steady
level of clpP expression (Figure 8A) because it is affected by many signals including heat and puromycin [60].
Interestingly, ylaD-deficient HBD02 and ylaD-overexpressing mutants increased and decreased clpP gene
expression by increasing concentrations of Mn and H
2
O
2
, respectively (Figure 8B), indicating that YlaC over-
production increases sigH and clpP levels during growth and sporulation and that YlaD acts as a critical sensor
of redox changes and Mn
2+
levels by the transcriptional activity of YlaC. These data were supported by
2D-PAGE analysis (Figure 8C). We propose that YlaC stimulates sporulation through ClpP induction, while
YlaD senses cellular signals, such as H
2
O
2
, produced from cell metabolism.
The clpP gene, which encodes the proteolytic component of Clp, is known to be induced after heat shock,
salt, and ethanol stress, as well as treatment with puromycin [60,61]. Two transcriptional start sites upstream of
the clpP gene are known to contain the consensus sequences of promoters recognized by σ
A
and σ
B
. YlaC
recognizes an unidentified promoter in the clpP gene and induces ClpP after treatment with H
2
O
2
.
High-resolution S1 nuclease mapping analysis was carried out to determine the transcription start point of the
YlaC-dependent promoter in the clpP gene. The clpP gene was induced after treatment with H
2
O
2
, and at least
three promoters were found upstream of the clpP gene (Supplementary Figure S8A). A + 1 site for
YlaC-dependent transcription was located 5–13 bp upstream of the +1 site of σ
A
. The YlaC-dependent pro-
moter of the clpP gene was found to overlap with the σ
A
-dependent promoter (Supplementary Figure S8B).
This result corresponds with that obtained from the in vitro transcription assay. Based on these data, we antici-
pate that YlaC can compete with σ
A
for the expression of the clpP gene.
Discussion
Building on our prior study, the present data confirm that the biophysically demonstrated mechanism is a
redox-sensitive ECF, σfactor-mediated, transcriptional redox regulation via the HX3CXXC motif featuring a
cysteine-containing Zn center as the redox switch. The mechanism was determined under oxidative stress and
Mn-substituted, Zn-limited conditions, with priority given to the control of the oxidation–reduction state of
anti-σfactor YlaD in B. subtilis. This is the first report of the involvement of oxidative stress-responsive B. sub-
tilis ECF σfactors in the mechanism of sporulation initiation.
Notably, the data reveal the actions responsible for redox-sensing of anti-σfactor YlaD. Its redox change
modulates the transcriptional activity of its σfactor, YlaC, by a Mn-dependent redox-sensing molecular switch.
As mentioned in the Introduction, metallo-regulators respond to their interacting partners, accompanied by the
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2142
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
activity of growth phase-related σfactors, in specific metal-dependent regulation during sporulation initiation.
This is why the potential roles of the transcriptional activity of metal-co-ordinated global regulators and σ
factors in response to environmental stresses, which are strongly associated with spore formation as an oxida-
tive stress response, have been elucidated in each developmental stage, particularly of the sporulation process,
including the vegetative and stationary phases, sporulation, spore maintenance, germination, and outgrowth
[62]. The inference is that the developmental phase of B. subtilis has very different morphological traits that
undergo separate pathways of differentiation [63]. Thus, oxidative stress-response mechanisms of the preven-
tion and/or repair of macromolecular damage are distinguished from different cell type-dependent σfactors
[16,64]. For instance, the σ
B
form of RNA polymerase is associated with the Zn-centered, PerR-mediated tran-
scription of katE,dps,ohrA, and ohrB in the stationary phase. Thus, the activity of stress-response global
Figure 7. β-Glu activity of the P
yla
:gusA fusion reporter and the expression of genes related to the initiation of sporulation during development.
(A)β-Glu activity of the P
yla
:gusA fusion reporter. Cells were harvested during the early-exponential phase (EP), mid-exponential phase (MP),
late-exponential phase (LE), early-stationary phase (ESP), mid-stationary phase (MSP), and late-stationary phase (LSP). Each growth phase is indicated
by an arrow. The curve and the bar represent the growth curve and β-Glu activity, respectively. Units of β-Glu activity were calculated as follows:
1000 × A
420
/reaction time (min) × OD
595
of culture. (B,C) Northern blot analysis. The mRNAs of kinA,spo0A,abrB,sigH,clpX,clpP,andylaC (control)
were probed in this experiment. All experiments were repeated at least three times independently.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2143
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
regulators and σfactors necessarily accompanies metal ion regulation and sporulation as the primary circuit
prior to maintenance or replacement of metal ions.
Knowledge gained from advanced genome sequencing projects and the subsequent establishment of a compre-
hensive mutant database has revealed that diverse bacterial ECF σfactors comprise the ECF subfamily of σ
factors. Although the general concept of functional roles of ECF σfactors in B. subtilis is not new, most studies
have concentrated on the regulatory aspects of anti-σfactor activity acting on the cell surface events in the regula-
tion of cell envelope stress responses using various ECF σfactor mutants or sequence- and/or structure-based pre-
dictions of protein–protein interactions on a genome-wide scale. The internal N-terminal region of the
membrane-embedded anti-σfactors interacts with their partners and cognate ECF σfactors to sequester it from
RNA polymerase. However, how the external domain senses the nature of the extracellular stress and transmits
the appropriate signal to the internal domain, followed by the release of the RNAP-binding σfactor, remains
unclear. Even though many ECF σfactor–anti-σfactor pairs have been identified through the genome database
and laboratory analyses [33], compared with other well-defined σfactors, little is known about the molecular
mechanism of regulation of ECF σfactors in cooperation with their anti-σfactors. One concern is about how
anti-σfactors receive and sense environmental signals. Another is the dissociation of the ECF σfactors. Genome
analysis data support the hypothesis of functional traits of the seven types of ECF RNAP σsubunits that are
known for B. subtilis. Similarly, the ECF σsubfamily of regulators tends to respond to cell envelope stresses.
Hence, for at least a decade preceding the present study, B. subtilis ECF σfactor YlaC and its interacting partner,
YlaD, in the ylaABCD operon have been tentatively regarded as having a unique redox regulatory role. Supporting
evidence has come from studies that deciphered the transcriptional characteristics of the ylaABCD operon through
experiments using ylaC- and/or ylaD-disrupted and/or overexpressing mutants [33,41,42].
In our case, the main possible underlying reason for this knowledge gap is poor experimental reproducibility
resulting from the relatively low yield of recombinant YlaD, its unstable protein structure, and instability of
B. subtilis RNA especially in ylaC and ylaD disruptants. Confronted by erratic reproducibility of experimental
results and difficulties in interpreting the methods used, attention has focused instead on novel supportive
Figure 8. The expression of clpP was regulated by YlaC under oxidized conditions and through the addition of Mn to the cell culture.
(A,B) Northern blot analysis. (A, top) and (B, top) MnCl
2
(0.1 mM) was added to cells in the exponential growth phase in SMM. Following the
addition of MnCl
2
,clpP mRNA was probed. (A, bottom) and (B, bottom) After the treatment of the cells with 0.1 mM H
2
O
2
,clpP mRNA was probed.
(C) Two-dimensional PAGE. Cells in the exponential growth phase were treated with 0.1 mM H
2
O
2
. The ClpP was identified by MALDI-TOF-based
peptide mass fingerprinting as described in the Experimental section. For each process condition, all experimental trials were repeated at least three
times independently.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2144
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
measures. Presently, gene transcripts of interest were measured using real-time RT-PCR. The results were not
always identical with results of northern analysis. The latter analysis showed similar and reproducible transcrip-
tion patterns, for reasons that remain unclear. A non-radioactive northern method employing digoxigenin
(DIG)-labeled DNA probes and chemiluminescence also repeatedly failed. Therefore, data comparisons are hin-
dered by the difficulty in matching data sets between biochemical analyses (i.e. results from RT-PCR,
DIG-labeled northern blot analysis, and northern analysis using α-
32
P labeled probes) and the problem of
measuring gene expressions in ylaC and ylaD mutants. As another reason for the dichotomy between the
anticipated and actual data, we previously illustrated that in the absence of Mn ions (Mn
2+
)in vitro, the formation
of the YlaC–YlaD
N
complex might be affected by redox changes caused by external oxidants, but not how they
affect this activity in vivo. We previously reported these redox changes in vitro in the presence and absence of oxi-
dants, but many experiments were again required because of poor data quality related to the short lifetime and
poor stability of the protein samples, and protein degradation and precipitation that occurred. Importantly, the
main reason has been the ongoing instability of the YlaD protein during experiments we have conducted over
many years. YlaD and the cysteine-substituted YlaD mutants that are required for the experiments were also quite
unstable under all the experimental and storage conditions. The optimal operating pH for recombinant YlaD
protein production in E. coli ranged from 5.5 to 8 with modification, which did not help in minimizing the
pH-dependent proteolytic activity or in maintaining the stability of the recombinant protein, especially the oxi-
dized and reduced forms of YlaD (data not shown). With many experimental trials, we obtained better quality
data in the presence of H
2
O
2
. In the same vein, CBBR-250 staining of YlaC metal-substituted YlaD (YlaD
Zn
,
YlaD
Mn
) and the YlaC–YlaD complex and staining of other proteins have been of questionable quality. Thus, to
complement our data, we had to consider other related experiments. These included UV CD spectra analysis and,
in particular, the presently described in vivo experiments on growth and sporulation. We had previously analyzed
the interaction between YlaC and YlaD in vitro. The present data are novel.
Anti-σfactors are believed to contain Zn as a cofactor [59,65], similar to redox-related metalloproteins [66].
The HX
3
CXXC motif is a redox-sensing domain and a metal-binding site. One example is RsrA, a ZAS-family
member from S.coelicolor that functions as a redox-sensing and Zn-based transcriptional regulator [44,65].
Others examples include Rhodobacter sphaeroides ChrR [37] and B.subtilis RsiW [67]. However, there has
been insufficient experimental evidence, particularly in B. subtilis, to prove the redox-sensing mechanism of
Zn-co-ordinated anti-σfactor YlaD, despite the detection of a highly conserved HX
3
CXXC motif that acts as a
redox switch platform similar to the aforementioned redox-sensing regulators. Several studies by Matsumoto
et al. [41] and Ryu et al. [42] have pointed to another method of analysis of the oxidative stress-sensing mech-
anism of B. subtilis ECF σfactors. The approach has not been tried to date. Currently, the best inspiration for
new ideas is the HX
3
CXXC motif in the RsrA of S. coelicolor, which interacts with the ECF σfactor, σ
R
,
through redox exchange [44,65]. In considering the still unclear mechanism of the association of redox changes
in YlaD, it is useful to consider our hypothesis of the crucial role of the transcriptional activity of YlaC linked
to the specific metal-dependent redox-sensing molecular switch in B. subtilis. To explore this line of reasoning,
convincing evidence was needed that YlaD is a member of the ZAS family. To the end, we measured the metal
ion content using a truncated YlaD variant lacking the C-terminal transmembrane domain to account for pre-
viously observed structural similarity and predicted oxidative stress-sensing regulation between B. subtilis YlaD
and S.coelicolor RsrA. Figure 1 and Table 2 represent convincing evidence that YlaD is indeed a member of
the Zn-dependent ZAS family. The data show that the reduced form of YlaD contains a near-stoichiometric
amount of both Zn and iron, in contrast with the absence of these ions in the oxidized form of YlaD.
Considering that the ylaABCD operon contains the P
yla
and Spx-dependent σ
A
promoters [41], our data
provide compelling evidence that the oxidized form of YlaD is released from the YlaC–YlaD complex and that
free YlaC auto-regulates its own promoter, P
yla
, because YlaC solely recognized the P
yla
promoter in vitro.
The established mechanism of the redox-sensing and Zn-co-ordinated anti-σfactor YlaD, which can regulate
the transcriptional activity of YlaC, is used by the ChrR and RsrA proteins to sense cellular redox variations
through a cysteine-containing Zn center. Thus, we next had to consider a cysteine and metal substitution strat-
egy for the target proteins to demonstrate the use of a cysteine-co-ordinating Zn center as the redox switch.
Many redox-regulated proteins, including GAPDH, OxyR, Yap1p, and p53, recruit one or more reactive cyst-
eine residues as efficient redox sensors. Other proteins, such as SoxR and FNR, use iron, which is a redox-
sensitive metal, as the redox sensor, and Zn as a redox-inert metal to ensure that all the redox chemistry occurs
at the co-ordinating cysteines, rather than the metal itself, and to minimize unwanted side reactions with water
or oxygen that results in the accumulation of additional ROS [68,69].
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2145
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
The expectation of cellular functions of Zn-binding proteins is bolstered by the demonstration that structural
Zn sites in proteins like Hsp33 [52], RsrA [27], and protein kinase C [53] commonly consist of
thiol-co-ordinated Zn sites as a cysteine-containing Zn center as the redox switch, which allows the sensing of
the inevitable release of ROS release and/or disulfide stress. This can occur because Zn acts as a Lewis acid and
accepts electron pairs from the thiol groups. The association of redox changes in B. subtilis YlaD was serendip-
itously inspired by the experimental evidence of the β-Glu activity of P
yla
:gusA and northern blot demonstra-
tion of the remarkable increase in the ylaC gene transcripts in the presence of MnCl
2
. The novel finding of the
loss of YlaD binding to YlaC in the presence of Mn may be one of the most crucial determinants of the YlaC–
YlaD interaction. Evidence for this statement comes from a recently discovered B. subtilis sporulation regula-
tory network, which differs from the known sporulation network in multiple endospore-forming bacteria that
features well-conserved and well-defined sporulation σfactors and the master regulator of sporulation, Spo0A,
along with strongly conserved sequential activation of these global regulators. These features differ markedly
from those for cells in the presence of Zn ions, which act as a promoter in the YlaC–YlaD interaction. The
finding that the ZAS-family member YlaD can sense a specific metal ion while functioning as a redox switch is
remarkable, although some Fur-family members, such as the PerR protein of B.subtilis and Staphylococcus
aureus, also have the same ability [58].
With the knowledge of the data acquisition power of experiments combining biochemical and biophysical
approaches, we experimentally planned and examined in detail how YlaD is able to alter its redox status to
affect YlaC transcriptional activity and whether the YlaC–YlaD interaction is affected by the active cysteine
and/or Zn ions are modified or substituted, and the influence of Mn
2+
concentrations. This approach was
rational, based on the demonstration that the introduction of Mn in response to a given oxidative stress is gen-
erally derepressed and that some types of iron-containing enzymes switch to a less redox-sensitive Mn form
[70]. Metabolic events in E. coli are iron-centric with a little conditional use of Mn. Also, a representative
Mn-dependent enzyme, Mn–superoxide dismutase (SOD), is usually non-metallated in unstressed cells. Unlike
E. coli,Mn
2+
is absolutely required for B. subtilis growth to maintain the intracellular content homeostasis of
Mn and iron ions. Mn serves as an important cofactor for metabolic enzymes, such as oxidoreductases, trans-
ferases, and hydrolases. Mn also prevents cells from oxidative damages, either as a cofactor for Mn-dependent
catalases and SOD or due to its inherent ability to quench free radical-mediated reactions. While Mn
2+
is
essential for the survival of most bacteria, including B. subtilis, it is toxic at elevated concentrations [17]. Cells
maintain Mn homeostasis by the tightly regulated expression of Mn
2+
transport systems. Therefore, based on
our previous [42] and present (Figures 1 and 2) data, we wondered whether YlaD expression in cells might
also regulate the in vivo YlaC transcriptional activity, specifically in cells grown in the presence of Mn
2+
and/or
under oxidative stress, both of which can affect YlaD activity. YlaD is an anti-σfactor for YlaC and transcrip-
tion from the P
yla
promoter in the yla operon is autoregulated by YlaC. Furthermore, YlaC is regulated by
redox changes in YlaD, prominently by cellular Mn
2+
(Figures 1 and 2). These observations prompted the sug-
gestion that B. subtilis YlaC interacts with YlaD under reducing conditions and, in turn, YlaC is inactivated by
interaction with YlaD. This would indicate that transcriptional activity of YlaC directly reflects redox changes
in YlaD, which might regulate the interaction with YlaC, depending on the Mn
2+
concentration in the cell.
Spore formation is stimulated by the addition of H
2
O
2
or MnCl
2
to the culture medium in ylaC and ylaD
mutants. Furthermore, the induction of sporulation through nutritional limitation and metal homeostasis alter-
ation has been well characterized [71]. However, the sporulation process, particularly initiation, via changes in
the redox state reflecting Zn, iron, and Mn concentrations remains unclear.
This study adopted a novel approach, which has never been adequately attempted in B. subtilis, and presents
evidence of redox regulation via ECF anti-σfactor YlaD against oxidative stresses in B. subtilis based on bio-
chemical and biophysical observations using recombinants of YlaC and YlaD and/or their mutants. We used
B. subtilis strains to explore how the transcriptional activity of the ECF σfactor YlaC is regulated by redox-
sensing roles of its anti-σfactor YlaD and how YlaD senses sporulation-inducing signals and affect redox
changes. Despite the paucity of literature supporting the involvement of ECF σfactors in oxidative stress-
sensing mechanisms in gram-positive bacteria, we utilized B. subtilis based on the finding that the S. coelicolor
redox-sensing anti-σfactor, RsrA, which possesses a thiol–disulfide redox switch that regulates σ
R
, since
mycothiol can reduce RsrA to bind σ
R
, so that the RsrA–σ
R
system senses the intracellular level of reduced
mycothiol. The finding is compelling evidence of a natural modulator of the ECF transcription system [65,72].
More support for our experimental evidence of YlaC–YlaD in this study comes from the description that the
activity of the ZAS-family member σ
W
, which contained a conserved HX
3
CXXC motif that forms disulfide
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2146
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
bonds [27,37], is induced by alkaline stress [38] and salt stress [39], rather than by oxidants. In contrast, YlaC,
which is another ECF σfactor in B. subtilis, is strongly induced by oxidants like diamide and H
2
O
2
, and contri-
butes to the resistance to these oxidants. YlaC transcriptional activity may be strongly associated with sporula-
tion given the sporulation rates in YlaC mutants. The mechanism of action appears to involve an interaction
between YlaC and the reduced form of the anti-σfactor, YlaD.
The foregoing observations have revealed the need for more functional studies regarding ECF σfactors in
bacteria. Nonetheless, the present results are a novel revelation of putative functions of the redox-sensing anti-σ
factor, YlaD, under the control of metal-sensing oxidative stress response. The biochemical and biophysical
approach involving metal substitution confirmed that YlaC, which interacts with YlaD, is a member of the ECF
subfamily of alternative σfactors, and that both YlaC and YlaD either drive or alter B. subtilis sporulation.
Furthermore, the present study is the first description of a redox-sensing ECF subfamily in the ylaABCD
operon under the control of a metal-sensing mechanism of increasing Mn
2+
concentration in B. subtilis.B. sub-
tilis is an exhaustively studied gram-positive model bacterium that is amenable to detailed genetic and biophys-
ical analyses of its oxidative stress response mechanisms [73,74]. The present observations using B. subtilis
provide an important piece of evidence in support of cellular utilization of alternative σsubunits to reprogram
RNA polymerase by triggering redox-sensing genes required for growth phase alteration, motility, stress
Figure 9. Proposed model for the function of YlaC and its regulation by anti-sigma-YlaC factor YlaD.
The revealed mechanism showed that the anti-σfactor YlaD controls the transcriptional activity of YlaC via both manganese-
dependent redox-sensing molecular switch and concomitant gene regulation of clpP and sigH in B. subtilis as described in the
text. Based on our finding, the results of this study demonstrates how metal-coordinated global regulators and σ- and anti-σ-
factors receive and sense various environmental stress signals directly associated with sporulation processes. We here
presented crucial evidence of the importance of redox regulation via anti-σfactor YlaD against oxidative stresses and metal
substitution governing sporulation regulatory network by biochemical and biophysical methods.
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2147
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
response, and sporulation [41,42]. Except for σ
Y
and σ
Z
, the cellular functions of other types of ECF σfactors
and their anti-σfactors have been further elucidated by biophysically demonstrating the YlaD redox state-
dependent modulation of YlaC transcriptional activity in B. subtilis. The findings with YlaC provide insight
and increased understanding of spore formation in B. subtilis. Concerning the contributing role of YlaC to
sporulation in B. subtilis, the ylaC gene is expressed during stationary growth, like other ECF σfactors.
Disruption of ylaC disruptant confers a seriously sporulation-defective phenotype [42]. In contrast with the
ylaD disruptant, a constitutively active mutant stimulated sporulation under oxidative stress and
Mn-supplemented conditions. These observations address how YlaC induces sporulation associated with
expression of genes related to the initiation of sporulation. The sigH and clpP genes, which can be induced by
various stress factors, were constitutively expressed in the ylaC-overexpressing mutant, while the ylaC-deletion
mutant showed very little expression of either of these genes. Furthermore, the clpP gene, which encodes a pro-
teolytic component of the Clp or Ti protease, is induced from its σ
A
-dependent promoter after exposure to
heat and puromycin [60]. Furthermore, ClpXP (a complex of ClpP and ClpX) controls the post-translation
activity of σ
H
[75]. The observations that the expression of the clpP gene was induced byH
2
O
2
and by the add-
ition of Mn, and that the ylaC disruptant had a low level of mRNA expression of the clpP gene, in contrast
with the ylaC-overexpressing mutant indicate that expression of the ylaC gene stimulates clpP gene expression.
This is consistent with mapping data of the 50-end of the clpP mRNA that revealed the transcriptional start site
recognized by YlaC 5–10 base pairs upstream of that of the σ
A
-dependent promoter. We show that YlaC stimu-
lates spore formation by inducing clpP gene expression. Thus, studies on ECF σfactors in B. subtilis are closely
linked to their roles contributing to sporulation. In the same vein, inactive YlaC bound to reduced YlaD is acti-
vated upon release from the YlaC–YlaD complex in response to oxidative stress. In this mechanism, the oxi-
dized YlaD forms a disulfide bond upon the release of Zn
2+
and the increasing concentration of Mn
2+
triggers
the dissociation of the YlaC–YlaD complex because Mn
2+
competes with other metals in binding to YlaD. The
resulting activated YlaC then associates with RNA polymerase and recognizes its specific promoters, such as
Pyla, as well as the promoter for the clpP gene. An increasing amount of ClpP stimulates cell sporulation
accompanied concomitantly by transcriptional changes in YlaC (Figure 9).
Abbreviations
β-Glu, β-glucuronidase; 2D-PAGE, two-dimensional polyacrylamide gel electrophoresis; AMS,
4-acetamido-40-maleimidylstilbene-2,20-disulfonic acid; DIG, digoxigenin; ECF, extracytoplasmic function; GST,
glutathione S-transferase; H
2
O
2
, hydrogen peroxide; IAA, iodoacetamide; Mn, manganese; PAGE,
polyacrylamide gel electrophoresis; RNAP, RNA polymerase; ROS, reactive oxygen species; SMM, Spizizen
minimal medium; SOD, superoxide dismutase; UV CD, ultraviolet circular dichroism; Zn, zinc.
Author Contribution
M.-K.K., H.-B.R., S.-H.S., J.-W.L., and S.-O.K. designed the research. M.-K.K., H.-B.R., and S.-H.S. performed
the research. M.-K.K., S.-H.S.., J.-W.L., and S.-O.K. analyzed the data. M.-K.K., H.-B.R., J.-W.L., and S.-O.K.
contributed new reagents/analytic tools, and M.-K.K., J.-W.L., and S.-O.K. wrote the paper.
Funding
This work was supported by the Research Fellowship of the BK21plus project.
Acknowledgements
We thank P. Setlow and P. J. Piggot for providing B. subtilis PS832 and plasmids.
Competing Interests
The Authors declare that there are no competing interests associated with the manuscript.
References
1 Pavlendová, N., Muchová, K. and Barák, I. (2007) Chromosome segregation in Bacillus subtilis.Folia Microbiol. 52, 563–572 https://doi.org/10.1007/
BF02932184
2 Cai, Y., Chandrangsu, P., Gaballa, A. and Helmann, J.D. (2017) Lack of formylated methionyl-tRNA has pleiotropic effects on Bacillus subtilis.
Microbiology 163, 185–196 https://doi.org/10.1099/mic.0.000413
3 Shin, S.-M., Song, S.-H., Lee, J.-W., Kwak, M.-K. and Kang, S.-O. (2017) Methylglyoxal synthase regulates cell elongation via alterations of cellular
methylglyoxal and spermidine content in Bacillus subtilis.Int. J. Biochem. Cell Biol. 91(Pt A), 14–28 https://doi.org/10.1016/j.biocel.2017.08.005
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2148
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
4 Woods, E.C. and McBride, S.M. (2017) Regulation of antimicrobial resistance by extracytoplasmic function (ECF) sigma factors. Microbes Infect. 19,
238–248 https://doi.org/10.1016/j.micinf.2017.01.007
5 Ilbert, M., Graf, P.C.F. and Jakob, U. (2006) Zinc center as redox switch—new function for an old motif. Antioxid. Redox Signal. 8, 835–846
https://doi.org/10.1089/ars.2006.8.835
6 Devkota, S.R., Kwon, E., Ha, S.C., Chang, H.W. and Kim, D.Y. (2017) Structural insights into the regulation of Bacillus subtilis SigW activity by
anti-sigma RsiW. PLoS ONE 12, e0174284 https://doi.org/10.1371/journal.pone.0174284
7 Helmann, J.D. (2014) Specificity of metal sensing: iron and manganese homeostasis in Bacillus subtilis.J. Biol. Chem. 289, 28112–28120 https://doi.
org/10.1074/jbc.R114.587071
8 Fimlaid, K.A. and Shen, A. (2015) Diverse mechanisms regulate sporulation sigma factor activity in the Firmicutes. Curr. Opin. Microbiol. 24,88–95
https://doi.org/10.1016/j.mib.2015.01.006
9 Imlay, J.A. (2003) Pathways of oxidative damage. Annu. Rev. Microbiol. 57, 395–418 https://doi.org/10.1146/annurev.micro.57.030502.090938
10 Kiley, P.J. and Storz, G. (2004) Exploiting thiol modifications. PLoS Biol. 2, e400 https://doi.org/10.1371/journal.pbio.0020400
11 Toledano, M.B., Delaunay, A., Monceau, L. and Tacnet, F. (2004) Microbial H
2
O
2
sensors as archetypical redox signaling modules. Trends Biochem. Sci.
29, 351–357 https://doi.org/10.1016/j.tibs.2004.05.005
12 Lee, J.-W. and Helmann, J.D. (2006) The PerR transcription factor senses H
2
O
2
by metal-catalysed histidine oxidation. Nature 440, 363–367
https://doi.org/10.1038/nature04537
13 Duarte, V. and Latour, J.-M. (2010) Perr vs OhrR: selective peroxide sensing in Bacillus subtilis.Mol. Biosyst. 6, 316–323 https://doi.org/10.1039/
B915042K
14 Zuber, P. (2004) Spx-RNA polymerase interaction and global transcriptional control during oxidative stress. J. Bacteriol. 186, 1911–1918 https://doi.org/
10.1128/JB.186.7.1911-1918.2004
15 Hecker, M. and Völker, U. (2001) General stress response of Bacillus subtilis and other bacteria. Adv. Microb. Physiol. 44,35–91 https://doi.org/10.
1016/S0065-2911(01)44011-2
16 Helmann, J.D., Wu, M.F.W., Gaballa, A., Kobel, P.A., Morshedi, M.M., Fawcett, P. et al. (2003) The global transcriptional response of Bacillus subtilis to
peroxide stress is coordinated by three transcription factors. J. Bacteriol. 185, 243–253 https://doi.org/10.1128/JB.185.1.243-253.2003
17 Guedon, E., Moore, C.M., Que, Q., Wang, T., Ye, R.W. and Helmann, J.D. (2003) The global transcriptional response of Bacillus subtilis to manganese
involves the MntR, Fur, TnrA and σB regulons. Mol. Microbiol. 49, 1477–1491 https://doi.org/10.1046/j.1365-2958.2003.03648.x
18 Chandrangsu, P., Rensing, C. and Helmann, J.D. (2017) Metal homeostasis and resistance in bacteria. Nat. Rev. Microbiol. 15, 338–350 https://doi.
org/10.1038/nrmicro.2017.15
19 Hantke, K. (2001) Iron and metal regulation in bacteria. Curr. Opin. Microbiol. 4, 172–177 https://doi.org/10.1016/S1369-5274(00)00184-3
20 Mills, S.A. and Marletta, M.A. (2005) Metal binding characteristics and role of iron oxidation in the ferric uptake regulator from Escherichia coli.
Biochemistry 44, 13553–13559 https://doi.org/10.1021/bi0507579
21 Helmann, J.D. (2006) Deciphering a complex genetic regulatory network: the Bacillus subtilis σW protein and intrinsic resistance to antimicrobial
compounds. Sci. Prog. 89, 243–266 https://doi.org/10.3184/003685006783238290
22 Daly, M.J., Gaidamakova, E.K., Matrosova, V.Y., Vasilenko, A., Zhai, M., Venkateswaran, A. et al. (2004) Accumulation of Mn(II) in Deinococcus
radiodurans facilitates gamma-radiation resistance. Science 306, 1025–1028 https://doi.org/10.1126/science.1103185
23 Outten, C.E. and O’Halloran, T.V. (2001) Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science 292, 2488–2492
https://doi.org/10.1126/science.1060331
24 Posey, J.E. and Gherardini, F.C. (2000) Lack of a role for iron in the Lyme disease pathogen. Science 288, 1651–1653 https://doi.org/10.1126/
science.288.5471.1651
25 Hightower, K.E. and Fierke, C.A. (1999) Zinc-catalyzed sulfur alkylation: insights from protein farnesyltransferase. Curr. Opin. Chem. Biol. 3, 176–181
https://doi.org/10.1016/S1367-5931(99)80030-1
26 Sobota, J.M., Gu, M. and Imlay, J.A. (2014) Intracellular hydrogen peroxide and superoxide poison 3-deoxy-D-arabinoheptulosonate 7-phosphate synthase,
the first committed enzyme in the aromatic biosynthetic pathway of Escherichia coli.J. Bacteriol. 196, 1980–1991 https://doi.org/10.1128/JB.01573-14
27 Paget, M.S.B., Bae, J.-B., Hahn, M.-Y., Li, W., Kleanthous, C., Roe, J.-H. et al. (2001) Mutational analysis of RsrA, a zinc-binding anti-sigma factor with
a thiol-disulphide redox switch. Mol. Microbiol. 39, 1036–1047 https://doi.org/10.1046/j.1365-2958.2001.02298.x
28 Li, W., Bottrill, A.R., Bibb, M.J., Buttner, M.J., Paget, M.S.B. and Kleanthous, C. (2003) The role of zinc in the disulphide stress-regulated anti-sigma
factor RsrA from Streptomyces coelicolor.J. Mol. Biol. 333, 461–472 https://doi.org/10.1016/j.jmb.2003.08.038
29 Feklístov, A., Sharon, B.D., Darst, S.A. and Gross, C.A. (2014) Bacterial sigma factors: a historical, structural, and genomic perspective. Annu. Rev.
Microbiol. 68, 357–376 https://doi.org/10.1146/annurev-micro-092412-155737
30 Lonetto, M.A., Brown, K.L., Rudd, K.E. and Buttner, M.J. (1994) Analysis of the Streptomyces coelicolor sigE gene reveals the existence of a subfamily
of eubacterial RNA polymerase sigma factors involved in the regulation of extracytoplasmic functions. Proc. Natl Acad. Sci. U.S.A. 91, 7573–7577
https://doi.org/10.1073/pnas.91.16.7573
31 Helmann, J.D. (2002) The extracytoplasmic function (ECF) sigma factors. Adv. Microb. Physiol. 46,47–110 https://doi.org/10.1016/S0065-2911(02)
46002-X
32 Asai, K. (2018) Anti-sigma factor-mediated cell surface stress responses in Bacillus subtilis.Genes Genet. Syst. 92, 223–234 https://doi.org/10.1266/
ggs.17-00046
33 Souza, B.M., de Paula Castro, T.L., Carvalho, R.D., Seyffert, N., Silva, A., Miyoshi, A. et al. (2014) σ
ECF
factors of gram-positive bacteria: a focus on
Bacillus subtilis and the CMNR group. Virulence 5, 587–600 https://doi.org/10.4161/viru.29514
34 Thackray, P.D. and Moir, A. (2003) SigM, an extracytoplasmic function sigma factor of Bacillus subtilis, is activated in response to cell wall antibiotics,
ethanol, heat, acid, and superoxide stress. J. Bacteriol. 185, 3491–3498 https://doi.org/10.1128/JB.185.12.3491-3498.2003
35 Ho, T.D., Hastie, J.L., Intile, P.J. and Ellermeier, C.D. (2011) The Bacillus subtilis extracytoplasmic function σfactor σ
V
is induced by lysozyme and
provides resistance to lysozyme. J. Bacteriol. 193, 6215–6222 https://doi.org/10.1128/JB.05467-11
36 Helmann, J.D. (2016) Bacillus subtilis extracytoplasmic function (ECF) sigma factors and defense of the cell envelope. Curr. Opin. Microbiol. 30,
122–132 https://doi.org/10.1016/j.mib.2016.02.002
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2149
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
37 Newman, J.D., Anthony, J.R. and Donohue, T.J. (2001) The importance of zinc-binding to the function of Rhodobacter sphaeroides ChrR as an
anti-sigma factor. J. Mol. Biol. 313, 485–499 https://doi.org/10.1006/jmbi.2001.5069
38 Wiegert, T., Homuth, G., Versteeg, S. and Schumann, W. (2001) Alkaline shock induces the Bacillus subtilis σW regulon. Mol. Microbiol. 41,59–71
https://doi.org/10.1046/j.1365-2958.2001.02489.x
39 Petersohn, A., Brigulla, M., Haas, S., Hoheisel, J.D., Völker, U. and Hecker, M. (2001) Global analysis of the general stress response of Bacillus subtilis.
J. Bacteriol. 183, 5617–5631 https://doi.org/10.1128/JB.183.19.5617-5631.2001
40 Asai, K., Yamaguchi, H., Kang, C.-M., Yoshida, K.-c., Fujita, Y. and Sadaie, Y. (2003) DNA microarray analysis of Bacillus subtilis sigma factors of
extracytoplasmic function family. FEMS Microbiol. Lett. 220, 155–160 https://doi.org/10.1016/S0378-1097(03)00093-4
41 Matsumoto, T., Nakanishi, K., Asai, K. and Sadaie, Y. (2005) Transcriptional analysis of the ylaABCD operon of Bacillus subtilis encoding a sigma factor
of extracytoplasmic function family. Genes Genet. Syst. 80, 385–393 https://doi.org/10.1266/ggs.80.385
42 Ryu, H.-B., Shin, I., Yim, H.-S. and Kang, S.-O. (2006) Ylac is an extracytoplasmic function (ECF) sigma factor contributing to hydrogen peroxide
resistance in Bacillus subtilis.J. Microbiol. 44, 206–216 PMID:16728958
43 Harwood, C., Cutting, S.M. and Chabert, R. (1990) Molecular Biological Methods for Bacillus. Wiley, Chichester
44 Kang, J.-G., Paget, M.S.B., Seok, Y.-J., Hahn, M.-Y., Bae, J.-B., Hahn, J.-S. et al. (1999) Rsra, an anti-sigma factor regulated by redox change. EMBO
J. 18, 4292–4298 https://doi.org/10.1093/emboj/18.15.4292
45 Van Dessel, W., Van Mellaert, L., Geukens, N., Lammertyn, E. and Anné, J. (2004) Isolation of high quality RNA from Streptomyces.J. Microbiol.
Methods 58, 135–137 https://doi.org/10.1016/j.mimet.2004.03.015
46 Kenney, T.J. and Moran, Jr, C.P. (1987) Organization and regulation of an operon that encodes a sporulation-essential sigma factor in Bacillus subtilis.
J. Bacteriol. 169, 3329–3339 https://doi.org/10.1128/jb.169.7.3329-3339.1987
47 Bollag, D.M., Rozycki, M.D. and Edelstein, S.J. (1996) Protein Methods, 2nd ed. Willey-Liss, NY
48 Karow, M.L., Glaser, P. and Piggot, P.J. (1995) Identification of a gene, spoIIR, that links the activation of sigma E to the transcriptional activity of sigma
F during sporulation in Bacillus subtilis.Proc. Natl Acad. Sci. U.S.A. 92, 2012–2016 https://doi.org/10.1073/pnas.92.6.2012
49 Milhaud, P. and Balassa, G. (1973) Biochemical genetics of bacterial sporulation. IV. Sequential development of resistances to chemical and physical
agents during sporulation of Bacillus subtilis.Mol. Gen. Genet. 125, 241–250 https://doi.org/10.1007/BF00270746
50 Shin, I., Ryu, H.-B., Yim, H.-S. and Kang, S.-O. (2005) Cytochrome c550 is related to initiation of sporulation in Bacillus subtilis.J. Microbiol. 43,
244–250 PMID:15995641
51 Herzberg, O. and James, M.N.G. (1988) Refined crystal structure of troponin C from Turkey skeletal muscle at 2.0 Å resolution. J. Mol. Biol. 203,
761–779 https://doi.org/10.1016/0022-2836(88)90208-2
52 Graf, P.C.F. and Jakob, U. (2002) Redox-regulated molecular chaperones. Cell. Mol. Life Sci. 59, 1624–1631 https://doi.org/10.1007/PL00012489
53 Gopalakrishna, R. and Jaken, S. (2000) Protein kinase C signaling and oxidative stress. Free Radic. Biol. Med. 28, 1349–1361 https://doi.org/10.1016/
S0891-5849(00)00221-5
54 Kunst, F., Ogasawara, N., Moszer, I., Albertini, A.M., Alloni G., Azevedo V. et al. (1997) The complete genome sequence of the gram-positive bacterium
Bacillus subtilis.Nature 390, 249–256 https://doi.org/10.1038/36786
55 Yoshimura, M., Asai, K., Sadaie, Y. and Yoshikawa, H. (2004) Interaction of Bacillus subtilis extracytoplasmic function (ECF) sigma factors with the
N-terminal regions of their potential anti-sigma factors. Microbiology 150, 591–599 https://doi.org/10.1099/mic.0.26712-0
56 Cao, M., Moore, C.M. and Helmann, J.D. (2005) Bacillus subtilis paraquat resistance is directed by σM, an extracytoplasmic function sigma factor, and
is conferred by YqjL and BcrC. J. Bacteriol. 187, 2948–2956 https://doi.org/10.1128/JB.187.9.2948-2956.2005
57 Vallee, B.L. and Falchuk, K.H. (1993) The biochemical basis of zinc physiology. Physiol. Rev. 73,79–118 https://doi.org/10.1152/physrev.1993.73.1.79
58 Lee, J.-W. and Helmann, J.D. (2007) Functional specialization within the Fur family of metalloregulators. BioMetals 20, 485–499 https://doi.org/10.
1007/s10534-006-9070-7
59 Zdanowski, K., Doughty, P., Jakimowicz, P., O’Hara, L., Buttner, M.J., Paget, M.S.B. et al. (2006) Assignment of the zinc ligands in RsrA, a
redox-sensing ZAS protein from Streptomyces coelicolor.Biochemistry 45, 8294–8300 https://doi.org/10.1021/bi060711v
60 Gerth, U., Krüger, E., Derré, I., Msadek, T. and Hecker, M. (1998) Stress induction of the Bacillus subtilis clpP gene encoding a homologue of the
proteolytic component of the Clp protease and the involvement of ClpP and ClpX in stress tolerance. Mol. Microbiol. 28, 787–802 https://doi.org/10.
1046/j.1365-2958.1998.00840.x
61 Gerth, U., Wipat, A., Harwood, C.R., Carter, N., Emmerson, P.T. and Hecker, M. (1996) Sequence and transcriptional analysis of clpX, a class-III
heat-shock gene of Bacillus subtilis.Gene 181,77–83 https://doi.org/10.1016/S0378-1119(96)00467-2
62 Zuber, P. (2009) Management of oxidative stress in Bacillus.Annu. Rev. Microbiol. 63, 575–597 https://doi.org/10.1146/annurev.micro.091208.
073241
63 Mutlu, A., Trauth, S., Ziesack, M., Nagler, K., Bergeest, J.-P., Rohr, K. et al. (2018) Phenotypic memory in Bacillus subtilis links dormancy entry and
exit by a spore quantity-quality tradeoff. Nat. Commun. 9,69https://doi.org/10.1038/s41467-017-02477-1
64 Ursula, J. and Dana, R. (2013) Oxidative Stress and Redox Regulation, Springer
65 Bae, J.-B., Park, J.-H., Hahn, M.-Y., Kim, M.-S. and Roe, J.-H. (2004) Redox-dependent changes in RsrA, an anti-sigma factor in Streptomyces
coelicolor: zinc release and disulfide bond formation. J. Mol. Biol. 335, 425–435 https://doi.org/10.1016/j.jmb.2003.10.065
66 Clarke, N.D. and Berg, J.M. (1998) Zinc fingers in Caenorhabditis elegans:finding families and probing pathways. Science 282, 2018–2022 https://doi.
org/10.1126/science.282.5396.2018
67 Schöbel, S., Zellmeier, S., Schumann, W. and Wiegert, T. (2004) The Bacillus subtilis σW, anti-sigma factor RsiW is degraded by intramembrane
proteolysis through YluC. Mol. Microbiol. 52, 1091–1105 https://doi.org/10.1111/j.1365-2958.2004.04031.x
68 Touati, D. (2000) Sensing and protecting against superoxide stress in Escherichia coli—how many ways are there to trigger soxRS response? Redox
Rep. 5, 287–293 https://doi.org/10.1179/135100000101535825
69 Kiley, P.J. and Beinert, H. (2003) The role of Fe–S proteins in sensing and regulation in bacteria. Curr. Opin. Microbiol. 6, 181–185 https://doi.org/10.
1016/S1369-5274(03)00039-0
70 Anjem, A. and Imlay, J.A. (2012) Mononuclear iron enzymes are primary targets of hydrogen peroxide stress. J. Biol. Chem. 287, 15544–15556
https://doi.org/10.1074/jbc.M111.330365
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society2150
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
71 Grossman, A.D. and Losick, R. (1988) Extracellular control of spore formation in Bacillus subtilis.Proc. Natl Acad. Sci. U.S.A. 85, 4369–4373
https://doi.org/10.1073/pnas.85.12.4369
72 Park, J.-H. and Roe, J.-H. (2008) Mycothiol regulates and is regulated by a thiol-specific antisigma factor RsrA and sigma(R) in Streptomyces coelicolor.
Mol. Microbiol. 68, 861–870 https://doi.org/10.1111/j.1365-2958.2008.06191.x
73 Antelmann, H., Hecker, M. and Zuber, P. (2008) Proteomic signatures uncover thiol-specific electrophile resistance mechanisms in Bacillus subtilis.
Expert Rev. Proteomics 5,77–90 https://doi.org/10.1586/14789450.5.1.77
74 Hecker, M. and Völker, U. (2004) Towards a comprehensive understanding of Bacillus subtilis cell physiology by physiological proteomics. Proteomics 4,
3727–3750 https://doi.org/10.1002/pmic.200401017
75 Liu, J., Cosby, W.M. and Zuber, P. (1999) Role of lon and ClpX in the post-translational regulation of a sigma subunit of RNA polymerase required for
cellular differentiation in Bacillus subtilis.Mol. Microbiol. 3, 415–428 https://doi.org/10.1046/j.1365-2958.1999.01489.x
76 Hanahan, D. (1983) Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557–580 https://doi.org/10.1016/S0022-2836(83)
80284-8
77 Studier, F.W. (1991) Use of bacteriophage T7 lysozyme to improve an inducible T7 expression system. J. Mol. Biol. 219,37–44 https://doi.org/10.
1016/0022-2836(91)90855-Z
78 Yanisch-Perron, C., Vieira, J. and Messing, J. (1985) Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and
pUC19 vectors. Gene. 33, 103–119 https://doi.org/10.1016/0378-1119(85)90120-9
© 2018 The Author(s). Published by Portland Press Limited on behalf of the Biochemical Society 2151
Biochemical Journal (2018) 475 2127–2151
https://doi.org/10.1042/BCJ20170911
