FEMS Microbiology Reviews, fux015, 41, 2017, 549–574
doi: 10.1093/femsre/fux015
Advance Access Publication Date: 11 April 2017
Review Article
REVIEW ARTICLE
Regulation of heat-shock genes in bacteria: from
signal sensing to gene expression output
Davide Roncarati∗
and Vincenzo Scarlato
Department of Pharmacy and Biotechnology (FaBiT), University of Bologna, 40126 Bologna, Italy
∗
Corresponding author: Department of Pharmacy and Biotechnology (FaBiT), University of Bologna, via Selmi, 3, 40126 Bologna, Italy.
Tel: (+39) 051 2099320; Fax: (+39) 051 2094286; E-mail: davide.roncarati@unibo.it
One sentence summary: The authors provide an updated review of the strategies employed by many important bacteria to finely regulate transcription
of heat-shock genes, including a comprehensive description of the several ways in which environmental cues are perceived and processed.
Editor: Franz Narberhaus
ABSTRACT
The heat-shock response is a mechanism of cellular protection against sudden adverse environmental growth conditions
and results in the prompt production of various heat-shock proteins. In bacteria, specific sensory biomolecules sense
temperature fluctuations and transduce intercellular signals that coordinate gene expression outputs. Sensory
biomolecules, also known as thermosensors, include nucleic acids (DNA or RNA) and proteins. Once a stress signal is
perceived, it is transduced to invoke specific molecular mechanisms controlling transcription of genes coding for heat-shock
proteins. Transcriptional regulation of heat-shock genes can be under either positive or negative control mediated by
dedicated regulatory proteins. Positive regulation exploits specific alternative sigma factors to redirect the RNA polymerase
enzyme to a subset of selected promoters, while negative regulation is mediated by transcriptional repressors. Interestingly,
while various bacteria adopt either exclusively positive or negative mechanisms, in some microorganisms these two
opposite strategies coexist, establishing complex networks regulating heat-shock genes. Here, we comprehensively
summarize molecular mechanisms that microorganisms have adopted to finely control transcription of heat-shock genes.
Keywords: heat-shock response; transcriptional regulation; alternative σ factors; HspR repressor; HrcA repressor; CtsR
repressor
INTRODUCTION
The heat-shock response is a protective mechanism that is crucial
for bacterial survival and adaptation to hostile environmental
conditions. This response is apparently universal as it has
been observed in every bacterial species investigated. It consists
of a set of well-coordinated responses and processes, mostly involving
the strictly regulated production of various heat-shock
proteins. These, in turn, mainly include molecular chaperones
and proteases, whose intracellular abundance rapidly increases
upon exposure to a variety of environmental stresses. The principal
function of heat-shock proteins is to assist protein folding.
They are also involved in rescue or degradation of denatured
proteins and deleterious misfolded aggregates. Upon
sudden temperature increase, heat-shock proteins transiently
accumulate in the cell. Once the organism has adapted to the
new temperature, the amount of heat-shock proteins decreases
to a steady-state level that is frequently greater than the initial
basal level in order to assist bacterial growth under non-optimal
environmental conditions.
Besides their roles in protecting cellular proteins from environmental
insults and in maintaining cellular homeostasis,
some heat-shock proteins are important virulence factors, while
others appear to affect pathogenesis indirectly. Although the
heat-shock response is universally conserved in both prokaryotes
and eukaryotes, the basic molecular mechanisms governing
the regulation of heat-shock genes differ considerably among
Received: 30 November 2016; Accepted: 14 March 2017
C FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
549
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
550 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
bacterial species. In particular, bacteria have evolved different
regulatory strategies that combine transcriptional and posttranscriptional
mechanisms to rapidly induce heat-shock protein
synthesis only when required.
In order to rapidly respond to environmental cues and to ensure
expression of heat-shock proteins only when appropriate,
bacteria employ specific sensory biomolecules that are able to
perceive temperature fluctuations and translate them into coordinated
gene expression outputs. In this respect, a wide repertoire
of thermal sensors has evolved among microorganisms.
Specifically, bacteria can sense temperature changes using nucleic
acids (DNA or RNA) or proteins.
In the present review article, we provide an updated, comprehensive
description of the strategies employed by many important
microorganisms to finely regulate transcription of genes
coding for heat-shock proteins. In the first part of the article,
we survey the molecular mechanisms of regulation (e.g. positive,
negative, simple and complex) that guarantee the proper
transcriptional output once stress signals are detected. In the
second part, we provide a comprehensive description of the
several ways in which environmental cues are perceived and
processed.
The heat-shock response: role and principal mediators
The heat-shock response was first described by Ritossa as a
phenomenon observed in the polytene chromosomes of salivary
glands of Drosophila melanogaster flies when moved from
their normal growth temperature to 37◦
C (Ritossa 1962). Triggered
by environmental stress insults, the heat-shock response
consists of the induction or upregulation of synthesis of a group
of proteins that help the cell to survive under conditions that
would normally be lethal. Actually, several stress conditions
are able to elicit heat-shock response, such as temperature
variations, osmotic changes, desiccation, antibiotics, solvents
or heavy metals. However, the heat-shock response triggered
by a sudden temperature increase has been widely used as a
model system for studying the impact of stress on biological
systems. Environmental stress conditions primarily affect cellular
protein homeostasis (maintenance of an overall balance of
folded proteins), posing a serious threat to their integrity. Heatshock
proteins are involved in several processes in bacterial
cells, including assisting the folding of newly synthesized proteins,
preventing aggregation of proteins under stress conditions
and recovering proteins that have been partially or completely
unfolded by stresses such as a sudden temperature increase.
During stress, in fact, spontaneous folding of newly synthetized
proteins is inefficient and error-prone, and a predominant fraction
of nascent polypeptides undergoes a chaperone-mediated
folding process (Mogk et al. 1999). Also, a large fraction of already
folded proteins gets partially or completely denatured and
becomes prone to form deleterious aggregates. Moreover, the
rates of transcription and translation decrease because the stability
of the RNA polymerase subunits and of the translation
factor EF-G are sensitive to temperature increase (Mogk et al.
1999).
Some heat-shock proteins are also required during normal
growth condition and they are abundant under all metabolic
conditions. GroEL and DnaK, the bacterial representatives of
the two major chaperone families Hsp60 and Hsp70, respectively,
play a key role in protein folding even during nonstressed
growth conditions, although their action becomes
more important during stress. They bind hydrophobic surfaces
of unfolded proteins and, together with their co-chaperones
(GroES and DnaJ-GrpE) and ATP hydrolysis, they promote
the acquisition of proper folding by the substrate polypeptides
through different mechanisms that reflect their different
architectures. While GroEL monomers assemble into
a cylindrical complex to form two heptameric rings that
enclose the entire substrate protein inside a large cavity, giving
the polypeptide a chance to fold without interacting with
any other protein, the DnaK chaperone is mostly monomeric,
exerting its action by binding short surface-exposed hydrophobic
amino-acid sequences (Xu, Horwich and Sigler 1997; Mogk
et al. 1999).
It is worth mentioning that the names groEL and groES should
be used for Escherichia coli genes only, because this nomenclature
was adopted following the observation that mutation of
these genes prevented the plating of several bacteriophages
(Georgopoulos et al. 1972; Takano and Kakefuda 1972). An alternative
used nomenclature has been proposed, in which homologs
of GroEL are called Cpn60, and those of GroES are named
Cpn10 (Coates, Shinnick and Ellis 1993). As both nomenclatures
are still widely used, in this review we decided to adopt the
nomenclature used for E. coli, GroEL and GroES.
Another group of heat-shock proteins expressed by the
bacterial cell and upregulated during stress is composed by
proteases. These proteins are responsible essentially for the
removal of damaged polypeptides from stressed cells. Some proteases
are multicomponent systems, in which a catalytic subunit
(e.g. ClpP and HslV) associates to substrate recognition
subunits (ClpA or ClpX for ClpP and HslU for HslV), which are
co-chaperones able to remodel substrate polypeptides upon
ATP hydrolysis and deliver them to proteolytic degradation
(Missiakas et al. 1996; Wawrzynow, Banecki and Zylicz 1996).
While these proteases assemble into complex ring-shaped
structures, other members of this group of heat-shock proteins
combine on a single polypeptide both chaperone and protease
activities (e.g. Lon, FtsH and the periplasmic serine protease
DegP).
The small heat-shock proteins constitute a highly heterogeneous
group of proteins whose expression is induced upon
stress challenges. Their principal role is to bind and protect unfolded
proteins, by holding them in a conformation not subjected
to degradation, until they will be efficiently refolded by
ATP-driven chaperones (Matuszewska et al. 2005).
Finally, some heat-shock proteins have been proposed as important
virulence factors, while others appear to affect pathogenesis
indirectly. In this respect, there are several examples
in which molecular chaperones seem to have taken on diverse
functions other than just protein folding. For example, GroEL1 of
Mycobacterium smegmatis, a GroEL paralog with no heat-shockrelated
functions, is involved in mycolic acid synthesis and
biofilm formation (Ojha et al. 2005). Moreover, the GroES homolog
of Helicobacter pylori, besides its prototypical role as cochaperonin,
plays a role in the storage and trafficking of Ni2+
ions, a crucial virulence determinant for this human gastric
pathogen (de Reuse, Vinella and Cavazza 2013). On the other
hand, there is evidence that molecular chaperones in many
cases might act as direct virulence factors. Even though it still
remains a controversial concept, it has been proposed that several
bacterial species use cell surface GroEL and DnaK chaperones
as adhesins during the interaction with host cells (reviewed
in Henderson and Martin 2011). Moreover, chaperones proteins
have been shown to have cell-to-cell signaling properties and, in
several cases, are able to induce the synthesis of proinflammatory
cytokines and promote apoptosis (Henderson and Martin
2011).
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 551
REGULATORY MECHANISMS OF HEAT-SHOCK
GENES
In bacteria, regulation of heat-shock genes can be positive or
negative. Positive regulation exploits alternative sigma factors
to specifically direct the transcriptional machinery to a subset
of selected promoters, while negative regulation is orchestrated
by transcriptional repressors. Interestingly, while several bacteria
adopt exclusively positive or negative mechanisms, in some
microorganisms these opposite strategies coexist, establishing
a complex regulatory network of heat-shock genes.
Positive transcriptional regulation of heat-shock genes
Positive transcriptional regulation of heat-shock genes relies on
the use of a dedicated alternative σ factor, the subunit of the
RNA polymerase that confers promoter recognition specificity to
the transcription enzyme. In general, this transcriptional control
mechanism exploits the ability of condition-specific alternative
σ factors to outcompete the housekeeping σ subunit normally
associated with RNA polymerase and redirects the enzyme to
a selected subset of heat-shock gene promoters, thereby reprogramming
cellular transcription. The strategy of transiently
increasing transcription of heat-shock genes being dependent
upon temperature stress through a dedicated σ factor is widely
used among bacteria, although the model organism Escherichia
coli has long served as the paradigm for this kind of regulation.
Escherichia coli heat-shock sigma-32 (σ32
)
The heat-shock response in E. coli is governed by the htpR gene
product, whose high amino acid sequence similarity with the
housekeeping σ70
was previously described (Landick et al. 1984).
The 32-kDa HtpR polypeptide was purified from crude extracts
and combined in vitro with RNA polymerase ‘core’ enzyme, consisting
of the four subunits α2, β and β’only (a preparation of
the RNA polymerase enzyme that retains catalytic activity but
is unable to initiate de novo transcription in vitro). This in vitro
transcription assay showed that HtpR was able to confer specificity
to RNA polymerase in recognizing heat-shock promoters,
without requiring the vegetative sigma factor, σ70
. Following the
demonstration that HtpR was a σ factor able to promote transcription
initiation at heat-shock promoters, its gene was renamed
rpoH (in analogy to the already characterized rpoD gene,
coding for the vegetative σ70
factor) and the gene product was
called sigma-32 (σ32
, sometimes also called σH
) (Grossman, Erickson
and Gross 1984). As stated above, to exert its function,
σ32
interacts with RNA polymerase core enzyme and directs the
transcription machinery to specific promoters. The vegetative σ
factor and almost all alternative σ factors constitute a homologous
set of proteins (except σ54
, which belongs to a separate
family), the σ70
family of proteins, with several regions of sequence
conservation. The extensive σ-core interaction in the
formation of the RNA polymerase holoenzyme (the enzymatic
form with α2ββ’σ subunit composition) involves the same regions
and even equivalent residues in σ32
and σ70
(Sharp et al.
1999). What specifically targets the transcription machinery to
heat-shock genes is the recognition of particular promoter elements
by σ32
DNA-binding domains (DBD) (Kumar et al. 1995).
Several genome-wide studies that addressed the definition of
heat-shock promoter elements confirmed that those recognized
by σ32
are different from those bound by RNA polymerase associated
with the housekeeping σ70
. Specifically, while the –35
box consensus sequence for σ32
binding, TTGAAA, is highly similar
to the σ70
TTGACA consensus, the σ32
specific CCCCATNT
-10 box shows no similarities to the σ70
TATAAT -10 consensus
hexamer (Nonaka et al. 2006; Wade et al. 2006; Koo et al. 2009).
Accordingly, a mutational analysis of residues belonging to σ32
regions 2.4 and 4.2 (conserved regions among the various σ
factors belonging to the σ70
family, which are involved in the
recognition of –10 and –35 core promoter elements, respectively)
showed that the recognition of the –35 box is mediated by similar
amino acids clustered in regions 4.2 of σ32
and σ70
, while
the interaction with the downstream –10 element involves different
residues (Kourennaia, Tsujikawa and Dehaseth 2005). The
rapid, transient transcriptional increase of heat-shock genes
upon shift to higher temperature is directly linked to the amount
of active σ32
in the cell. Normally present in very small amounts
and with a half-life of just a few minutes at 37◦
C, the σ32
level
rapidly and transiently increases during heat-shock as a result of
changes in its rate of synthesis and stability (Grossman et al 1987;
Lesley, Thompson and Burgess 1987; Straus, Walter and Gross
1987). Even though the cellular level of σ32
depends mostly on
post-transcriptional regulatory mechanisms, the control of rpoH
transcription appears to be quite complex. The region upstream
of the rpoH coding sequence harbors five promoters (named P1
and P3 through P6) recognized by different σ factors and responsive
to several stimuli (Fig. 1, left panel). Specifically, P4 and P5
are transcribed by σ70
-associated RNA polymerase and P3 and
P6 are bound by σ24
and σ54
, respectively, while P1 can be transcribed
by σ70
or σS
depending on the growth phase (Erickson
et al. 1987; Erickson and Gross 1989; Wang and Kaguni 1989a;
Nagai et al. 1990; Missiakas and Raina 1998; Pallen 1999; Janaszak
et al. 2009). In addition, the P6 promoter is catabolite sensitive
and its transcription is dependent on cAMP levels and
on the CRP activator (Nagai et al. 1990), while cAMP-CRP together
with CytR anti-activator form a repression complex overlapping
P3, P4 and P5 (Kallipolitis and Valentin-Hansen 1998).
Also DnaA, the DNA-binding protein involved in initiation of
chromosomal DNA replication, takes part in the regulation of
rpoH by repressing the P3 and P4 promoters (Wang and Kaguni
1989b) (Fig. 1, left panel). This complex regulatory network is responsive
to various environmental signals and conditions, providing
the proper amount of rpoH transcript and, accordingly,
an extensive adaptability of the stress response. The hypothesis
that the transient increase of cellular σ32
amount following
heat stress occurs at the translational level has been proposed
according to independent observations. First, the striking
increase of the rate of synthesis of σ32
upon heat-shock
does not correlate with a significant increase of rpoH transcription
(Erickson et al. 1987). Second, to understand the contribution
of transcriptional and/or post-transcriptional mechanisms
to the expression of σ32
, the activity of a reporter gene (lacZ encoding
the β-galactosidase enzyme) was monitored using transcriptional
and translational fusions with the 5 region of rpoH.
Results from these analyses showed that temperature regulated
increase of σ32
was not dependent on the rpoH promoter,
but was controlled by the translational signals of RpoH mRNA,
suggesting a mechanism of translational control for σ32
during
heat-shock (Straus, Walter and Gross 1987). Several studies
have identified the regions of rpoH mRNA important for
proper σ32
regulation, suggesting a mechanism involving the
mRNA secondary structure formed by the 5 coding sequence
of rpoH close to the translation start codon and, thereby, affecting
translation efficiency (Kamath-Loeb and Gross 1991; Nagai,
Yuzawa and Yura 1991; Yuzawa et al. 1993). All these findings
converged in the current model of temperature-mediated translational
regulation of σ32
. Specifically, during normal growth
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
552 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
Figure 1. Transcriptional and translational control of σ32
expression. Left panel: schematic representation of the region upstream of the rpoH coding sequence (depicted
by a gray block arrow) containing five promoters (indicated by bent arrows and named P1 and P3 to P6) recognized by different σ factors (indicated underneath
each promoter name). Transcription from some of these promoters is also regulated by transcriptional regulators: the arrowhead denotes positive regulation, while
hammerheads indicate negative regulation. Central panel: schematic representation of the RpoH mRNA secondary structure at physiological temperature. Transcript
sequences of crucial importance for the efficient initiation of translation close to the ribosome-binding site and the AUG start codon are sequestered in double-stranded
structures, which leads to inefficient ribosome binding (gray dashed circle indicates the portion of the transcript containing these translation initiation sequences).
Right panel: enlarged view of the RpoH mRNA region involved in the initiation of translation. While during normal growth conditions ribosome binding is hindered by
secondary structure formation, at high temperature the partial melting of the mRNA secondary structure enhances ribosome entry and translational initiation. Black
lines indicate Watson–Crick base pairing.
conditions, the rpoH mRNA folds in a complex secondary
structure (schematically represented in Fig. 1, central panel) that
hides sequences crucial for the efficient initiation of translation
surrounding the ribosome-binding site and the AUG start codon.
Conversely (Fig. 1, right panel), at high temperature the partial
melting of the mRNA secondary structure enhances ribosome
entry and translational initiation without the involvement of additional
cellular components (Morita et al. 1999). rpoH represents
the first described example of a class of temperature-sensing
RNA sequences known as RNA thermometers (Kortmann and
Narberhaus 2012), which provide a rapid post-transcriptional
mechanism to control the synthesis of heat-shock proteins (discussed
below). The temperature-mediated positive regulation
of σ32
synthesis through structured mRNA combines with the
complex control of the heat-shock σ factor activity and stability,
mediated by the concerted action of chaperones and proteases.
The finding that E. coli strains carrying mutations in the
dnaK, dnaJ and grpE genes were characterized by enhanced expression
of heat-shock proteins at low temperature and by a prolonged
transcriptional response upon heat challenge suggested
a role for the DnaK-DnaJ-GrpE machinery as a negative regulator
of heat-shock response (Straus, Walter and Gross 1990).
The key role of the DnaK system in the regulation of heat-shock
gene expression was further characterized, and a direct physical
ATP-dependent interaction between σ32
and DnaK, DnaJ and
GrpE chaperones was demonstrated both in vitro and in vivo
(Gamer, Bujard and Bukau 1992; Liberek et al. 1992). In addition,
the σ32
functionality is modulated by another major heat-shock
protein, the chaperonin GroEL, as its depletion determined enhanced
transcription of heat-shock genes, while its overexpression
established the opposite effect (Kanemori, Mori and Yura
1994; Guisbert et al. 2004; Patra et al. 2015). The activity of σ32
appears
to be dependent on its direct interaction with free chaperones
(Guisbert et al. 2004). When unfolded proteins accumulate
in the cell, they bind to and sequester chaperones, allowing σ32
to dissociate as a stable protein. In other words, it is widely accepted
that the degree of sequestration of the DnaKJE/GroESL
chaperones through binding to misfolded proteins determines
the level of transcription of heat-shock genes, linking in this way
the transcriptional response to the cellular folding state of proteins
(Tomoyasu et al. 1998). The effect of chaperones on σ32
stability
is directly linked to specific cellular proteases involved in
its degradation. In particular, it was shown that an E. coli strain
depleted of the membrane-bound ATP-dependent FtsH metalloprotease
(formerly named HflB) accumulated σ32
and induced
the heat-shock response. Furthermore, the half-life of the highly
unstable σ factor increased by a factor of up to 12-fold in mutants
with reduced FtsH function (Herman et al. 1995). Moreover,
in vitro assays with purified components confirmed the direct
association and the FtsH-mediated processing of σ32
(Tomoyasu
et al. 1995). Also, additional cellular proteases, such as the
ATP-dependent protease HslVU (also known as ClpQY), seem to
play a role in the heat-shock response by modulating in vivo the
turnover of σ32
(Kanemori et al. 1997). All the experimental data
converged to a model for the homeostatic control of σ32
concentration
in the cell in response to environmental stimuli. Normally
present in low amounts and highly unstable, upon temperature
upshift, σ32
levels rapidly increase as a consequence
of enhanced mRNA translation and transient stabilization (positive
feedforward mechanism). This induction phase is followed
by a recovery/shut-off phase, mediated by chaperones and proteases
(known as negative feedback loops), in which the activity
and the stability of σ32
rapidly decrease to a new steady-state
level. Even though the role of the DnaK and GroE chaperone
systems in modulating σ32
activity and in promoting proteasemediated
degradation was well documented in vivo, some contradictory
observations had for many years hampered its recapitulation
in vitro and the detailed mechanism remained elusive.
Specifically, several studies, focused on the identification and
the characterization of mutants with altered homeostatic regulation
(known as ‘dysregulation mutants’), led to the definition
of a short region of σ32
(named homeostatic control region) that
appeared to be important for both stability and activity of σ32
(Obrist and Narberhaus 2005; Yura et al. 2007). Escherichia coli cells
expressing heat-shock σ factor with mutations in the homeostatic
control region showed increased σ32
activity and stability,
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 553
Figure 2. Representation of the current model for activity and stability control of σ32
, depicting its interaction with chaperones, its transit to the inner membrane and
its FtsH protease-mediated degradation. Gray arrowheads denote negative modulation.
consistent with its inferred role in the feedback control mechanism
(Obrist and Narberhaus 2005; Yura et al. 2007). Surprisingly,
purified dysregulation mutant σ32
isoforms showed normal
binding to chaperones in in vitro interaction assays and were
not inhibited by chaperones when assayed by in vitro transcription
experiments (Yura et al. 2007; Suzuki et al. 2012). These discrepancies
suggested that the negative feedback control of σ32
mediated by DnaK and GroE chaperones must be more complex
than previously believed and that there must be at least an additional
step involving some other cellular components besides
chaperone binding. This missing step was recently characterized
by Lim et al. (2013). In particular, they showed that σ32
is
associated to the inner membrane instead of being distributed
throughout the cytoplasm, as previously thought. This membrane
association is mediated by the signal recognition particle
(SRP) in combination with the SRP receptor (SR). This cotranslational
protein targeting machinery (SRP-SR), normally acting
only on periplasmic and membrane proteins, specifically targets
σ32
to the inner membrane, which is crucial for proper homeostatic
control upon heat-shock. The homeostatic control region
of σ32
is directly involved in this process. Using in vivo crosslinking
assays, it was recently shown that the homeostatic control
region of σ32
encompasses the binding site recognized by
a subunit of SRP (Miyazaki et al. 2016). These latter findings allowed
for revision of the model for activity and stability control
of σ32
as shown in Fig. 2. It is worth noting that membrane localization
of σ32
may allow the heat-shock regulator to monitor protein
folding status both in the cytoplasm and in the membrane,
linking cytoplasmic and membrane protein damage to the heatshock
response.
The regulation of gene expression through the use of alternative
sigma factors guarantees a global transcriptional reprogramming
of the bacterial transcriptome in response to changing
environmental conditions. In this light, it is not surprising
that σ32
has a broad regulon, controlling the transcription of
hundreds of genes upon temperature increase. Several genomewide
studies focusing on the characterization of the σ32
regulon
monitored gene expression changes upon inducible overexpression
of σ32
using microarrays (Zhao, Liu and Burgess 2005;
Nonaka et al. 2006). Besides known genes encoding proteases
and chaperones involved in protein homeostasis, these analyses
allowed the identification of σ32
-dependent targets involved
in other key cellular processes. These processes include protection
of both DNA and RNA, enhancement of transcription
and translation at high temperature, as well as genes coding
for other transcriptional regulators, thus propagating the response.
Moreover, the finding that several σ32
-dependent genes
have important roles in membrane protection and functionality
suggested for the first time an intimate link between σ32
regulation
and membrane homeostasis (Nonaka et al. 2006).
Combining chromatin immunoprecipitation with microarrays
(ChIP-to-chip), Wade et al. (2006) identified in vivo several new
σ32
promoters and compared their locations with the positions
of σ70
promoters determined in a separate study. Notably, they
demonstrated that the majority of σ32
promoters overlaps with
σ70
promoters, with promoter elements typical of one σ factor
interspersed within the specific elements of the other σ, and that
they can be transcribed, both in vivo and in vitro, by the RNA polymerase
associated with σ32
or σ70
(Wade et al. 2006). Even though
these data confirmed previous observations on rrnBP1 and gapA
promoters (Newlands et al. 1993; Charpentier and Branlant 1994),
the functional overlap between σ32
and σ70
was found to be
much more extensive than expected. This promoter organization
allows basal σ70
-dependent expression of many genes under
normal conditions of growth to be boosted by σ32
upon temperature
challenge.
Considering the huge reprogramming of E. coli transcriptome
imposed by σ32
, a strategy that directly or indirectly targets this
pleiotropic regulator might be exploited by any invading organism
to manipulate host response. In fact, the first J-domain protein
(i.e. a protein that contain a J-domain able to stimulate
DnaK activity) encoded by a bacteriophage has recently been isolated
(Perrody et al. 2012). Notably, this protein, named Rki, is expressed
by the phage RB43 early after infection of E. coli and, by
specifically interacting and interfering with the DnaK chaperone
function, can stabilize σ32
, thereby promoting accumulation of
heat-shock proteins. The authors proposed that σ32
stabilization
by Rki could help phage growth, likely promoting middle- and
late-gene transcription and/or facilitating the proper folding of
phage proteins as a consequence of the GroESL chaperonin accumulation
(Perrody et al. 2012).
Escherichia coli heat-shock sigma-E (σE
)
Besides σ32
, the master regulator of heat-shock response, in
E. coli a second heat-shock regulon is governed by another
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
554 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
Figure 3. Schematic representation of the mechanism of regulation of σE
activity.
During normal growth conditions (A), σE
is sequestered to the inner membrane
by the concerted action of the antisigma factor RseA and the periplasmic
co-antisigma factor RseB. As a consequence, genes belonging to the σE
regulon
are transcriptionally inactive. Upon stress (B), RseA is completely degraded
by a cascade of regulated proteolytic events in which DegS and RseP membraneassociated
proteases are primarily involved. Following the complete degradation
of RseA by cytoplasmic proteases, σE
is available for interaction with the RNA
polymerase enzyme and the transcription of σEspecific
promoters is rapidly
activated.
alternative sigma factor named σE
(also known as σ24
). This
24-kDa alternative σ factor, encoded by the leading gene of the
tetracistronic rpoE-rseA-rseB-rseC operon, was initially isolated
as the factor responsible for the transcription of rpoH P3 promoter
(as reported in Fig. 1, left panel) with distinct promoter
specificity from the known sigma factors (Mecsas et al. 1993). It is
a member of the extracytoplasmic function subfamily of sigma
factors that function as effector molecules responding to extracytoplasmic
stimuli such as cell wall and membrane stress and
the oxidation state (Missiakas and Raina 1998; Helmann 2002).
Specifically, E. coli σE
is heat inducible and responsive to stress
in the extracytoplasmic compartment of the bacterial cell (like,
for example, accumulation of misfolded proteins translocated
across the cytoplasmic membrane into the periplasmic space)
and its activity is governed by a complex mechanism based on
regulated proteolysis (Mecsas et al. 1993). Under normal growth
conditions, σE
is tethered to the membrane and kept inactive by
the antisigma factor RseA (Fig. 3A). RseB and RseC, co-expressed
in the same operon together with σE
and RseA, also take part
in σE
regulation. Specifically, while the mechanism of positive
regulation exerted by the inner-membrane-associated RseC protein
on σE
activity still remains to be fully characterized (Missiakas
et al. 1997), it was demonstrated that RseB is located in
the periplasmic space and acts as a co-antisigma factor when
a direct interaction with RseA is established (Cezairliyan and
Sauer 2007). Upon stress, RseA is completely degraded by the
sequential action of different proteases and σE
is free to interact
with the RNA polymerase core enzyme to direct transcription
of a specific subset of promoters characterized by peculiar
–10 and –35 consensus elements (Rhodius et al. 2006; Ades
2008) (Fig. 3B). In detail, upon stress signal perception, the first
event of the regulatory cascade is the activation of DegS, a
membrane-associated protease that cleaves the periplasmic domain
of RseA only upon dissociation of RseB from the antisigma
factor (Walsh et al. 2003; Grigorova et al. 2004; Lima et al. 2013). A
second membrane-associated protease named RseP further digests
RseA in its transmembrane region, releasing in the cytoplasm
a truncated C-terminal portion of the antisigma factor
still associated to σE
(Kanehara, Ito and Akiyama 2002, 2003).
The last step of the σE
activation process involves other cellular
proteases such as ClpP/X-A, Lon and/or HslUV that completely
degrade the remaining undigested part of RseA, freeing
σE
in the cytoplasm (Chaba et al. 2007)(Fig. 3B). This membrane
localization of σE
, a strategy also employed to maintain
other alternative σ factors in their inactive form when their
activity is not necessary, differs from the membrane-localized
state of σ32
described above, a mechanism used for the dynamic
regulation of the principal E. coli regulator. The σE
regulon
consists of tens of genes, whose function is almost completely
devoted to folding of proteins in the cell envelope and
to biosynthesis/transport of lipopolysaccharide (LPS), a component
of the outer membrane of Gram-negative bacteria (Dartigalongue,
Missiakas and Raina 2001; Rhodius et al. 2006). In
fact, in addition to several genes involved in LPS biosynthesis,
some σE
-regulated genes encode periplasm-localized proteins
that act on misfolded polypeptides (such as, for example, the
periplasmic protease HtrA/DegP), while other regulon members
code for cytoplasmic proteins involved in coordinating environmental
stimuli with expression of the σE
regulon (Dartigalongue,
Missiakas and Raina 2001; Rhodius et al. 2006). However, as in
the case of σ32
described above, an extensive functional overlap
has been observed between σE
and the housekeeping σ70
. In
fact, a substantial fraction of σE
-transcribed promoters can also
be bound by σ70
, demonstrating that in E. coli the wide overlap
of alternative σ factors with σ70
is not limited to σ32
(Wade et al.
2006).
Negative transcriptional regulation of heat-shock genes
In addition to positive regulation by means of specialized σ factors,
another strategy to control heat-shock genes transcription
relies on dedicated repressors. Under normal laboratory
growth conditions, these DNA-binding regulators bind specific
operators and repress transcription of heat-shock genes, while
upon heat stress, the transcription of these genes is rapidly induced.
Promoters driving transcription of genes controlled by
heat-shock repressors harbor –10 and –35 elements recognized
and bound by RNA polymerase associated with the vegetative
σ factor. The first indications of the existence of alternative
mechanisms of transcriptional regulation that are different from
heat-shock σ factors came from studies of Bacillus subtilis stress
response. In this bacterium, it was initially thought that the σ28
containing
RNA polymerase controlled the transcription of heatshock
genes because of its overlapping promoter specificity with
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 555
E. coli σ32
-containing RNA polymerase (Briat, Gilman and Chamberlin
1985). Subsequent studies demonstrated that σ28
was dedicated
to transcription of flagellar genes and to some genes
coding for chemotaxis effectors and that different B. subtilis
chaperone genes, groESL and dnaK among others, were preceded
by vegetative promoter sequences (Li and Wang 1992; Wetzstein
et al. 1992). Afterwards, Chang et al. (1994) demonstrated that
transcription of groESL was dependent on the vegetative sigma
factor, σ43
, indicating that other mechanisms of heat-shock gene
regulation, different from positive strategies based on alternative
σ factors, must exist. In the subsequent years, several heatshock
repressor proteins were identified and their mechanisms
of regulation via interaction with conserved DNA elements located
in the promoters of heat-shock genes were characterized.
A common feature of heat-shock repressors is, besides the negative
regulation of target genes, that they repress their own promoter.
This regulatory scheme, known as negative autoregulation,
is widely found among transcriptional repressors (negative
autoregulation has been observed for about 50% of E. coli repressors)
and guarantees fast response kinetics upon signal perception
(Alon 2007). Consequently, upon heat stress and subsequent
rapid induction of the controlled promoters, repressor concentration
rapidly increases and reaches a threshold amount necessary
for the re-establishment of the repressed state. In this way,
the induction phase that rapidly appears upon heat stress is immediately
followed by a shut-off phase of transcription.
Currently, it appears well established that negative transcriptional
regulation through dedicated repressors constitutes an
important and widespread mechanism employed by bacteria to
control heat-shock gene expression, not necessarily as an alternative
to positive regulation described above.
HrcA repressor
The observation that a conserved sequence element is embodied
in the promoter regions controlling the transcription of class
I heat-shock genes of B. subtilis (consisting of two operons, the bicistronic
groES-groEL and the heptacistronic dnaK operon) paved
the way to the identification and characterization of the HrcA
repressor, one of the most widespread heat-shock negative regulators
in bacteria. Specifically, an inverted repeat (IR) consisting
of 9 bp separated by a 9-bp spacer was identified in the DNA region
close to the transcription start site of dnaK and groESL genes
(Schmidt et al. 1992; Wetzstein et al. 1992). The role of this IR as a
negative cis-element in the control of chaperone expression was
demonstrated by introducing mutational changes within the upstream,
downstream or both halves of the IR preceding dnaK
and observing increased expression at low temperature and a
reduction in the stimulation of the operon after heat-shock (Zuber
and Schumann 1994). Considering that the conserved IR
was observed in many different bacterial species, always within
promoters controlling chaperone genes, it was named CIRCE
for controlling inverted repeat of chaperone expression. Even
though the possibility was initially considered that the CIRCE sequence
could act alone through the formation of a temperaturedependent
secondary structure, the demonstration of involvement
of a DNA-binding protein in CIRCE-dependent negative
regulation followed soon after. In particular, the inactivation of
orf39, the first gene of the B. subtilis dnaK operon, led to the upregulation
of dnaK and groESL even at physiological growth temperature
(Schulz, Tzschaschel and Schumann 1995). Moreover,
the characterization of regulatory mutants constitutively overexpressing
groE and dnaK at low temperature revealed that all
mutations map within orf39. The finding that cell extracts of
E. coli overproducing Orf39 were able to specifically retard DNA
probes harboring a CIRCE sequence in electrophoretic mobility
shift assays (EMSA) clearly suggested that Orf39 was the negative
regulator of chaperones expression in B. subtilis, likely acting
as a CIRCE-specific DNA-binding repressor (Yuan and Wong
1995b). Similar observations were made for Caulobacter crescentus,
where disruption of an orf39 homologous gene led to the
conclusion that it encoded a negative regulator and, thus, it was
named HrcA for heat-shock regulation at CIRCE elements (Roberts
et al. 1996). Direct interaction between HrcA protein and the
CIRCE element was demonstrated in B. subtilis by showing that
a recombinant purified HrcA was able to retard a labeled probe
containing a CIRCE element in the EMSA assay (Mogk et al. 1997).
Detailed biochemical studies aimed at the molecular level dissection
of the HrcA-CIRCE interaction have been hampered in
many cases by its instability and strong tendency to form insoluble
aggregates. For this reason, HrcA binding to target promoters
has been characterized in detail only in a limited number
of cases. For example, in Helicobacter pylori the development of a
dedicated expression and purification protocol allowed demonstration
by DNase I footprinting that HrcA binds operator elements
similar to the CIRCE sequences overlapping both groESL
and dnaK core promoters (Roncarati et al. 2007a,b). On both target
promoters, the protected regions cover about 30 bp, suggesting
that a dimer of HrcA interacts with the DNA. A similar situation
was also observed for Chlamydia trachomatis, where HrcA
showed full protection of the CIRCE element and neighboring
nucleotides on the dnaK promoter, covering about 40 bp of the
DNA probe (Wilson and Tan 2004). In this latter microorganism,
in vitro transcription assays of selected promoters carried
out in the presence of different concentrations of purified HrcA
demonstrated that it was able to specifically repress dnaK transcription
in a concentration-dependent manner (Wilson and Tan
2002). The position of the CIRCE element is almost always in a
region close to the transcription start site, overlapping the portion
of the DNA contacted by the RNA polymerase during initiation
of transcription. The mechanism of repression exerted by
HrcA upon binding to the CIRCE element relies on the physical
occupancy of the core promoter with a consequent impediment
of RNA polymerase binding under physiological growth conditions
(Fig. 4A). Consistent with this model, when the position of
the CIRCE element of the B. subtilis groE promoter was shifted
from the original position to + 4 further downstream to position
+ 25 with respect to the transcription start site, the negative
regulatory effect mediated by this cis-acting element was
abolished (Yuan and Wong 1995a). HrcA-mediated repression of
heat-shock genes is generally responsive to different kinds of
environmental stresses. In many cases, when bacterial cells are
exposed to stressful conditions such as high temperature, high
salt concentration or a condition that provokes accumulation
of misfolded proteins in the cytoplasm, the transcription from
HrcA-controlled promoters is rapidly derepressed (Schmidt, Hertel
and Hammes 1999; Laport et al. 2004). Even though it is well
established that CIRCE acts as a specific HrcA-binding site at the
DNA level, it has also been observed that the IR affects the stability
of the RNA transcript. Specifically, in Rhodobacter capsulatus,
the secondary structure imposed by the CIRCE element on the
groESL mRNA appears to differentially affect transcript stability
at various temperatures (J¨ager, J¨ager and Klug 2004). To date, the
crystal structure of the HrcA repressor of the hyperthermophile
Thermotoga maritima provides the only detailed structural information
for this heat-shock regulator. The 2.2 ˚A resolution crystal
structure shows that TmHrcA forms a dimeric structure, and
each monomer is composed of a winged helix-turn-helix (HTH)
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
556 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
Figure 4. Negative regulation of heat-shock gene transcription by HrcA, HspR and CtsR repressors. (A) Under physiological growth conditions, HrcA binds the CIRCE
conserved IR and represses transcription by hindering RNA polymerase promoter binding (depicted by black arrowhead). GroES-GroEL chaperonin interacts with HrcA
and stimulates its DNA-binding activity. This GroE-mediated feedback control of HrcA activity is negatively modulated by accumulation of unfolded proteins during
stress. (B) HspR-mediated repression of heat-shock genes at normal growth temperature is due to binding of the regulator to conserved HAIR sequences close to the
core promoter, thereby interfering with RNA polymerase binding (indicated by a black arrowhead). In S. coelicolor, DnaK chaperone interacts with HspR and positively
modulates its DNA-binding activity. Upon direct interaction between HspR and DnaK, the repressor becomes competent for promoter binding, while DnaK acts as
a corepressor of HspR, with no direct contact to DNA. The HspR-DnaK functional interaction is negatively modulated by the accumulation of misfolded proteins in
the cytoplasm that titrate away the chaperone. (C) Current model for the mechanism of repression and for the heat-dependent regulation of CtsR repressor activity.
During control conditions, CtsR binds to conserved IR sequences and represses transcription of heat-shock genes (depicted by a black arrowhead). Following a sudden
increase of temperature, CtsR undergoes a heat-induced structural alteration and loses affinity for DNA (indicated by a gray dashed line). During environmental stress
conditions, upon temperature-dependent autophosphorylation, McsB becomes active and, assisted by its co-activator McsA, functions as an adaptor to target CtsR to
ClpP/C-mediated proteolytic degradation. Upon dephosphorylation by the cognate phosphatase YwlE, McsB becomes deactivated.
N-terminal DBD, a central GAF-like domain (probably involved in
dimerization rather than in cGMP binding) and a C-terminal inserted
dimerizing domain. However, the conformation of both
DBDs in the dimer suggests that the TmHrcA is incompatible
with DNA binding and may represent an inactive form of this
protein (Liu et al. 2005). The detailed structure of the DBD, combined
with results of mutagenesis studies of conserved residues
(Hitomi et al. 2003; Wiegert and Schumann 2003), suggests that
HrcA binding to DNA should involve the interaction of the recognition
helix of the HTH domain with the DNA major groove and
the interaction of a loop connecting two α-helices with the DNA
backbone. It is worth noting that the amino acid sequence similarity
of HrcA proteins from different species is surprisingly low.
The overall sequence identity is often <30% with the exception
of HrcA proteins from related organisms (Narberhaus 1999).
Hence, the molecular features derived from the solved structure
described above may be different among the various HrcA regulators.
In order to identify additional genes regulated by HrcA besides
the typical groE and dnaK genes, genome-wide expression
studies have been combined with the search of conserved CIRCE
elements and with in vitro DNA-binding studies. What generally
emerges from such approaches is that HrcA is involved,
directly or indirectly, in the regulation of several genes linked
to various cellular processes. For example, a microarray-based
transcriptome analysis performed in Listeria monocytogenes
revealed that HrcA regulates, albeit indirectly, genes involved
in stress response, metabolism, translation and DNA replication
(Yuewei et al. 2007). Similar pleiotropic effects of hrcA mutation
were observed with analogous approaches in other microorganisms,
including H. pylori and Lactobacillus plantarum (Roncarati
et al. 2007a; Van Bokhorst-van de Veen et al. 2013). However,
the number of genes directly regulated by HrcA appears to
be restricted and limited to genes involved in stress response.
For example, in the human pathogen Mycoplasma genitalium,
the genes encoding the stress proteases Lon and ClpB possess
a conserved CIRCE element within their promoter regions,
suggesting a HrcA-mediated negative regulation (Musatovova,
Dhandayuthapani and Baseman 2006). Moreover, various lines
of evidence indicate that the expression of clp genes in several
Lactobacillus species is regulated by HrcA, instead of CtsR, the
conserved clp gene regulator of low G+C Gram-positive bacteria
(Suokko et al. 2005, 2008), while in Streptococcus salivarus these
two heat-shock repressors are cooperative in controlling clpP
gene transcription (Chastanet and Msadek 2003).
HspR repressor
The heat-shock transcriptional repressor HspR was first discovered
and characterized in the Streptomyces genus. Specifically,
it was shown that, in Streptomyces coelicolor, the distal gene of
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 557
the heat-inducible dnaK operon encoded a novel heat-shock protein,
similar to the GlnR repressors of Bacillus spp. and to the
MerR family of transcriptional regulators. In vitro DNA-binding
assays were carried out to demonstrate that HspR was able to
bind to three inverted repeat sequences (IR1, IR2 and IR3) in
the dnaK promoter region, centred at positions –75, –49 and +4
with respect to the transcriptional start site. This strongly suggests
a direct role for HspR in heat-shock gene regulation (Bucca
et al. 1995). Genetic and biochemical evidence of HspR regulatory
function followed soon after. In particular, it was shown that disruption
of the hspR gene led to high-level constitutive transcription
of dnaK operon. Moreover, the addition of an E. coli purified
recombinant his-tagged HspR to in vitro transcription assays
clearly demonstrated repression of dnaK transcription, providing
additional evidence for direct involvement of HspR in dnaK regulation
(Bucca, Hindle and Smith 1997). Similar observations were
made in the same year in S. albus (Grandvalet, Servant and Mazodier
1997), where HspR was shown to bind an IR identical to
IR3 of S. coelicolor in the promoter region of the protease gene
clpB (Grandvalet, de Cr´ecy-Lagard and Mazodier 1999). Genes
similar to hspR were observed in several other bacterial species,
and HspR-binding sites were detected upstream of various heatshock
genes of these microorganisms. The HspR-binding site
was named HAIR (for HspR Associated Inverted Repeats) and a
consensus sequence was proposed (Grandvalet, de Cr´ecy-Lagard
and Mazodier 1999). So far, HspR repressors have been found and
characterized, to various extents, in several bacterial species
and it is now well established that this represents, in addition
to the widespread HrcA repressor, a widely employed system
to negatively control heat-shock genes transcription. The HAIR
sequences have been found, in most cases, in the vicinity of
the core promoter region, often overlapping the –10 and/or –35
boxes (Bucca et al. 1995; Grandvalet, de Cr´ecy-Lagard and Mazodier
1999; Stewart et al. 2002). This observation suggests that
HspR prevents RNA polymerase binding at physiological temperature
by a steric hindrance mechanism at the promoter (Fig. 4B).
In some cases, however, HspR-HAIR interaction takes place far
upstream of the core promoter region, in an atypical position
for a transcriptional repressor (Spohn et al. 2004; Schmid et al.
2005). The long distance between the HAIR sequence and the
core promoter –10/–35 boxes supports the hypothesis of a complex
mechanism for HspR-mediated transcriptional repression,
which encompasses more than simple steric occlusion of the
promoter. In this respect, dual repression of some heat-shock
genes mediated by HspR and HrcA in H. pylori represents a clear
example of this mechanism. The distal binding sites of HspR
on groESL and dnaK promoters (between positions –43/–120 and
–78/–149 with respect to the transcriptional start site, respectively)
combine with proximal HrcA/CIRCE interactions, occluding
core promoter regions (Roncarati et al. 2007a, discussed in
detail below). Another intriguing example of complex HspRmediated
regulation has been observed in Mycobacterium tuberculosis.
In this intracellular human pathogen, the acr2 gene, encoding
a member of the widespread α-crystallin family of molecular
chaperones, is transcriptionally controlled by HspR in combination
with the response regulator PhoP (Singh et al. 2014). In this
case, the heat-shock repressor HspR binds to the HAIR sequence
located in the core promoter region (Stewart et al. 2002), while
PhoP-binding region maps downstream from the transcriptional
start site. Moreover, a direct HspR-PhoP interaction has been
demonstrated, as well as the crucial importance of the simultaneous
presence of both regulators for heat-shock-dependent in
vivo regulation of acr2 (Singh et al. 2014). Accordingly, the current
model for acr2 regulation postulates that under normal growth
condition both the PhoP and HspR, bound to their target sites,
form a higher-order DNA-protein structure that prevents RNA
polymerase binding and the direct HspR-PhoP interaction provides
additional stability to this complex.
Initially identified as the regulator of major chaperone genes
such as dnaK in Streptomycetes spp., the identification of additional
genes controlled by HspR in several different bacterial
species was enabled by the advent of microarray techniques. In
S. coelicolor, DNA microarray-based analysis of gene expression
in wild-type and hspR-disruption mutant strains led to the identification
of 17 genes controlled, directly or indirectly, by HspR
(Bucca et al. 2003). This analysis was refined in a subsequent
work by the same group, by implementation of a high-density
microarray (Bucca et al. 2009). Specifically, gene expression analysis
was combined with immunoprecipitation of in vivo HspRbinding
sites and hybridization of the recovered DNA fragments
on a DNA microarray chip (ChIP-on-chip). From this analysis, it
emerged that, in S. coelicolor, HspR directly represses transcription
of genes encoding chaperones (dnaK-grpE-dnaJ-hspR operon)
and heat-shock proteases (lon and clpB), as well as one ribosomal
RNA gene and two tRNA genes, thus broadening, for the
first time, the regulatory role of HspR in this microorganism. In
H. pylori, to investigate if the regulatory function of HspR is restricted
to promoters of chaperone genes or if it is directly involved
in controlling additional genes, an in vitro selection of
genomic DNA fragments bound by purified HspR protein was
developed (Delany et al. 2002). Besides the three previously characterized
binding sites in the promoter regions of the multicistronic
operons coding for the components of the DnaK and
GroE machineries (Spohn and Scarlato 1999), two novel binding
sites were identified, located in the 3 region of both speA and tlpB
genes, which encode an arginine decarboxylase and a methylaccepting
chemotaxis protein, respectively. Moreover, sequence
alignment of novel HspR-binding sites highlighted HAIR-like sequences
as well as conserved nucleotides extending outside the
previously proposed consensus binding sequence. However, parallel
macroarray hybridization of cDNA probes deriving from
H. pylori wild-type and hspR-mutant strains failed to observe
deregulation of these novel target genes, suggesting the existence
of a minority of non-canonical binding sites, apparently
not associated with regulatory functions (Roncarati et al. 2007a).
In the last 15 years, the identification of members of the HspR
regulon was also pursued through array-based whole transcriptome
analyses in several other unrelated bacterial species including
Campylobacter jejuni, M. tubercolosis and Deinococcus radiodurans
(Stewart et al. 2002; Andersen et al. 2005; Schmid et al.
2005; Holmes, Penn and Lund 2010). Overall, these studies suggested
that HspR directly represses, alone or in combination
with other transcriptional regulators, the transcription of a limited
set of genes that encode the major cellular chaperones and
heat-shock proteases. However, indirect effects of hspR disruption
affecting transcript abundance of genes involved in diverse
cellular processes not strictly related to heat-shock have been
observed. For example, in H. pylori and in the closely related
C. jejuni, hspR deletion led to substantially lower expression of
many genes coding for proteins involved in cell motility, and accordingly,
hspR-mutant strain showed reduced motility (Andersen
et al. 2005; Roncarati et al. 2007a). One hypothesis for the
intersection of HspR-regulated heat-shock response and flagellar
assembly may be that accumulation of chaperone proteins
in the mutant strain alters the assembly of the flagellar apparatus
and/or increases the activity of dedicated flagellar transcriptional
regulators, which in turn establishes negative feedback
for the programmed transcription of motility genes. Another
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
558 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
interesting example of HspR-mediated interplay between the
heat-shock response and an unrelated central cellular process
was observed in Bifidobacterium breve UCC2003 (Zomer et al. 2009).
In this microorganism, an SOS response is induced by oxidative
stress as well as by heat-shock. Repression of DNA damage
repair genes is mediated by the LexA regulator, whose selfcleavage
and consequent activation is promoted by the RecA
protein. Induction of RecA expression is probably under the control
of HspR under heat-shock conditions, an observation that
directly links the repressor to a process employed by Bifidobacterium
to respond to an excess of DNA damage (Zomer and van
Sinderen 2010).
CtsR repressor
The characterization of B. subtilis heat-shock response revealed
the existence of a subgroup of stress-responsive genes whose
transcription is governed by two alternative promoters: one recognized
by the RNA polymerase complexed with the vegetative
sigma factor σA
and the other is dependent on the general stress
sigma factor σB
. The transcription from both the vegetative σA
and
σB
-dependent promoters of this class III subgroup of genes
(consisting of trxA, clpC and clpP operons) was induced by various
stress signals, even though the induction pattern differed
with respect to the particular gene and stress conditions (Kr ¨uger,
Msadek and Hecker 1996; Gerth et al. 1998; Scharf et al. 1998). The
observation that vegetative promoters of the above-mentioned
genes remained stress inducible even on a σB
-mutant background,
together with the lack of a CIRCE consensus sequence
typical of B. subtilis HrcA-regulated class I genes, prompted scientists
to search for a novel heat-shock repressor responsible
for this regulation. The first gene of the clpC operon, named orf1,
was shown to encode a product with a predicted HTH DBD, suggesting
a regulatory role for the protein (Kr ¨uger et al. 1997). This
hypothesis was confirmed by the observation that transcript
amounts of the class III clpC and clpP genes were significantly
upregulated in a orf1-mutant strain compared to the wild-type
under normal growth conditions, while class I (dnaK) and class
II (ctc) genes were similarly expressed between the two strains
(Kr ¨uger and Hecker 1998). The experimental confirmation of
Orf1 being a direct transcriptional repressor came from EMSA assays
carried out on a radiolabeled clpC operon promoter (Kr ¨uger
and Hecker 1998). Subsequent studies by Derr´e, Rapoport and
Msadek (1999) expanded the analysis of targets directly bound
by the repressor, thereafter renamed CtsR for Class three stress
gene repressor. Specifically, it was shown that CtsR binds a directly
repeated heptanucleotide sequence positioned in close
proximity of the transcription start site of clpP and clpC operons
(Derr´e, Rapoport and Msadek 1999). Moreover, CtsR binding
was shown to take place on five distinct binding sites nearby
the clpE transcription start point, also postulating the existence
of complex promoter architectures (Derr´e et al. 1999). Both CtsR
homologs and target sequences were found upstream of clp and
other heat-shock genes of several Gram-positive bacteria, an observation
that suggested CtsR as a widespread transcriptional
repressor involved in heat-shock regulation. In this respect, it
was subsequently shown that, in L. monocytogenes, CtsR regulation
is strikingly similar to that in B. subtilis, describing for the
first time a stress response regulatory gene in a human pathogen
(Nair et al. 2000). Specifically, it was demonstrated that L. monocytogenes
CtsR represses the transcription of clpP, clpC, clpE and
clpB genes by directly binding to their respective vegetative promoters
(Nair et al. 2000; Chastanet et al. 2004). Considering that
the positions of the binding sites characterized so far on several
promoters of various bacterial species overlap the core promoter
region, CtsR should exert its repressive function by sterically occluding
RNA polymerase binding to DNA (Fig. 4C). The first indications
that the prototypical B. subtilis CtsR could bind DNA as a
dimer, evinced from analysis of the binding sites and from some
biochemical studies (Derr´e et al. 2000), were further confirmed
by the determination of the crystal structure of the repressor
bound to a 26 bp DNA derivative of the clpC promoter (Fuhrmann
et al. 2009). The crystal structure revealed that each protomer
is characterized by an N-terminal winged HTH DBD and a
C-terminal dimerization domain composed by four α-helices organized
as a four-helix bundle. DNA recognition is based on the
insertion of the HTH domain in the major groove and a concomitant
interaction of the extended β-hairpin wing with the minor
groove. Some crucial residues, located within the HTH domain
and involved in DNA binding, were identified and found to be
highly conserved in the CtsR protein family (Fuhrmann et al.
2009). The CtsR repressor is historically considered the master
regulator of the cellular protein quality control genes of low-GC
Gram-positive bacteria. This definition is based on the observation
that CtsR typically controls the expression of genes encoding
stress proteases mainly involved in degradation of misfolded/denatured
polypeptides as well as of deleterious protein
aggregates. For example, in the model organism B. subtilis, but
also in the human pathogen L. monocytogenes and in the probiotic
bacterium Lactococcus lactis, CtsR essentially controls the
transcription of several components of the Clp machinery (Derr´e
et al. 1999; Nair et al. 2000; Varmanen, Ingmer and Vogensen 2000;
Chastanet et al. 2004), while the regulation of the classical molecular
chaperones, such as DnaK and GroE, is HrcA dependent. In
some cases, however, CtsR is also able to regulate the expression
of HrcA-dependent genes. In Staphylococcus aureus, for example,
dnaK and groE operons are coregulated by both CtsR and
HrcA. These regulators contact adjacent binding sites and act
synergistically to maintain low basal levels of expression of both
operons in the absence of stress (Chastanet, Fert and Msadek
2003). In some other instances, as in Oenococcus oeni, CtsR solely
regulates clp machinery genes and also GroE/DnaK molecular
chaperones. This lactic acid bacterium lacks hrcA in its genome
and CtsR has taken over the regulation of typical HrcA targets
(Grandvalet et al. 2005). Interestingly, some studies have revealed
new non-canonical genes regulated by CtsR. These include
genes encoding small heat-shock proteins, such as Hsp18
in O. oeni and Hsp1 in Lactobacillus plantarum, the membranebound
FtsH protease in L. plantarum and the tatAC genes that
code for two minimal translocases responsible for the twin arginine
translocation (Tat) pathway in L. monocytogenes (Grandvalet
et al. 2005; Hu et al. 2007; Fiocco et al. 2009, 2010). The CtsR
master regulator binds the promoters of target genes and represses
their transcription under physiological growth conditions.
Intriguingly, upon heat-shock, CtsR loses DNA-binding activity
and the non-functional protein undergoes regulated proteolytic
degradation mediated by the ClpCP protease (Elsholz et al.
2010) (Fig. 4C). The first indications that CtsR is a specific target
of ClpPC under stress conditions came from in vivo stability assays
performed in B. subtilis wild-type and in clpP or clpC mutant
strains (Kr ¨uger et al. 2001). Specifically, Kr ¨uger and colleagues
demonstrated that, after inhibiting translation with tetracycline,
CtsR was much more labile in the wild-type strain, compared to
mutant strains. It also became clear that the products of two
genes belonging to the CtsR operon, McsA and McsB, play a crucial
role in the proteolytic turnover of the repressor. In particular,
mcsB encodes a heat-activated kinase that targets nonfunctional
heat-inactivated CtsR, while mcsA codes for an activator of McsB
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 559
(Kirstein et al. 2005, 2007). The current model for CtsR proteolytic
degradation, represented in Fig. 4C, postulates that, under normal
growth conditions, McsB is kept inactive by interaction with
ClpC. Upon heat exposure, McsB is titrated away from ClpC and
becomes activated by temperature-dependent autophosphorylation
and by interaction with McsA activator. Following this sophisticated
dual activation step, McsB acts as an adaptor, targeting
non-functional CtsR species and efficiently driving them
to degradation. Active McsB adaptor is downregulated by either
dephosphorylation by the cognate phosphatase YwlE or by rapid
degradation of the phosphorylated form (Elsholz et al. 2010).
RheA repressor
RheA represents an example of heat-shock transcriptional repressor
with a restricted distribution among bacteria, being
found only in S. albus. The first indications that S. albus possessed
an additional heat-shock transcriptional regulator, in
addition to HrcA and HspR, came from the analysis of a DNA
region upstream of the hsp18 gene, encoding a heat-inducible
protein belonging to the small heat-shock protein family. This
open reading frame, named orfY and coding for a 225-amino
acid polypeptide, is transcribed from a vegetative promoter in
the opposite orientation to hsp18. Moreover, disruption of orfY
led to the accumulation of hsp18 transcript, even at low temperature,
suggesting a direct or indirect involvement of OrfY in the
transcriptional regulation of hsp18 (Servant and Mazodier 1996).
Additional biochemical and genetic studies provided evidence
that OrfY was the direct repressor of hsp18. Thereafter, OrfY was
named RheA for repressor of heat-shock protein eighteen. In particular,
expression of S. albus RheA in E. coli determined repression
of transcription from the hsp18 promoter fused to lacZ as
reporter gene. Moreover, a DNA probe encompassing the hsp18
promoter was specifically retarded by a crude cell extract containing
RheA in EMSA assays (Servant, Rapoport and Mazodier
1999). Considering that the transcription start sites of the divergent
hsp18 and rheA genes are separated only by 25 nucleotides
and that the –10 boxes of the two promoters partially overlap, it
was not surprising to discover that RheA also negatively autoregulates
its own transcription (Servant, Rapoport and Mazodier
1999). The RheA repressor binds two IRs (a perfect IR TGTCATCN5-GATGACA,
flanked by an imperfect IR GTCATC-N5-GACGAC)
located in the intergenic region of the divergent rheA and hsp18
core promoters, thereby occluding the access to RNA polymerase
and repressing both genes’ transcription under normal growth
conditions (Servant, Grandvalet and Mazodier 2000). RheA represents
the first described transcriptional repressor of small heatshock
proteins, anticipating the similar scenario confirmed in O.
oeni and Clostridium acetobutylicum, where hsp18 is transcriptionally
repressed by CtsR (Derr´e, Rapoport and Msadek 1999).
Feedback regulatory circuits modulate the activity of
heat-shock repressors
Similar to the complex homeostatic control of σ32
in E. coli described
above, feedback regulatory circuits of heat-shock transcriptional
repressor activity mediated by chaperones have been
observed in several bacterial species.
Soon after the identification and characterization of the HrcA
repressor as a negative regulator of class I heat-shock genes in B.
subtilis, experimental evidence pointed out the role of the GroE
chaperonin as a modulator of HrcA regulon (Mogk et al. 1997).
In particular, depletion of the intracellular level of GroESL was
associated with high expression of the HrcA-controlled dnaK
operon at all temperatures. By contrast, GroESL overexpression
led to the hyper-repression of transcription from the target promoter.
Moreover, EMSA assays showed that the affinity of HrcA
binding to a DNA probe containing a CIRCE operator is increased
in the presence of GroESL. It is worth noting that the shifted fragment
migrated to the same position in the absence or presence
of GroEL, indicating a role for the chaperonin in the modulation
of HrcA DNA-binding activity rather than functioning as a corepressor
(Mogk et al. 1997). Following the formal demonstration
of the direct interaction between HrcA and GroE in vitro (Reischl,
Wiegert and Schumann 2002), a model was proposed for their
functional interplay (Fig. 4A). This model, known as the ‘titration
model’, postulates that HrcA is kept in an active conformation
by GroESL, competent for binding to CIRCE elements. Upon
stress insult, the accumulation of misfolded polypeptides in the
cell sequesters and titrates away the chaperonin, which can no
longer interact with the HrcA repressor. Without the presence
of bound chaperonin, HrcA loses DNA-binding affinity and repression
of class I heat-shock genes is relieved. This model is
also supported by the finding that addition of ethanol, treatment
with puromycin (that causes accumulation of truncated
polypeptides in the cytoplasm) and overexpression of substrates
of GroESL derepress the HrcA regulon (Mogk et al. 1998). The
same feedback mechanism operates in C. trachomatis (Wilson
et al. 2005), where the interaction between HrcA and GroE was
investigated in more detail. It is worth noting that C. trachomatis
harbors in its genome three genes coding for GroEL homologs:
groEL1, groEL2 and groEL3. These genes are expressed constitutively
throughout the developmental cycle of the bacterium
(Karunakaran et al. 2003) and differentially regulated in response
to heat-shock. In fact, while transcription of groEL1 is negatively
regulated by the HrcA-CIRCE system and induced upon temperature
increase, transcription of groEL2 and groEL3 is not responsive
to heat-shock nor to HrcA regulation (Karunakaran et al.
2003; Hanson and Tan 2015). In this system, the characteristic
additional C-terminal tail of chlamydial HrcA interfered with repressor
binding to the CIRCE element in vitro and also with HrcAmediated
transcriptional repression in vitro and in vivo (Chen,
Wilson and Tan 2011). Intriguingly, the negative effect exerted
by the HrcA C-terminal inhibitory region could be counteracted
by GroE. Specifically, it was shown that recombinant GroEL was
able to enhance HrcA activity (HrcA-binding activity to CIRCE elements
in vitro as well as HrcA-mediated repression in vivo) and
that this effect was more pronounced on the full-length HrcA
rather than on the truncated version (without the C-terminal
tail) of the repressor. Intriguingly, the positive effect of recombinant
GroEL on HrcA-binding activity to its operator was shown to
be ATP independent, suggesting a non-canonical mechanism of
action of the chaperonin (Chen, Wilson and Tan 2011). The characterization
of the GroESL-mediated feedback control of HrcA
repressor activity in several distant bacterial species, like for example
in C. crescentus and H. pylori (Susin et al. 2004; Roncarati
et al. 2007a), suggests that it may represent a general mechanism
to post-transcriptionally control repressor functionality in
response to environmental stimuli.
The hypothesis that homeostatic control of transcriptional
regulators by chaperones might represent a novel common
theme in heat-shock gene regulation is confirmed by the characterization
of the feedback control of the heat-shock master
repressor HspR by DnaK. The DnaK-HspR functional interaction
was initially characterized in S. coelicolor. It was shown that the
in vitro formation of a stable complex between HspR and its DNA
target was DnaK dependent, suggesting that DnaK stimulates
HspR DNA-binding activity (Bucca et al. 2000). Surprisingly, the
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
560 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
DnaK-mediated modulation of HspR was shown to be independent
from co-chaperones DnaJ and GrpE and required no
addition of ATP, thereby excluding a mechanism based on DnaK
chaperoning function. In vitro EMSA assays showed the formation
of a ternary complex composed of DnaK interacting with
HspR bound to DNA. Thus, S. coelicolor DnaK acts as a corepressor
of HspR, with no direct contact with DNA (Bucca et al. 2000).
This role of DnaK in heat-shock gene regulation was further
confirmed both in vitro and in vivo. In particular, it was shown
that DnaK was able to stimulate HspR-mediated repression in
in vitro transcription assays (Bucca et al. 2000). Moreover, in vivo
depletion of cellular DnaK led to high-level expression of HspRrepressed
promoters at normal growth temperature (Bucca et al.
2003). All these findings converged in a model of DnaK feedback
control of HspR activity that resembles, with some differences,
the titration model valid for GroE-HrcA (Fig. 4B). That is, under
normal environmental conditions DnaK interacts with HspR and
the ternary complex binds to target DNA, resulting in tight transcriptional
repression. During heat-shock, DnaK is sequestered
by unfolded proteins that accumulate in the cytoplasm and
HspR is less efficiently able to repress the transcription of target
genes whose expression results induced. More recently, a similar
DnaK-mediated feedback loop controlling HspR activity has
been investigated and characterized in M. tubercolosis, although
with some important differences. In this bacterium, DnaK
interacts directly with the HspR-HAIR complex and stimulates
repressor DNA-binding activity in an ATP-independent manner
(Parijat and Batra 2015). However, in this pathogen HspR modulation
does not depend solely on DnaK as in S. coelicolor. Evidence
supports an important role played also by the GroE chaperonins
GroEL1 and GroEL2 in the HspR activation process (Das Gupta,
Bandyopadhyay and Das Gupta 2008; Bandyopadhyay et al. 2012).
In the context of feedback regulation of heat-shock repressors,
H. pylori represents a peculiar example in which two chaperones
modulate the activity of two distinct transcriptional
regulators. As described above, H. pylori uses both HrcA and
HspR repressors for the regulation of heat-shock response. However,
while HrcA activity is feedback regulated by the GroESL
chaperonin according to the typical ‘titration model’ proposed
for B. subtilis, the DNA-binding activity of HspR is negatively
modulated by the heat-shock protein CbpA, rather than by
DnaK (Fig. 5). In analogy with the homologous protein of E. coli
(Chae et al. 2004; Bird et al. 2006), the HspR-repressed cbpA gene
is thought to encode a dual-function protein, working both as
a DnaJ-like co-chaperone of DnaK and as a nucleoid-associated
protein involved in nucleoid structuring function. In order to
characterize a possible post-transcriptional control over HspR
by the DnaK chaperone system, surprisingly it was found that,
upon direct protein–protein interaction, CbpA alone was able to
negatively modulate HspR binding to target promoters in vitro
without contacting the DNA, but only when the repressor was
not bound to its operators. In addition, overexpression of CbpA
led to deregulation of heat-shock response in vivo (Roncarati,
Danielli and Scarlato 2011). These findings suggest important
considerations and add new perspectives in heat-shock gene
regulation. First of all, the effect of CbpA on HspR DNA binding
(negative effect) is opposite to the modulation typically exerted
by DnaK or GroE on HspR and HrcA (positive effect), implying
that the accepted model for chaperone feedback regulation of
transcriptional repressors is insufficient to explain CbpA-HspR
functional interaction in H. pylori. A possible explanation is that
CbpA regulation of HspR-binding activity is adopted by H. pylori
to fine-tune the regulatory network shutoff response governing
heat-shock response (Roncarati, Danielli and Scarlato 2011).
Second, considering the putative functions of CbpA as both a
co-chaperone and a nucleoid-associated protein (preliminary results
suggest that CbpA of H. pylori possesses both functions; D.
Roncarati, personal communication), it would be informative to
characterize the functional interplay between HspR and CbpA.
In other words, an intriguing hypothesis is that the heat-shock
regulator HspR itself, aside from its role in the repression of transcription
regulated by CbpA, might influence the co-chaperone
and/or nucleoid-associated activity of CbpA by direct protein–
protein interaction (Fig. 5). This would be a novel example in
which heat-shock gene regulation intersects with distinct cellular
functions, as for example the maintenance and regulation
of the bacterial nucleoid.
Cross-comparison of heat-shock transcriptional
repressors
Experimental data available so far allow to highlight similarities
as well as differences among the major heat-shock
transcriptional repressors described in the previous sections.
Starting from the mechanism of regulation, they all repress
Figure 5. Model for HrcA and HspR-dependent heat-shock gene regulation in H. pylori. Two dedicated transcriptional repressors, HrcA and HspR, directly repress three
multicistronic operons containing the major chaperone genes of H. pylori. Specifically, the cbp operon encodes a homolog of the E. coli curved DNA-binding protein A
(CbpA), the heat-shock transcriptional repressor HspR and a protein with putative helicase function; the dnaK operon encodes the heat-shock transcriptional repressor
HrcA, the co-chaperone GrpE and the DnaK chaperone; the gro operon encodes the chaperonin system GroES-GroEL. Positive regulation of HrcA activity by the GroESL
chaperonin in depicted by a gray dashed arrow, while negative modulation of HspR by CbpA is highlighted by a grey dashed hammerhead. The hypothetical negative
effect of HspR on CbpA activity is indicated by a grey dashed hammerhead marked with a question mark.
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 561
transcription, under normal growth conditions, by sterically occluding
RNA polymerase binding to target promoters. In almost
all cases, in fact, these repressors interact with the target
promoters by binding conserved sequences that overlap core
promoter elements, such as –10 and –35 boxes. A closer inspection
of their DNA-binding architecture reveals some differences.
While HrcA-regulated promoters typically contain a single
CIRCE IR and are bound by a dimeric form of the repressor,
other promoters are controlled in a more complex manner by
HspR and CtsR repressors. For instance, B. subtilis clpE promoter,
but also S. coelicolor and H. pylori dnaK promoters harbor multiple
repressors’ binding sites spread all over the 5 control region,
suggesting the existence of complex mechanisms of transcriptional
regulation. Another common feature shared by all heatshock
repressors is their negative autoregulation. Negative autoregulation
is a recurrent regulatory pattern, also called network
motif, found in ∼40% of the known transcriptional regulators
in E. coli (Thieffry et al. 1998). This network motif has been
suggested to increase the homeostasis of the autoregulated gene
product in the context of stochastic gene expression noise, to
speed up the response kinetics upon signal perception and to
provide more linear dose responses (Becskei and Serrano 2000;
Rosenfeld, Elowitz and Alon 2002; Nevozhay et al. 2009). In the
context of heat-shock gene regulation, negative autoregulation
ensures tight control of steady-state repressors amount under
normal environmental conditions, followed by a prompt
induction upon heat-shock of regulators’ gene transcription.
Many more peculiarities emerge when considering the posttranscriptional
regulation of repressors activity and stability/turnover
mediated by chaperones and proteases, respectively.
As shown in Fig. 4, while CtsR and RheA activity does
not appear to be feedback modulated by molecular chaperones,
both HrcA and HspR interplay with distinct and specific chaperone
systems, whose actions add an additional level of regulation.
On the other hand, the repressors’ turnover that takes
place following heat stress has been deeply characterized for
CtsR (shown in Fig. 4C), while detailed information for the other
heat-shock repressors is still missing. In this respect, Jastrab
et al. (2015) have recently identified a novel pathway responsible
for the degradation of the heat-shock repressor HspR in M.
tubercolosis. Specifically, they showed that the heat-shock repressor
is actively degraded by the prokaryotic proteasome encoded
by this human pathogen, and that degradation is enhanced by
an ATP-independent proteasome activator, called PafE, and by
HspR denaturation.
Combinations of heat-shock transcriptional regulators
The transcriptional regulation of heat-shock genes displays
varying degrees of complexity among various microorganisms,
reflecting the extreme diversity of genetic regulatory mechanisms
in bacteria.
In several bacteria, heat-shock transcriptional regulation is
managed exclusively by a dedicated σ factor that enables the
RNA polymerase enzyme to transcribe specific genes important
for the response to temperature stress (Fig. 6A, left panel, and
Table 1). In such simple cases, all the thermoresponsive genes
are under the control of a single regulator. For example, in Pseudomonas
aeruginosa and in Vibrio cholerae, transcriptional regulation
of heat-shock genes is governed by an alternative sigma
factor specific for heat stress, similar in sequence and function
to E. coli σ32
(Allan et al. 1988; Slamti, Livny and Waldor 2007).
As described above, the model organism E. coli also adopts exclusively
positive regulators for the heat-shock response. Here,
two alternative σ factors (σ32
and σE
) are involved in heat-shock
regulation, with σ32
playing a major role in sensing cytoplasmic
and inner-membrane stimuli, while σE
is committed to extracytoplasmic
(envelope) stress response (Fig. 6A, left panel, and
Table 1).
In other bacterial species, the coexistence of positive and
negative control mechanisms, together regulating the expression
of distinct sets of heat-shock genes, has been observed. For
example, in the Gram-positive pathogen L. monocytogenes, there
are three groups of heat-shock genes, each one controlled by
a different strategy (Fig. 6A, central panel). In detail, genes belonging
to class I (encoding members of the major chaperone
systems DnaK and GroE) and to class III (encoding chaperones
and ATP-dependent Clp proteases) are negatively regulated by
HrcA and by the CtsR heat-shock repressor, respectively. In contrast,
genes belonging to class II (coding for general stress proteins)
are positively controlled by the alternative sigma factor σB
(Table 1) (van der Veen et al. 2007). A similar situation, in which
positive and negative mechanisms are both employed to control
separate sets of genes, has been described in other bacteria
such as B. subtilis and Agrobacterium tumefaciens (Table 1) (Nakahigashi
et al. 1999; Schumann 2003). Intriguingly, in cyanobacteria,
both positive and negative strategies of transcriptional regulation
combine to control expression of some heat-shock genes
(reviewed by Rajaram, Chaurasia and Apte 2014). For instance,
in Synechocystis PCC6803, the groESL operon is negatively regulated
by the HrcA/CIRCE system and positively regulated by the
alternative sigma factors SigE and SigB (Nakamoto, Suzuki and
Kojima 2003; Singh et al. 2006; Kojima and Nakamoto 2007). Furthermore,
the histidine kinase Hik34 is involved in the negative
regulation of the chaperonin gene (Slabas et al. 2006; ˇCerveny
et al. 2015). In addition to transcriptional mechanisms, RNAbased
regulation of heat-shock genes has also been reported in
cyanobacteria. Expression of the hsp17 gene, coding for a small
heat-shock protein also known as HspA, is transcriptionally dependent
on the alternative sigma factors SigE and SigB (Singh
et al. 2006; Tuominen et al. 2006, 2008), and post-transcriptionally
controlled by a temperature-sensitive RNA structure in the
5 -UTR portion of the gene (Table 1) (Kortmann et al. 2011).
The regulatory schemes found in bacteria that adopt only
negative mechanisms of transcriptional regulation could be
very simple. However, combinations of repressors and overlaps
of their regulons are not rare and sometimes lead to
complex regulatory networks. One of the simplest regulatory
scheme is employed, for example, by the human pathogen
M. genitalium (Fig. 6A, right panel). This Gram-positive bacterium,
which has the smallest genome among self-replicating
free living organisms, possesses a single housekeeping σ factor
and manages heat-shock gene regulation exclusively through
the HrcA repressor (Table 1) (Musatovova, Dhandayuthapani
and Baseman 2006). Similarly, a single transcriptional repressor,
CtsR, controls heat-shock response in the lactic acid bacterium
O. oeni (Grandvalet et al. 2005) (Fig. 6A, right panel,
and Table 1). A common manner of combining negative regulators
is based on the exploitation of several transcriptional
repressors, each specifically controlling a particular set of
genes. In B. subtilis and closely related species, the HrcA regulator
is in charge of keeping groESL and dnaK genes repressed
under normal growth conditions, while Clp proteaseencoding
genes are regulated solely by the master repressor CtsR
(Table 1) (Chastanet et al. 2001). An extreme example is represented
by S. albus, a soil bacterium that uses an arsenal of three
different repressors to specifically regulate distinct sets of heatshock
genes without any cross-regulation and regulons overlap
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
562 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
Figure 6. Combinations of heat-shock transcriptional regulators and complex regulatory circuits. (A) While in some bacterial species heat-shock transcription is
controlled exclusively by positive (left panel) or negative (right panel) mechanisms of regulation, in several cases positive and negative control strategies coexist (central
panel). Black arrows indicate positive regulation, while black hammerheads show negative regulation. (B) Examples in which distinct transcriptional repressors are
employed to control heat-shock gene expression. In S. albus HrcA, HspR and RheA repressors regulate separate sets of genes, whereas in S. pneumoniae HrcA and
CtsR regulons partially overlap. In S. aureus, the HrcA regulon is completely embedded within the CtsR regulon. (C) Schematic representation of the incoherent feedforward
loop governing heat-shock gene regulation in S. aureus and H. pylori. A master regulator, shown in red (HspR for H. pylori or CtsR for S. aureus), represses several
heat-shock genes. Among these genes, the groESL operon is also repressed by the HrcA regulator, which is, in turn, under the control of the master repressor.
(Fig. 6A, right panel, and Fig. 6B). Specifically, while groEL regulation
involves HrcA (Grandvalet, Rapoport and Mazodier 1998),
dnaK and clpB genes are repressed by HspR (Bucca, Hindle and
Smith 1997; Grandvalet, Servant and Mazodier 1997) and the
hsp18 gene is transcriptionally controlled by RheA (Table 1) (Servant,
Rapoport and Mazodier 1999). However, in several other
bacterial species, the regulons of two different heat-shock repressors
partially overlap and this results in some genes being
simultaneously controlled by more than one regulatory protein,
rendering these systems more complicated than the examples
described above. HrcA is most commonly involved in those situations,
partnering with CtsR in some bacteria, while in some
other cases it interacts directly with HspR. A typical example
is the dual regulation of groESL by CtsR and HrcA in S. pneumoniae
(Fig. 6B). In this microorganism, the groESL 5 region harbors
a highly conserved CIRCE operator, which is the specific recognition
sequence for HrcA-binding mapping immediately downstream
from the transcription start site. Also, there is an upstream
CtsR-binding site overlapping the –10 and –35 hexamers
(Chastanet et al. 2001). The two binding sites are in close proximity,
separated by only 16 bp. This tandem operator arrangement,
involving a crucial region for transcription initiation, results in
the S. pneumoniae groESL operon being under dual repression by
both HrcA and CtsR (Fig. 6B). Accordingly, it was demonstrated
that, in a mutant strain lacking CtsR, transcription of groESL was
just slightly increased at 37◦
C, consistent with functional HrcA
repression (Chastanet et al. 2001). This non-redundant dual repression
mechanism, also employed very similarly by S. salivarus
for the regulation of groESL and clpP (Table 1) (Chastanet and
Msadek 2003), allows transcription of these genes to be maintained
at very low levels in the absence of stress by the synergistic
action of both CtsR and HrcA and to be fully induced by
both regulators upon signal perception. Also, in S. aureus, CtsR
and HrcA combine to control the expression of groESL and dnaK,
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 563
Table 1. Regulatory mechanisms controlling heat-shock genes’ transcription in a panel of bacteria.
Organism Transcriptional regulator Regulated genes
Agrobacterium tumefaciens RpoH groESL, dnaK-dnaJ, grpE, clpP and others
HrcA groESL
Bacillus subtilis HrcA Class I: groESL, dnaK-dnaJ, grpE, and others
σB
Class II: genes coding for general stress proteins
CtsR Class III: clpP, clpE, ctsR-mcsA-mcsB-clpC and others
Unknown
CssS/CssR Class IV: htpG
Unknown Class V: htrA, htrB and others
Class VI: ftsH, clpX and others
Escherichia coli σ32
(σH
, RpoH) groESL, dnaK-dnaJ, grpE, ibpA and others
σE
(σ24
) degP, clpX, lon and others
Helicobacter pylori HspR cbpA-hspR-helicase, groESL, hrcA-grpE-dnaK
HrcA groESL, hrcA-grpE-dnaK
Listeria monocytogenes HrcA Class I: groESL, dnaK, and others
σB
Class II: genes coding for general stress proteins
CtsR Class III: clpP, clpB, hslU and others
Mycoplasma genitalium HrcA dnaK, lon, clpB
Oenococcus oeni HrcA groESL, dnaK, clpP, hsp18 and others
Pseudomonas aeruginosa σH
(RpoH) groESL, dnaK, and others
Staphylococcus aureus HrcA hrcA-dnaK, groESL
CtsR clpP, clpC, clpB, hrcA-dnaK, groESL
σB
genes coding for general stress proteins
Streptococcus pneumoniae HrcA groESL, dnaK-dnaJ, grpE
CtsR groESL, clpP, clpC, clpE
Streptococcus salivarus HrcA clpP, groESL, dnaK
CtsR clpP, groESL, other clp genes
Streptomyces albus HrcA groESL1, groEL2, dnaJ2
HspR dnaK, clpB
RheA hsp18
Synechocystis PCC6803 HrcA groESL, groEL-2, dnaK2 and others
SigB groESL, groEL-2, dnaK2, hspA, htpG and others
SigE groESL, groEL-2, dnaK2, hspA, htpG and others
HiK34 groESL, dnaK2, htpG
Vibrio cholerae σ32
(RpoH) groESL, dnaK-dnaJ, lon, clpB, and others
but they interact in an even more peculiar and complex way
(Fig. 6B). In particular, S. aureus CtsR alone represses the transcription
of several clp genes, but also combines with HrcA to
control expression mediated by the dnaK and the groESL operons.
Moreover, since hrcA is the first gene of the dnaK operon,
this results in the HrcA regulon being entirely embedded within
the CtsR regulon (Table 1) (Chastanet, Fert and Msadek 2003). As
in S. pneumoniae, CtsR- and HrcA-specific operators are organized
in tandem and overlap the region adjacent to the transcription
start site, allowing non-redundant dual repression during normal
growth. Strikingly, an almost identical regulatory network
has been described in the distantly related Gram-negative human
pathogen H. pylori, where HrcA combines with the HspR
repressor (Fig. 5). Here, HspR alone represses transcription of
its own tricistronic operon (Spohn and Scarlato 1999), while, in
combination with HrcA, it controls the expression of groESL and
hrcA-grpE-dnaK operons (Spohn et al. 2004). Hence, similarly as
for S. aureus, this results in the H. pylori HrcA regulon being embedded
within the regulon of a second master heat-shock repressor
(i.e. HspR instead of CtsR). A detailed analysis of in vitro
DNA binding by HrcA and HspR revealed that their operators
are arranged in tandem, consistent with the typical architecture
for dually regulated promoters, but that the position of the
HspR operator maps to far upstream from the core promoter
to an atypical position for a repressor (Roncarati et al. 2007a).
Moreover, even though both repressors are required for regulation
(a single hspR or hrcA mutation leads to complete derepression
of the dually controlled promoters), binding in vitro to their
respective adjacent operators occurs in an independent, noncooperative
manner. An interesting explanation for the peculiar
interaction of heat-shock repressors found in H. pylori and S. aureus
can be inferred by considering a logical scheme involving
both regulators (Fig. 6C). In both organisms, there is a master
regulator (HspR or CtsR) which directly regulates another regulator
(HrcA) and a target gene (groESL), which, in turn, is regulated
by both HrcA and the master regulator (HspR or CtsR).
Moreover, considering that all three regulatory interactions are
repressive, these circuits seem to represent rare examples of incoherent
type-2 feed forward loops (Alon 2007; Danielli, Amore
and Scarlato 2010). This peculiar network motif is employed to
modulate the dynamic behavior of the circuit, greatly speeding
up the transcriptional response of target genes upon input stress
signals. Some experimental observations of H. pylori sustain this
hypothesis. In fact, transcription of groESL was rapidly induced (2
min) upon heat-shock in a wild-type genetic background, while
in a mutant strain in which the binding site of HspR on the
groESL promoter was deleted (and, so, the direct connection between
HspR and groESL was interrupted), the derepression of
groESL transcription was observed only 60 min after temperature
challenge (Spohn et al. 2004). Even though further experimental
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
564 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
characterizations are needed to dissect the dynamic properties
of regulatory circuits such as those described above, they could
reflect specific evolutionary adaptations to particular needs encountered
by bacteria in their specific niches.
The examples described above provide a complex picture
of heat-shock genes’ regulation and raise interesting questions
about their role and evolution. Even though, in many cases, clear
experimental evidences are still missing, some speculations can
be put forward. For instance, the presence in the same organism
of diverse regulatory mechanisms involved in the control of distinct
as well as overlapping sets of genes might have evolved
to allow bacteria a proper response to distinct stresses. Alternatively,
such regulatory systems might account for a ‘differential
response’ in terms of both stress severity and differential
pattern of gene expression in response to a stress insult.
The latter aspect has been observed in S. albus, where different
sets of heat-shock genes are controlled by different repressors
(Fig. 6A). Specifically, two different regulatory patterns have
been observed, consisting of constitutive synthesis of GroEL and
Hsp18 (regulated by HrcA and RheA, respectively) at high temperature
and transient heat-shock-induced synthesis of HspRregulated
ClpB and DnaK proteins (Mazodier et al. 1991). A heatshock
intensity-dependent response (differential response) of
some regulatory systems or individual genes has been documented
in different bacterial species, including M. tuberculosis
and Corynebacterium glutamicum (Young and Garbe 1991;
Engels et al. 2004). The use of diverse regulatory systems by the
same organism to control different heat-shock genes becomes
particularly relevant and interesting when it involves the differential
regulation of multiple copies of genes coding for the
GroEL chaperonins. It has been observed that a significant proportion
of bacterial genomes contains two or more chaperonin
genes and, in several instances, these genes appear to be differentially
regulated (Lund 2009). The current notion is that,
where present, these multiple chaperonins encoded by duplicated
groE genes have evolved a degree of subfunctionalization
and the differential regulation observed may reflect the various
contexts in which they are needed or the ecological niche they
have to cope with. For example, in Bradyrhizobium japonicum, a
nitrogen-fixing and roots-nodulating bacterium belonging to the
Alphaproteobacteria group, the seven chaperonin genes present
in its genome show complex pattern of transcriptional regulation,
not limited to heat-shock response. At least one of these
chaperonin genes has been shown to be transcriptionally regulated
by the NifA activator and by σ54
, which are both involved
in the regulation of nitrogen fixation genes under limited oxygen
conditions (Fischer 1994). Genetic analyses provide additional
evidences for a link between the regulation of some chaperonins,
root nodulation and nitrogen fixation, even though the
specificity of chaperonin novel functions appeared not absolute
(Lund 2009). A similar scenario of differential regulation of multiple
chaperonin genes linked to the evolution of chaperonins’
specialized functions has also been proposed for other bacterial
groups, including Cyanobacteria and Chlamydia (Lund 2009).
MECHANISMS OF HEAT SENSING
The ability of bacteria to rapidly respond to sudden temperature
increase depends on heat-sensing mechanisms that integrate
environmental cues to activate appropriate response pathways.
To date, various mechanisms of thermoregulation have
been described in bacteria and they involve nearly all classes
of biomolecules including lipids, proteins and nucleic acids
(i.e. both DNA and RNA). All of these classes can act as thermosensors
that detect changes in the environmental temperature
and initiate relevant cellular responses. Heat-sensing
mechanisms can be direct, by which the temperature directly affects
the activity of the sensing biomolecule, or can be indirect,
by which the consequences of a sudden temperature increase
(for example, the accumulation of misfolded proteins in the cytoplasm)
are detected. Even though temperature is a ubiquitous
signal that influences several cellular pathways, this chapter focuses
mainly on sensing mechanisms that trigger heat-shock
gene expression, and just a few examples of thermosensors involved
in the regulation of virulence determinants will be described
(extensively reviewed by Klinkert and Narberhaus 2009;
Shapiro and Cowen 2012).
Temperature sensing through RNA
The most rapid way of changing gene expression in response to
temperature variations is based on a cis-regulatory element that
is part of the mRNA encoding the heat-shock protein to be regulated.
This mechanism of thermoregulation guarantees a very
fast response upon signal perception because temperature affects
the translation efficiency of both the intracellular pool of
mRNA molecules as well as transcription already in progress.
The general principle is based on the formation of zipper-like,
temperature-sensitive secondary structures that characterize
the mRNAs subjected to this kind of regulation (Fig. 7A). In particular,
at physiological temperature the 5 region of this category
of mRNA forms a structure that hinders sequence elements
crucial for the initiation of translation, such as the ribosomebinding
site (also known as the Shine-Dalgarno sequence) and
the translation start codon. When such sequence elements are
involved in a secondary structure, the recognition and binding
of the transcript by the ribosome is hampered, thereby negatively
affecting translation of the mRNA. Upon a temperature
increase, the secondary structure goes through a rearrangement
or partial melting. As a consequence, ribosomes can easily gain
access to the 5 mRNA region and translation is enhanced. Several
temperature-sensing RNA sequences, also known as RNA
thermometers, have been discovered and characterized in some
details in the last two decades and were recently reviewed in
a comprehensive and detailed article by Kortmann and Narberhaus
(2012). The first RNA thermometer was discovered in E. coli
and regulates the expression of the heat-shock alternative sigma
factor σ32
, already discussed above. It represents one of the most
complex RNA thermometers known so far, with an extended
secondary structure that is not confined only to the RpoH mRNA
5 region (Fig. 1A). Specifically, although in the characterized
secondary structure the Shine-Dalgarno sequence is partially
exposed, translation of RpoH mRNA is hindered at low temperature
by an intramolecular pairing between the 5 untranslated
region and a portion of the coding sequence immediately
downstream the AUG start codon (Kortmann and Narberhaus
2012). However, much simpler RNA thermometers exist, as exemplified
by the one controlling the expression of the Bradyrhizobium
japonicum hspA gene. This cis-regulatory sequence constitutes
the founding member of the most abundant class of
RNA thermometers called ROSE for repression of heat-shock
genes expression. ROSE elements are typically involved in the
regulation of small heat-shock proteins and they have a length
ranging from 60 to about 120 nucleotides. ROSE elements were
noticed as a conserved DNA element preceding several heatshock
genes in various rhizobia. Initially, it was speculated that
they act at the DNA level and their regulatory mechanism was
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 565
Figure 7. Mechanisms of heat sensing. (A) Sensing through RNA involves temperature-sensitive secondary structures in the 5 region of the mRNA subjected to this
kind of regulation. During normal growth conditions, a secondary structure forms and sequence elements crucial for efficient initiation of translation are masked and
poorly accessible for ribosome binding. Upon temperature increase, a structural rearrangement or resolution of the structure exposes such sequence elements and
translation is enhanced. Symbols: SD, Shine-Dalgarno sequence element; AUG, translation start codon. (B) Temperature sensing through DNA: model of temperaturedependent
regulation of virF transcription in S. flexneri. The DNA region flanked by the H-NS binding sites assumes a curvature at low temperature that allows contacts
between H-NS dimers bound to separate binding sites and the formation of a repressive nucleoprotein complex. At higher temperature, this curvature weakens and the
promoter is more accessible to RNA polymerase for virF gene transcription. (C) Transcriptional regulators as intrinsic heat sensors. Intrinsic heat-sensing repressors
are competent for DNA binding at permissive temperature (green oval), and transcription of heat-shock genes is repressed. Upon heat challenge, a temperatureinduced
structural transition (shown by a green rectangle) lead to a decrease of repressor DNA-binding activity and transcription is derepressed. (D) Indirect heatsensing
mechanism mediated by chaperones. The DNA-binding activity of transcriptional regulators is modulated by chaperones. During normal growth conditions,
the chaperone interacts with the heat-shock regulator and exerts its regulatory function. Several positively modulated repressors (such as HrcA and HspR in some
bacterial species) gain DNA-binding activity upon chaperone interaction, while in other instances the interaction with the chaperone results in the sequestration of
the activator (for example, E. coli σ32
). Following heat stress, chaperones are sequestered by misfolded proteins that accumulate in the cytoplasm and transcriptional
regulators are released and their DNA-binding activity results are positively or negatively affected.
thought to be dependent on the binding of a repressor protein
(Narberhaus et al. 1998). Noting that it was possible to predict a
similar secondary structure for all 15 known ROSE elements, a
post-transcriptional mechanism was proposed and verified by
a detailed mutational analysis of a ROSE-hspA-lacZ translational
fusion (Nocker et al. 2001). Subsequent extensive molecular and
structural characterizations revealed some key features of such
kinds of RNA thermometers (Chowdhury et al. 2003, 2006). Interestingly,
it was shown that ROSE-based RNA thermometers
comprise two to four stem loops and that they can gradually respond
to temperature variation and provide differential levels
of expression regulation according to the severity of heat stress.
Moreover, the dissection of the crucial intramolecular interactions
involved in secondary structure formation revealed that,
even though ROSE elements are characterized by a conserved
short stretch of nucleotides (U(U/C)GCU), non-canonical base
pairs also seem to be crucial for fine differential temperature
sensing. Additional examples of RNA thermometers containing
ROSE or ROSE-like elements have also been found in other
bacterial species such as Caulobacter, Salmonella and Escherichia
coli. In the latter species, a ROSE-like RNA thermometer controls
the temperature-dependent expression of the small heatshock
protein IbpA (coding for an inclusion body binding protein),
whose transcription is σ32
dependent (Nocker et al. 2001;
Waldminghaus et al. 2009). Another class of RNA thermometers
is based on a short stretch of four conserved uridines that base
pair with the AGGA nucleotide sequence constituting the ShineDalgarno
sequence. This element, called fourU, was initially
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
566 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
characterized in Salmonella enterica, where it controls the
temperature-dependent expression of the agsA gene that encodes
for a small heat-shock protein (Waldminghaus et al.
2007). Contrary to ROSE elements, the formation of the doublestranded
RNA secondary structure in fourU elements is governed
solely by canonical base pairing (Kortmann and Narberhaus
2012). Moreover, a peculiar feature of the fourU elements,
derived during a detailed structural investigation of the agsA
thermometer, is that the melting process in response to temperature
fluctuation is dependent on the concentration of Mg2+
ions, due to the presence of an Mg2+
-binding site placed directly
on the fourU element (Rinnenthal et al. 2011). It is worth
noting that thermoregulation based on RNA-sensing structures
has not evolved exclusively with respect to heat-shock genes.
Several pathogenic bacteria, which are subjected to shifts in
temperature during their life cycle (from environmental temperature
of 28◦
C–30◦
C to body temperature of 37◦
C or even
higher during infection), can modulate the expression of some
virulence-associated genes in response to temperature variations
using RNA thermometers (Grosso-Becera, Serv´ın-Gonz´alez
and Sober´on-Ch´avez 2015). For example, in the obligate commensal
of the human nasopharynx Neisseria meningitidis, RNA
thermometers were found in the 5 region of three genes involved
in the meningococcal resistance against immune killing.
Specifically, cssR, fHbp and lst genes (the first coding for an enzyme
involved in the exopolysaccharide biosynthesis, the second
coding for the host complement regulator factor H-binding
protein and the third encoding a protein necessary for LPS
sialylation) harbor RNA thermometers adjacent to the ShineDalgarno
sequence that enhance their expression at 42◦
C, a temperature
that is reached in the nasopharynx during Neisseria and
viral co-infections (Loh et al. 2013; Barnwal et al. 2016).
Temperature sensing through DNA
In some cases, temperature variations can be directly sensed by
the DNA of the bacterial cell. Several physiological metabolic
pathways of a microorganism are influenced by the external
conditions experienced in its ecological niche. In turn, the
metabolic state of the cell is influenced, including the ADP to
ATP ratio. As a consequence, enzymes that require ATP as a cofactor,
such as DNA gyrase, will be affected. This enzyme, whose
activity is strictly linked to the topological state of the DNA,
is dependent on ATP and inhibited by ADP (so change in the
ATP:ADP ratio influences DNA gyrase activity). For this reason,
external fluctuating conditions that affect metabolic processes,
including osmotic and heat-shock, can ultimately influence the
global level of DNA supercoiling (Hsieh, Burger and Drlica 1991;
Dorman and Corcoran 2009). Considering that DNA supercoiling
can influence gene transcription, DNA can be considered
as a thermosensor of environmental temperature change acting
through variations of the global topological state of the chromosome
in response to external stimuli. One of the primary parameters
of DNA topology that responds to temperature changes
is plasmid supercoiling. In particular, it was demonstrated both
in mesophilic and hyperthermophilic bacteria that sudden temperature
variations lead to transient changes in plasmid DNA
topology and this effect, in turn, has a significant impact on transcription
efficiency (L´opez-Garc´ıa and Forterre 1997). In some
other cases, however, local DNA structures mediate temperature
sensing and affect transcription of neighboring genes. Some
DNA sequences identified in E. coli and in several other bacteria
are characterized by a sequence specific topological conformation
that is able to assume various conformations in response to
temperature variation. When these DNA regions are proximal to
promoters, the local conformational variation induced by heatshock
is transduced into a modulation of RNA polymerase binding
efficiency, thereby affecting the transcription of the downstream
genes (Nickerson and Achberger 1995). An example is
represented by the region upstream of the plc gene, encoding
the phospholipase C (PLC) in Clostridium perfrigens (Katayama
et al. 1999). In this case, the three phased A-tracts that precede
the PLC encoding gene confer a strongly bent conformation to
this DNA sequence. By using in vitro transcription assays, it was
demonstrated that the stimulatory effect on the promoter activity
of the A-tract sequence was temperature dependent, probably
due to changes in the bending angle upon temperature fluctuations.
Besides the direct activation of transcription mediated
by bent DNA through the facilitation of RNA polymerase binding,
temperature-dependent local DNA curvatures can indirectly
regulate the efficiency of transcription by affecting the interaction
of proteins with a regulatory role on gene expression. In this
respect, nucleoid-associated proteins such as IHF, Fis, HU and
H-NS have been shown to be involved in this kind of process.
One of the best studied examples concerns the temperaturedependent
regulation of virF transcription in Shigella flexneri, a
pathogenic bacterium able to invade human intestinal epithelium
(Fig. 7B). The AraC-like VirF regulator, a protein that triggers
the activation of several genes with invasive functions, must
be expressed only after the shift from the outside environment
to the host. It has been demonstrated that the transcription
of the virF promoter is kept repressed at non-permissive temperatures
(below 32◦
C) by the nucleoid-associated protein HNS
(Colonna et al. 1995). Moreover, it was shown that the virF
promoter harbors two H-NS binding sites separated by an intrinsically
bent region, whose existence was predicted in silico
and demonstrated by in vitro experimental results (Falconi et al.
1998; Prosseda et al. 2004). Intriguingly, upon temperature increase,
this DNA region undergoes an abrupt structural transition
(at ∼32◦
C) that affects H-NS accessibility to target DNA sites
and represses the virF promoter (Prosseda et al. 2004). A similar
interplay between temperature-dependent DNA bending and
H-NS binding was demonstrated to modulate hemolysin gene
expression in the model organism E. coli (Madrid et al. 2002) and
to enhance the expression of a type III secretion system above
30◦
C in S. enterica (Duong et al. 2007).
To summarize, even though several examples of
temperature-dependent regulation of gene expression point
to the role of DNA as a sensor biomolecule, it appears that
this mechanism of heat sensing is much more pertinent to
regulating the expression of virulence genes rather than in the
heat-shock response.
Sensing through proteins
Bacteria can also sense and respond to temperature fluctuations
in their ecological niches by using proteins as heat sensors.
Different classes of proteins, including kinases, heat-shock
transcriptional repressors and chaperones, have been characterized
as sensors of temperature changes (Klinkert and Narberhaus
2009). However, the two latter categories of proteins
are primarily employed to transduce heat stress signals that
trigger a transcriptional heat-shock response. Intrinsic heatsensing
transcriptional repressors are able to directly modulate
the transcription of target genes in response to temperature fluctuation.
Specifically, they are competent for promoter binding
and repress gene transcription only at physiological temperature.
Upon heat-shock, these thermosensing repressors undergo
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 567
a conformational change that lowers relative binding affinity
for their operators. As a consequence, target gene transcription
becomes derepressed (Fig. 7C). With respect to the previous
point, CtsR, the global repressor of heat-shock genes of Bacillus
subtilis and other low-GC Gram-positive bacteria, is one of the
best characterized examples (Elsholz et al. 2010). In vitro DNAbinding
assays carried out at different temperatures showed
that CtsR-binding activity to the clpC promoter drastically drops
under heat-shock conditions (50◦
C), compared to normal growth
temperature at 37◦
C. Interestingly, the temperature-dependent
loss of DNA-binding activity of CtsR was demonstrated to be
reversible. In fact, when CtsR was first incubated at a nonpermissive
temperature and then at 37◦
C, the regulator regained
DNA-binding activity. In the case of CtsR, the temperaturedependent
conformational change responsible for the loss of
DNA-binding activity was shown to be limited to a short glycinerich
loop region within the DBD, constituting the precise functional
site for heat sensing (Elsholz et al. 2010). The Streptomyces
albus heat-shock repressor RheA described above was shown to
be capable of sensing temperature variations in the absence of
other factors, behaving similarly to CtsR. Upon heat challenge,
RheA loses DNA-binding activity and the transition from active
to inactive form was related to a temperature-induced reversible
conformational change (Servant, Grandvalet and Mazodier
2000). A similar paradigm of temperature-dependent
DNA-binding capacity was also clearly demonstrated for the
virulence-associated RovA regulator of Yersinia pestis and for the
TlpA regulator of S. enterica serovar Typhimurium (Hurme et al.
1997; Herbst et al. 2009). In the latter example, in particular, temperature
reversibly affects the DNA-binding competent coiledcoil
domains of the oligomerized state of the protein, thereby
affecting DNA-binding ability. Another widespread heat-shock
repressor, HrcA, can act, in some cases, as an intrinsic protein
thermometer. The observation that the thermal denaturation
profiles of B. subtilis and of B. thermoglucosidasius HrcACIRCE
complexes (assayed by light scattering) were highly consistent
with the different growth temperatures of the two
microorganisms led to the proposal that the HrcA repressor
might have a role as a thermosensor (Hitomi et al. 2003). A
direct role as thermosensor was recently demonstrated for
the HrcA repressor of Helicobacter pylori. Specifically, in vitro
DNaseI footprinting assay results showed that HrcA-mediated
DNA binding is strictly temperature dependent (Roncarati,
Danielli and Scarlato 2014). When H. pylori HrcA was exposed
to temperatures above 37◦
C (physiological growth temperature),
it dramatically lowered binding affinity to CIRCE operators
and became essentially inactive. Intriguingly, the loss
of binding activity upon heat-shock treatment appeared to
be irreversible in vitro, likely as a consequence of the major
structural change as the temperature exceeded 40◦
C. However,
the situation is different in vivo, where HrcA is able to recover
its repressive function after the temperature challenge
(Roncarati, Danielli and Scarlato 2014). Furthermore, it was
demonstrated that the GroESL chaperonin plays a key role and
that it is able to restore HrcA-binding activity lost upon heat
challenge. The working model for HrcA-mediated heat sensing
in H. pylori postulates that, after a sudden increase of temperature,
heat-mediated inactivation of the regulator leads to
promoter derepression and increased transcription of the heatshock
gene promoters. Then, following the induction step, at
least a fraction of the denatured repressor is refolded by GroESL
and can take part, together with the newly synthetized HrcA,
in transcriptional regulation of the target genes. Considering
the key role of GroESL in regulating HrcA-binding activity upon
heat sensing, this functional interaction could be exploited by
H. pylori to adjust the transcriptional response under various
stress conditions. That is to say that, depending on the severity
of a particular stress stimulus, the amount of GroE chaperonin
available for the functional interaction with HrcA will
change, influencing the restoration of target gene regulation. A
different situation was observed for Chlamydia trachomatis HrcAdependent
heat-shock gene regulation. In this obligate intracellular
human pathogen, HrcA binds to CIRCE operators within
groESL and dnaK promoters and represses their transcription under
normal growth conditions. Heat-shock gene transcription is
then induced upon a sudden temperature increase. However, it
was shown that HrcA-binding activity to CIRCE was not affected
by elevated temperature in vitro (Wilson and Tan 2004). Recently,
Hanson and Tan (2015) developed a method based on ChIP to
study heat-shock gene regulation during an intracellular infection,
and they were able to monitor in vivo HrcA binding to target
promoters at various temperatures. This approach allowed
demonstration that, upon a sudden temperature increase, the
in vivo induction of heat-shock gene transcription was due to a
decrease of HrcA binding to groESL and dnaK promoters. The discrepancy
between the in vitro and in vivo results for HrcA temperature
sensitivity suggests that elevated temperature is not sufficient
per se to modulate chlamydial HrcA-binding capacity and
implies that additional in vivo factors must be involved in the
regulation of the HrcA-mediated heat-shock response. Based on
previous observations of a functional interaction between HrcA
and GroE (Wilson et al. 2005), it has been proposed that the chaperonin
is the actual heat sensor of the regulatory circuit. The
latter example represents a well-characterized mechanism employed
in several cases that link environmental stress signals
to a corresponding transcriptional response. This signal-sensing
system does not directly detect the temperature increase; rather,
the consequences of heat-shock on cellular proteins are sensed.
Specifically, this strategy, represented in Fig. 7D, is based on
the direct interaction between heat-shock transcriptional regulators
and the major cellular chaperones. DNA-binding activity
of transcriptional regulators is positively or negatively regulated
by chaperones that are available for interaction during normal
growth conditions. Upon sudden temperature upshift, chaperones
are titrated away by denatured proteins that accumulate
in the cytoplasm and transcriptional regulators are released. As
described above (Figs 2 and Fig. 4A and B), the activity of HrcA
and HspR repressors of several bacterial species, but also of the
heat-shock σ32
of E. coli, is regulated in this way by the principal
cellular chaperones such as the GroE chaperonin and DnaKDnaJ-GrpE
system (Straus, Walter and Gross 1990; Babst, Hennecke
and Fischer 1996; Mogk et al. 1997; Bucca et al. 2000; Reischl,
Wiegert and Schumann 2002; Wilson et al. 2005). According
to this indirect heat-sensing mechanism, chaperones, and not
transcriptional regulators, are the actual ‘stress sensors’.
An interesting aspect emerging from the examples of the
sensing mechanisms described above is that, in several cases,
the regulation of the heat-shock response involves both direct
temperature sensing (through RNA thermometers and
intrinsic heat-sensing transcriptional regulators) and indirect
chaperone-mediated sensing of unfolded proteins. These composed
systems allow bacteria to respond differentially, depending
on severity and type of environmental stresses. For example,
RNA thermometers can detect minor temperature variations
allowing the system to immediately respond, before cellular
processes are altered. In this respect, E. coli σ32
regulation represents
a perfect example. Here temperature-regulated translation
of RpoH mRNA combines with two feedback loops that
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
568 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
control σ32
activity and stability. An interesting implication of
these combined systems of signal sensing is that cells might
discriminate heat stress from any other type of stress. Consequently,
bacteria might differentially induce the expression
of specific set of genes, thus increasing specificity of stress
response.
CONCLUSIONS
This review highlights the importance of a sudden response of
various bacterial species to temperature changes and summarizes
the key mechanisms that diverse bacteria adopt to counteract
imminent cellular damage. This highly conserved defense
mechanism against environmental stress, known as the heatshock
response, appears to be controlled by a plethora of different
strategies in bacteria. Control of heat-shock gene transcription
is successfully and efficiently achieved by the use of
specialized regulatory proteins exerting their positive or negative
action on the initiation of transcription by the RNA polymerase
enzyme. Strikingly, positive and negative mechanisms
of regulation, alone or in combination, are employed by microorganisms
to rapidly reprogram gene transcription upon stress
perception. Additionally, control of gene transcription can be
combined with posttranscriptional mechanisms of protein ex-
pression.
The model organism Escherichia coli has long served as
the central paradigm for understanding the mechanism of
stress induced activation of heat-shock gene transcription
based on the use of alternative sigma factors. After decades of
intense research on the heat-shock sigma factor σ32
, it appears
that a complex multilayered regulatory cascade orchestrates
σ32
homeostasis, in which transcriptional and translational
control combines with chaperone mediated modulation of σ32
stability. Moreover, the recent discovery of inner membrane
localization of the heat-shock sigma factor mediated by the
SRP-SR co-translational targeting system adds another key part
to this intricate story. However, further efforts are needed to
understand several mechanistic aspects of the transit of σ32
to
the membrane.
Negative regulation of heat-shock gene transcription through
dedicated repressors has been studied in detail in several bacterial
species. Different aspects of the major heat-shock repressors
have been characterized, such as their DNA-binding
mechanism, the characterization of their regulon and their biochemical
properties (such as, for example, their heat-sensing
capabilities). In this respect, it is worth noting that the way
in which the master heat-shock repressor HspR senses fluctuations
of environmental temperature remains elusive. In fact,
while for HrcA, CtsR and RheA several studies pinpointed the direct
or indirect (chaperone-mediated) mechanisms of heat sensing,
the HspR repressor appears to be a heat stable protein and
its interplay with the DnaK chaperone complex could be experimentally
demonstrated in only a few cases. Further studies
should be carried out to understand how temperature can modulate
the DNA-binding activity of HspR.
In several cases, heat-shock regulatory circuits appear to be
built on very complex structures in which mechanisms of transcriptional
control combine with several feedback loops, whose
effects on the properties of the regulatory systems are hardly
predictable. In this respect, the question arises as to whether
the use of such complex regulation strategies in the heat-shock
systems is the result of evolutionary accidents or it derives from
specifically designed solutions to different requirements in the
field of heat-shock response. Rigorous analyses have been carried
out with systems biology approaches and computational
simulations of E. coli heat-shock response (El-Samad et al. 2005;
Kurata et al. 2006). As described above, the E. coli heat-shock response
involves complicated interactions, where both the activity
and stability of the alternative sigma factor σ32
are feedback
controlled by chaperones and proteases, respectively. From
these studies, it emerges that the use of feedback loops gives
valuable properties to the system, in terms in primis of robustness,
defined as the capability of a system to operate reliably
when its physical parameters vary within their expected ranges.
Furthermore, the use of FtsH protease-mediated degradation
feedback appears to provoke a faster response to a heat stimulation
and to limit the effects of biochemical noise (El-Samad
et al. 2005; Kurata et al. 2006). Further similar studies will undoubtedly
be crucial for the investigation and characterization
of other complex heat-shock regulatory schemes.
Another intriguing aspect of heat-shock gene regulation that
deserves further characterization concerns the combination of
different transcriptional regulators that establish highly complex
regulatory schemes. In particular, the cases of Staphylococcus
aureus and Helicobacter pylori, for which two repressors are
employed and the regulon of one repressor is embedded within
the regulon of the other repressor, represent rare examples of
regulatory networks whose functional implications remain to be
defined.
The recent development and application of the ‘omics’
techniques will allow further detailed characterization and understanding
of the heat-shock regulons specifically governed by
various heat-shock regulators. Functional genomics studies are
expected to deepen and expand our knowledge of the impact of
heat stress on the bacterial transcriptome and of the regulatory
mechanisms that control such responses.
FUNDING
This work was supported by Grants from the Italian Ministry of
Education, University and Research (2010P3S8BR 003) and by a
grant from the University of Bologna.
Conflict of interest. None declared.
REFERENCES
Ades SE. Regulation by destruction: design of the σE
envelope
stress response. Curr Opin Microbiol 2008;11:535–40.
Allan B, Linseman M, MacDonald LA et al. Heat shock response
of Pseudomonas aeruginosa. J Bacteriol 1988;170:3668–74.
Alon U. Network motifs: theory and experimental approaches.
Nat Rev Genet 2007;8:450–61.
Andersen MT, Brøndsted L, Pearson BM et al. Diverse roles
for HspR in Campylobacter jejuni revealed by the proteome,
transcriptome and phenotypic characterization of an hspR
mutant. Microbiology 2005;151:905–15.
Babst M, Hennecke H, Fischer HM. Two different mechanisms
are involved in the heat-shock regulation of chaperonin
gene expression in Bradyrhizobium japonicum. Mol Microbiol
1996;19:827–39.
Bandyopadhyay B, Das Gupta T, Roy D et al. DnaK dependence of
the mycobacterial stress responsive regulator HspR is mediated
through its hydrophobic C-terminal tail. J Bacteriol
2012;194:4688–97.
Barnwal RP, Loh E, Godin KS et al. Structure and mechanism
of a molecular rheostat, an RNA thermometer that modulates
immune evasion by Neisseria meningitidis. Nucleic Acids
Res 2016;44:9426–37.
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 569
Becskei A, Serrano L. Engineering stability in gene networks by
autoregulation. Nature 2000;405:590–3.
Bird JG, Sharma S, Roshwalb SC et al. Functional analysis of
CbpA, a DnaJ homolog and nucleoid-associated DNA-binding
protein. J Biol Chem 2006;281:34349–56.
Briat JF, Gilman MJ, Chamberlin MJ. Bacillus subtilis σ28
and Escherichia
coli σ32
(htpR) are minor factors that display an overlapping
promoter specificity. J Biol Chem 1985;260:2038–41.
Bucca G, Brassington AME, Hotchkiss G et al. Negative feedback
regulation of dnaK, clpB and lon expression by the DnaK chaperone
machine in Streptomyces coelicolor, identified by transcriptome
and in vivo DnaK-depletion analysis. Mol Microbiol
2003;50:153–66.
Bucca G, Brassington AME, Sch¨onfeld HJ et al. The HspR regulon
of Streptomyces coelicolor: a role for the DnaK chaperone as a
transcriptional co-repressor. Mol Microbiol 2000;38:1093–103.
Bucca G, Ferina G, Puglia AM et al. The dnaK operon of Streptomyces
coelicolor encodes a novel heat-shock protein which
binds to the promoter region of the operon. Mol Microbiol
1995;17:663–74.
Bucca G, Hindle Z, Smith CP. Regulation of the dnaK operon of
Streptomyces coelicolor A3(2) is governed by HspR, an autoregulatory
repressor protein. J Bacteriol 1997;179:5999–6004.
Bucca G, Laing E, Vassilis M et al. Development and application of
versatile high density microarrays for genome-wide analysis
of Streptomyces coelicolor: characterization of the HspR regulon.
Genome Biol 2009;10:R5.
ˇCerven´y J, Sinetova MA, Zavˇrel T et al. Mechanisms of high temperature
resistance of Synechocystis sp. PCC 6803: an impact
of Histidine Kinase 34. Life (Basel) 2015;5:676–99.
Cezairliyan BO, Sauer RT. Inhibition of regulated proteolysis by
RseB. P Natl Acad Sci USA 2007;104:3771–6.
Chaba R, Grigorova IL, Flynn JM et al. Design principles of the
proteolytic cascade governing the sigmaE-mediated envelope
stress response in Escherichia coli: keys to graded, buffered,
and rapid signal transduction. Gene Dev 2007;21:124–36.
Chae C, Sharma S, Hoskins JR et al. CbpA, a DnaJ homolog, is a
DnaK co-chaperone, and its activity is modulated by CbpM. J
Biol Chem 2004;279:33147–53.
Chang BY, Chen KY, Wen YD et al. The response of a Bacillus
subtilis temperature-sensitive sigA mutant to heat stress.
J Bacteriol 1994;176:3102–10.
Charpentier B, Branlant C. The Escherichia coli gapA gene is transcribed
by the vegetative RNA polymerase holoenzyme E
Sigma 70 and by the heat shock RNA polymerase E Sigma
32. J Bacteriol 1994;176:830–9.
Chastanet A, Derre I, Nair S et al. clpB, a novel member of
the Listeria monocytogenes CtsR regulon, is involved in virulence
but not in general stress tolerance. J Bacteriol 2004;186:
1165–74.
Chastanet A, Fert J, Msadek T. Comparative genomics reveal
novel heat shock regulatory mechanisms in Staphylococcus
aureus and other Gram-positive bacteria. Mol Microbiol
2003;47:1061–73.
Chastanet A, Msadek T. ClpP of Streptococcus salivarius is a novel
member of the dually regulated class of stress response
genes in gram-positive bacteria. J Bacteriol 2003;185:683–7.
Chastanet A, Prudhomme M, Claverys JP et al. Regulation of
Streptococcus pneumoniae clp genes and their role in competence
development and stress survival. J Bacteriol 2001;183:
7295–307.
Chen AL, Wilson AC, Tan M. A Chlamydia-specific C-terminal region
of the stress response regulator HrcA modulates its repressor
activity. J Bacteriol 2011;193:6733–41.
Chowdhury S, Maris C, Allain FH et al. Molecular basis for
temperature sensing by an RNA thermometer. EMBO J
2006;25:2487–97.
Chowdhury S, Ragaz C, Kreuger E et al. Temperature-controlled
structural alterations of an RNA thermometer. J Biol Chem
2003;278:47915–21.
Coates AR, Shinnick TM, Ellis RJ. Chaperonin nomenclature. Mol
Microbiol 1993;8:787.
Colonna B, Casalino M, Fradiani PA et al. H-NS regulation of virulence
gene expression in enteroinvasive Escherichia coli harboring
the virulence plasmid integrated into the host chromosome.
J Bacteriol 1995;177:4703–12.
Danielli A, Amore G, Scarlato V. Built shallow to maintain homeostasis
and persistent infection: insight into the transcriptional
regulatory network of the gastric human pathogen Helicobacter
pylori. PLoS Pathog 2010;6:e1000938.
Dartigalongue C, Missiakas D, Raina S. Characterization of
the Escherichia coli sigma E regulon. J Biol Chem 2001;276:
20866–75.
Das Gupta T, Bandyopadhyay B, Das Gupta SK. Modulation of
DNA-binding activity of Mycobacterium tuberculosis HspR by
chaperones. Microbiology 2008;154:484–90.
Delany I, Spohn G, Rappuoli R et al. In vitro selection of high affinity
HspR-binding sites within the genome of Helicobacter pylori.
Gene 2002;283:63–9.
de Reuse H, Vinella D, Cavazza C. Common themes and unique
proteins for the uptake and trafficking of nickel, a metal essential
for the virulence of Helicobacter pylori. Front Cell Infect
Microbiol 2013;3:94.
Derr´e I, Rapoport G, Devine K et al. ClpE, a novel type of HSP100
ATPase, is part of the CtsR heat shock regulon of Bacillus subtilis.
Mol Microbiol 1999;32:581–93.
Derr´e I, Rapoport G, Msadek T. CtsR, a novel regulator of stress
and heat shock response, controls clp and molecular chaperone
gene expression in gram-positive bacteria. Mol Microbiol
1999;31:117–31.
Derr´e I, Rapoport G, Msadek T. The CtsR regulator of stress response
is active as a dimer and specifically degraded in vivo
at 37 degrees C. Mol Microbiol 2000;38:335–47.
Dorman CJ, Corcoran CP. Bacterial DNA topology and infectious
disease. Nucleic Acids Res 2009;37:672–8.
Duong N, Osborne S, Bustamante VH et al. Thermosensing
coordinates a cis-regulatory module for transcriptional
activation of the intracellular virulence system in Salmonella
enterica serovar Typhimurium. J Biol Chem 2007;282:
34077–84.
El-Samad H, Kurata H, Doyle JC et al. Surviving heat shock: control
strategies for robustness and performance. P Natl Acad
Sci USA 2005;102:2736–41.
Elsholz AK, Michalik S, Z ¨uhlke D et al. CtsR, the Gram-positive
master regulator of protein quality control, feels the heat.
EMBO J 2010;29:3621–9.
Engels S, Schweitzer JE, Ludwig C et al. clpC and clpP1P2 gene expression
in Corynebacterium glutamicum is controlled by a regulatory
network involving the transcriptional regulators ClgR
and HspR as well as the ECF sigma factor sigmaH. Mol Microbiol
2004;52:285–302.
Erickson JW, Gross CA. Identification of the σE
subunit of
Escherichia coli RNA polymerase: a second alternative sigma
involved in high temperature gene expression. Gene Dev
1989;3:1462–71.
Erickson JW, Vaughn V, Walter WA et al. Regulation of the promoters
and transcripts of rpoH, the Escherichia coli heat shock
regulatory gene. Gene Dev 1987;1:419–32.
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
570 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
Falconi M, Colonna B, Prosseda G et al. Thermoregulation
of Shigella and Escherichia coli EIEC pathogenicity.
A temperature-dependent structural transition of DNA modulates
accessibility of virF promoter to transcriptional repressor
H-NS. EMBO J 1998;17:7033–43.
Fiocco D, Capozzi V, Collins M et al. Characterization of the CtsR
stress response regulon in Lactobacillus plantarum. J Bacteriol
2010;192:896–900.
Fiocco D, Collins M, Muscariello L et al. The Lactobacillus plantarum
ftsH gene is a novel member of the CtsR stress response
regulon. J Bacteriol 2009;191:1688–94.
Fischer HM. Genetic regulation of nitrogen fixation in rhizobia.
Microbiol Rev 1994;58:352–86.
Fuhrmann J, Schmidt A, Spiess S et al. McsB is a protein arginine
kinase that phosphorylates and inhibits the heat shock regulator
CtsR. Science 2009;324:1323–7.
Gamer J, Bujard H, Bukau B. Physical interaction between heat
shock proteins DnaK, DnaJ, and GrpE and the bacterial heat
shock transcription factor sigma 32. Cell 1992;69:833–42.
Georgopoulos CP, Hendrix RW, Kaiser AD et al. Role of the host
cell in bacteriophage morphogenesis: effects of a bacterial
mutation on T4 head assembly. Nat New Biol 1972;239:38–41.
Gerth U, Kr ¨uger E, Derr`e I et al. 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 1998;28:787–802.
Grandvalet C, Coucheney F, Beltramo C et al. CtsR is the master
regulator of stress response gene expression in Oenococcus
oeni. J Bacteriol 2005;187:5614–23.
Grandvalet C, de Cr´ecy-Lagard V, Mazodier P. The ClpB ATPase of
Streptomyces albus G belongs to the HspR heat shock regulon.
Mol Microbiol 1999;31:521–32.
Grandvalet C, Rapoport G, Mazodier P. hrcA, encoding the repressor
of the groEL genes in Streptomyces albus G, is associated
with a second dnaJ gene. J Bacteriol 1998;180:5129–34.
Grandvalet C, Servant P, Mazodier P. Disruption of hspR, the repressor
gene of the dnaK operon in Streptomyces albus G. Mol
Microbiol 1997;23:77–84.
Grigorova IL, Chaba R, Zhong HJ et al. Fine-tuning of the
Escherichia coli σE
envelope stress response relies on multiple
mechanisms to inhibit signal-independent proteolysis
of the trans- membrane anti-sigma factor, RseA. Gene Dev
2004;18:2686–97.
Grossman AD, Erickson JW, Gross CA. The htpR gene product
of E. coli is a sigma factor for heat-shock promoters. Cell
1984;38:383–90.
Grossman AD, Straus DB, Walter WA et al. Sigma 32 synthesis can
regulate the synthesis of heat shock proteins in Escherichia
coli. Gene Dev 1987;1:179–84.
Grosso-Becera MV, Serv´ın-Gonz´alez L, Sober´on-Ch´avez G. RNA
structures are involved in the thermoregulation of bacterial
virulence-associated traits. Trends Microbiol 2015;23:
509–18.
Guisbert E, Herman C, Lu CZ et al. A chaperone network controls
the heat shock response in E. coli. Gene Dev 2004;18:2812–21.
Hanson BR, Tan M. Transcriptional regulation of the Chlamydia
heat shock stress response in an intracellular infection. Mol
Microbiol 2015;97:1158–67.
Helmann JD. The extracytoplasmic function (ECF) sigma factors.
Adv Microb Physiol 2002;46:47–110.
Henderson B, Martin A. Bacterial virulence in the moonlight:
multitasking bacterial moonlighting proteins are virulence
determinants in infectious disease. Infect Immun
2011;79:3476–91.
Herbst K, Bujara M, Heroven AK et al. Intrinsic thermal sensing
controls proteolysis of Yersinia virulence regulator RovA. PLoS
Pathog 2009;5:e1000435.
Herman C, Th´evenet D, D’Ari R et al. Degradation of sigma 32, the
heat shock regulator in Escherichia coli, is governed by HflB. P
Natl Acad Sci USA 1995;92:3516–20.
Hitomi M, Nishimura H, Tsujimoto Y et al. Identification of a
helix-turn-helix motif of Bacillus thermoglucosidasius HrcA essential
for binding to the CIRCE element and thermostability
of the HrcA-CIRCE complex, indicating a role as a thermosensor.
J Bacteriol 2003;185:381–5.
Holmes CW, Penn CW, Lund PA. The hrcA and hspR regulons of
Campylobacter jejuni. Microbiology 2010;156:158–66.
Hsieh LS, Burger RM, Drlica K. Bacterial DNA supercoiling and
[ATP]/[ADP]. Changes associated with a transition to anaerobic
growth. J Mol Biol 1991;219:443–50.
Hurme R, Berndt KD, Normark SJ et al. A proteinaceous gene regulatory
thermometer in Salmonella. Cell 1997;90:55–64.
Hu Y, Raengpradub S, Schwab U et al. Phenotypic and transcriptomic
analyses demonstrate interactions between the transcriptional
regulators CtsR and Sigma B in Listeria monocytogenes.
Appl Environ Microb 2007;73:7967–80.
Janaszak A, Nadratowska-Wesolowska B, Konopa G et al. The P1
promoter of the Escherichia coli rpoH gene is utilized by sigma
70-RNAP or sigma s-RNAP depending on growth phase. FEMS
Microbiol Lett 2009;291:65–72.
J¨ager S, J¨ager A, Klug G. CIRCE is not involved in heat-dependent
transcription of groESL but in stabilization of the mRNA
5 -end in Rhodobacter capsulatus. Nucleic Acids Res 2004;32:
386–96.
Jastrab JB, Wang T, Murphy JP et al. An adenosine triphosphateindependent
proteasome activator contributes to the virulence
of Mycobacterium tuberculosis. P Natl Acad Sci USA
2015;112:E1763–72.
Kallipolitis BH, Valentin-Hansen P. Transcription of rpoH, encoding
the Escherichia coli heat-shock regulator sigma32, is negatively
controlled by the cAMP-CRP/CytR nucleoprotein complex.
Mol Microbiol 1998;29:1091–9.
Kamath-Loeb AS, Gross CA. Translational regulation of sigma 32
synthesis: requirement for an internal control element. J Bacteriol
1991;173:3904–6.
Kanehara K, Ito K, Akiyama Y. YaeL (EcfE) activates the σE
pathway
of stress response through a site-2 cleavage of anti-σE
,
RseA. Gene Dev 2002;16:2147–55.
Kanehara K, Ito K, Akiyama K. YaeL proteoylsis of RseA is controlled
by the PDZ domain of YaeL and a Gln-rich region of
RseA. EMBO J 2003;22:6389–98.
Kanemori M, Mori H, Yura T. Effects of reduced levels of GroE
chaperones on protein metabolism: enhanced synthesis of
heat shock proteins during steady-state growth of Escherichia
coli. J Bacteriol 1994;176:4235–42.
Kanemori M, Nishihara K, Yanagi H et al. Synergistic roles of
HslVU and other ATP-dependent proteases in controlling in
vivo turnover of sigma32 and abnormal proteins in Escherichia
coli. J Bacteriol 1997;179:7219–25.
Karunakaran KP, Noguchi Y, Read TD et al. Molecular analysis
of the multiple GroEL proteins of Chlamydiae. J Bacteriol
2003;185:1958–66.
Katayama S, Matsushita O, Jung CM et al. Promoter upstream
bent DNA activates the transcription of the Clostridium
perfringens phospholipase C gene in a low temperaturedependent
manner. EMBO J 1999;18:3442–50.
Kirstein J, Dougan DA, Gerth U et al. The tyrosine kinase McsB is a
regulated adaptor protein for ClpCP. EMBO J 2007;26:2061–70.
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 571
Kirstein J, Z ¨uhlke D, Gerth U et al. A tyrosine kinase and its activator
control the activity of the CtsR heat shock repressor in
B. subtilis. EMBO J 2005;24:3435–45.
Klinkert B, Narberhaus F. Microbial thermosensors. Cell Mol Life
Sci 2009;66:2661–76.
Kojima K, Nakamoto H. A novel light- and heat-responsive regulation
of the groE transcription in the absence of HrcA or
CIRCE in cyanobacteria. FEBS Lett 2007;581:1871–80.
Koo BM, Rhodius VA, Campbell EA et al. Dissection of recognition
determinants of Escherichia coli sigma32 suggests a composite
-10 region with an ’extended -10 motif and a core -10
element. Mol Microbiol 2009;72:815–29.
Kortmann J, Narberhaus F. Bacterial RNA thermometers: molecular
zippers and switches. Nat Rev Microbiol 2012;10:255–65.
Kortmann J, Sczodrok S, Rinnenthal J et al. Translation on demand
by a simple RNA-based thermosensor. Nucleic Acids Res
2011;39:2855–68.
Kourennaia OV, Tsujikawa L, Dehaseth PL. Mutational analysis
of Escherichia coli heat shock transcription factor sigma 32 reveals
similarities with sigma 70 in recognition of the -35 promoter
element and differences in promoter DNA melting and
-10 recognition. J Bacteriol 2005;187:6762–9.
Kr ¨uger E, Hecker M. The first gene of the Bacillus subtilis clpC
operon, ctsR, encodes a negative regulator of its own operon
and other class III heat shock genes. J Bacteriol 1998;180:
6681–8.
Kr ¨uger E, Msadek T, Hecker M. Alternate promoters direct stressinduced
transcription of the Bacillus subtilis clpC operon. Mol
Microbiol 1996;20:713–23.
Kr ¨uger E, Msadek T, Ohlmeier S et al. The Bacillus subtilis clpC
operon encodes DNA repair and competence proteins. Microbiology
1997;143 (Pt 4):1309–16.
Kr ¨uger E, Z ¨uhlke D, Witt E et al. Clp-mediated proteolysis in
Gram-positive bacteria is autoregulated by the stability of a
repressor. EMBO J 2001;20:852–63.
Kumar A, Grimes B, Logan M et al. A hybrid sigma subunit directs
RNA polymerase to a hybrid promoter in Escherichia coli. J Mol
Biol 1995;246:563–71.
Kurata H, El-Samad H, Iwasaki R et al. Module-based analysis of
robustness tradeoffs in the heat shock response system. PLoS
Comput Biol 2006;2:e59.
Landick R, Vaughn V, Lau ET et al. Nucleotide sequence of the
heat shock regulatory gene of E. coli suggests its protein product
may be a transcription factor. Cell 1984;38:175–82.
Laport MS, Lemos JA, Bastos Md Mdo C et al. Transcriptional analysis
of the groE and dnaK heat-shock operons of Enterococcus
faecalis. Res Microbiol 2004;155:252–8.
Lesley SA, Thompson NE, Burgess RR. Studies of the role of the
Escherichia coli heat shock regulatory protein sigma 32 by the
use of monoclonal antibodies. J Biol Chem 1987;262:5404–7.
Li M, Wang SL. Cloning and characterization of the groESL operon
from Bacillus subtilis. J Bacteriol 1992;174:3981–92.
Liberek K, Galitski TP, Zylicz M et al. The DnaK chaperone modulates
the heat shock response of Escherichia coli by binding
to the sigma 32 transcription factor. P Natl Acad Sci USA
1992;89:3516–20.
Lim B, Miyazaki R, Neher S et al. Heat shock transcription factor
σ32
co-opts the signal recognition particle to regulate protein
homeostasis in E. coli. PLoS Biol 2013;11:e1001735.
Lima S, Guo MS, Chaba R et al. Dual molecular signals mediate
the bacterial response to outer-membrane stress. Science
2013;340:837–41.
Liu J, Huang C, Shin DH et al. Crystal structure of a heat-inducible
transcriptional repressor HrcA from Thermotoga maritima:
structural insight into DNA binding and dimerization. J Mol
Biol 2005;350:987–96.
Loh E, Kugelberg E, Tracy A et al. Temperature triggers immune
evasion by Neisseria meningitidis. Nature 2013;502:237–40.
L´opez-Garc´ıa P, Forterre P. DNA topology in hyperthermophilic
archaea: reference states and their variation with growth
phase, growth temperature, and temperature stresses. Mol
Microbiol 1997;23:1267–79.
Lund PA. Multiple chaperonins in bacteria–why so many? FEMS
Microbiol Rev 2009;33:785–800.
Madrid C, Nieto JM, Paytubi S et al. Temperature- and
H-NS-dependent regulation of a plasmid-encoded virulence
operon expressing Escherichia coli hemolysin. J Bacteriol
2002;184:5058–66.
Matuszewska M, Kuczy ´nska-Wi´snik D, Laskowska E et al. The
small heat shock protein IbpA of Escherichia coli cooperates
with IbpB in stabilization of thermally aggregated
proteins in a disaggregation competent state. J Biol Chem
2005;280:12292–8.
Mazodier P, Guglielmi G, Davies J et al. Characterization
of the groEL-like genes in Streptomyces albus. J Bacteriol
1991;173:7382–6.
Mecsas J, Rouviere PE, Erickson JW et al. The activity of sigma
E, an Escherichia coli heat inducible sigma-factor, is modulated
by expression of outer membrane proteins. Genes Dev
1993;7(12B):2618–28.
Missiakas D, Mayer MP, Lemaire M et al. Modulation of the Escherichia
coli σE
(RpoE) heat- shock transcription-factor activity
by the RseA, RseB and RseC proteins. Mol Microbiol
1997;24:355–71.
Missiakas D, Raina S. The extracytoplasmic function sigma factors:
role and regulation. Mol Microbiol 1998;28:1059–66.
Missiakas D, Schwager F, Betton JM et al. Identification and characterization
of HsIV HsIU (ClpQ ClpY) proteins involved in
overall proteolysis of misfolded proteins in Escherichia coli.
EMBO J 1996;15:6899–909.
Miyazaki R, Yura T, Suzuki T et al. A novel SRP recognition sequence
in the homeostatic control region of heat shock transcription
factor σ(32). Sci Rep 2016;6:24147.
Mogk A, Homuth G, Scholz C et al. The GroE chaperonin machine
is a major modulator of the CIRCE heat shock regulon of Bacillus
subtilis. EMBO J 1997;16:4579–90.
Mogk A, Tomoyasu T, Goloubinoff P et al. Identification of thermolabile
Escherichia coli proteins: prevention and reversion of
aggregation by DnaK and ClpB. EMBO J 1999;18:6934–49.
Mogk A, V¨olker A, Engelmann S et al. Nonnative proteins induce
expression of the Bacillus subtilis CIRCE regulon. J Bacteriol
1998;180:2895–900.
Morita MT, Tanaka Y, Kodama TS et al. Translational induction of
heat shock transcription factor sigma32: evidence for a builtin
RNA thermosensor. Gene Dev 1999;13:655–65.
Musatovova O, Dhandayuthapani S, Baseman JB. Transcriptional
heat shock response in the smallest known self-replicating
cell, Mycoplasma genitalium. J Bacteriol 2006;188:2845–55.
Nair S, Derr´e I, Msadek T et al. CtsR controls class III heat shock
gene expression in the human pathogen Listeria monocytogenes.
Mol Microbiol 2000;35:800–11.
Nakahigashi K, Ron EZ, Yanagi H et al. Differential and independent
roles of a sigma(32) homolog (RpoH) and an HrcA repressor
in the heat shock response of Agrobacterium tumefaciens.
J Bacteriol 1999;181:7509–15.
Nakamoto H, Suzuki M, Kojima K. Targeted inactivation of the
hrcA repressor gene in cyanobacteria. FEBS Lett 2003;549:
57–62.
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
572 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
Nagai H, Yano R, Erickson JW et al. Transcriptional regulation
of the heat shock regulatory gene rpoH in Escherichia coli: involvement
of a novel catabolite-sensitive promoter. J Bacteriol
1990;172:2710–7.
Nagai H, Yuzawa H, Yura T. Interplay of two cis-acting mRNA
regions in translational control of sigma 32 synthesis during
the heat shock response of Escherichia coli. P Natl Acad Sci USA
1991;88:10515–9.
Narberhaus F. Negative regulation of bacterial heat shock genes.
Mol Microbiol 1999;31:1–8.
Narberhaus F, K¨aser R, Nocker A et al. A novel DNA element that
controls bacterial heat shock gene expression. Mol Microbiol
1998;28:315–23.
Nevozhay D, Adams RM, Murphy KF et al. Negative autoregulation
linearizes the dose-response and suppresses the
heterogeneity of gene expression. P Natl Acad Sci USA
2009;106:5123–8.
Newlands JT, Gaal T, Mecsas J et al. Transcription of the Escherichia
coli rrnB P1 promoter by the heat shock RNA polymerase
(E sigma 32) in vitro. J Bacteriol 1993;175:661–8.
Nickerson CA, Achberger EC. Role of curved DNA in binding
of Escherichia coli RNA polymerase to promoters. J Bacteriol
1995;177:5756–61.
Nocker A, Hausherr T, Balsiger S et al. A mRNA-based thermosensor
controls expression of rhizobial heat shock genes.
Nucleic Acids Res 2001;29:4800–7.
Nonaka G, Blankschien M, Herman C et al. Regulon and promoter
analysis of the E. coli heat-shock factor, sigma32, reveals
a multifaceted cellular response to heat stress. Gene Dev
2006;20:1776–89.
Obrist M, Narberhaus F. Identification of a turnover element in
region 2.1 of Escherichia coli sigma32 by a bacterial one-hybrid
approach. J Bacteriol 2005;187:3807–13.
Ojha A, Anand M, Bhatt A et al. GroEL1: a dedicated chaperone
involved in mycolic acid biosynthesis during biofilm formation
in mycobacteria. Cell 2005;123:861–73.
Pallen M. RpoN-dependent transcription of rpoH? Mol Microbiol
1999;31:393.
Parijat P, Batra JK. Role of DnaK in HspR-HAIR interaction of Mycobacterium
tuberculosis. IUBMB Life 2015;67:816–27.
Patra M, Roy SS, Dasgupta R et al. GroEL to DnaK chaperone
network behind the stability modulation of σ(32) at physiological
temperature in Escherichia coli. FEBS Lett 2015;589:
4047–52.
Perrody E, Cirinesi AM, Desplats C et al. A bacteriophage-encoded
J-domain protein interacts with the DnaK/Hsp70 chaperone
and stabilizes the heat-shock factor σ32
of Escherichia coli.
PLoS Genet 2012;8:e1003037.
Prosseda G, Falconi M, Giangrossi M et al. The virF promoter in
Shigella: more than just a curved DNA stretch. Mol Microbiol
2004;51:523–37.
Rajaram H, Chaurasia AK, Apte SK. Cyanobacterial heat-shock
response: role and regulation of molecular chaperones. Microbiology
2014;160:647–58.
Reischl S, Wiegert T, Schumann W. Isolation and analysis of
mutant alleles of the Bacillus subtilis HrcA repressor with reduced
dependency on GroE function. J Biol Chem 2002;277:
32659–67.
Rhodius VA, Suh WC, Nonaka G et al. Conserved and variable
functions of the sigmaE stress response in related genomes.
PLoS Biol 2006;4:e2.
Rinnenthal J, Klinkert B, Narberhaus F et al. Modulation of the
stability of the Salmonella fourU-type RNA thermometer. Nucleic
Acids Res 2011;39:8258–70.
Ritossa F. A new puffing pattern induced by temperature shock
and DNP in Drosophila. Experientia 1962;18:571–73.
Roberts RC, Toochinda C, Avedissian M et al. Identification of
a Caulobacter crescentus operon encoding hrcA, involved in
negatively regulating heat-inducible transcription, and the
chaperone gene grpE. J Bacteriol 1996;178:1829–41.
Roncarati D, Danielli A, Scarlato V. CbpA acts as a modulator of
HspR repressor DNA binding activity in Helicobacter pylori. J
Bacteriol 2011;193:5629–36.
Roncarati D, Danielli A, Scarlato V. The HrcA repressor is the
thermosensor of the heat-shock regulatory circuit in the
human pathogen Helicobacter pylori. Mol Microbiol 2014;92:
910–20.
Roncarati D, Danielli A, Spohn G et al. Transcriptional regulation
of stress response and motility functions in Helicobacter pylori
is mediated by HspR and HrcA. J Bacteriol 2007a;189:7234–43.
Roncarati D, Spohn G, Tango N et al. Expression, purification
and characterization of the membrane-associated HrcA
repressor protein of Helicobacter pylori. Protein Expr Purif
2007b;51:267–75.
Rosenfeld N, Elowitz MB, Alon U. Negative autoregulation speeds
the response times of transcription networks. J Mol Biol
2002;323:785–93.
Scharf C, Riethdorf S, Ernst H et al. Thioredoxin is an essential
protein induced by multiple stresses in Bacillus subtilis. J Bacteriol
1998;180:1869–77.
Schmid AK, Howell HA, Battista JR et al. HspR is a global negative
regulator of heat shock gene expression in Deinococcus
radiodurans. Mol Microbiol 2005;55:1579–90.
Schmidt A, Schiesswohl M, Volker U et al. Cloning, sequencing,
mapping, and transcriptional analysis of the groESL operon
from Bacillus subtilis. J Bacteriol 1992;174:3993–99.
Schmidt G, Hertel C, Hammes WP. Molecular characterisation of
the dnaK operon of Lactobacillus sakei LTH681. Syst Appl Microbiol
1999;22:321–8.
Schulz A, Tzschaschel B, Schumann W. Isolation and analysis of
mutants of the dnaK operon of Bacillus subtilis. Mol Microbiol
1995;15:421–9.
Schumann W. The Bacillus subtilis heat shock stimulon. Cell Stress
Chaperones 2003;8:207–17.
Servant P, Grandvalet C, Mazodier P. The RheA repressor is the
thermosensor of the HSP18 heat shock response in Streptomyces
albus. P Natl Acad Sci USA 2000;97:3538–43.
Servant P, Mazodier P. Heat induction of hsp18 gene expression
in Streptomyces albus G: transcriptional and posttranscriptional
regulation. J Bacteriol 1996;178:7031–6.
Servant P, Rapoport G, Mazodier P. RheA, the repressor of hsp18
in Streptomyces albus G. Microbiology 1999;145:2385–91.
Shapiro RS, Cowen LE. Thermal control of microbial development
and virulence: molecular mechanisms of microbial
temperature sensing. MBio 2012;3: e00238–12.
Sharp MM, Chan CL, Lu CZ et al. The interface of sigma with
core RNA polymerase is extensive, conserved, and functionally
specialized. Gene Dev 1999;13:3015–26.
Singh AK, Summerfield TC, Li H et al. The heat shock response
in the cyanobacterium Synechocystis sp. Strain PCC 6803 and
regulation of gene expression by HrcA and SigB. Arch Microbiol
2006;186:273–86.
Singh R, Anil Kumar V, Das AK et al. A transcriptional corepressor
regulatory circuit controlling the heat-shock
response of Mycobacterium tuberculosis. Mol Microbiol
2014;94:450–65.
Slabas AR, Suzuki I, Murata N et al. Proteomic analysis of the heat
shock response in Synechocystis PCC6803 and a thermally
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
Roncarati and Scarlato 573
tolerant knockout strain lacking the histidine kinase 34 gene.
Proteomics 2006;6:845–64.
Slamti L, Livny J, Waldor MK. Global gene expression and phenotypic
analysis of a Vibrio cholerae rpoH deletion mutant. J
Bacteriol 2007;189:351–62.
Spohn G, Danielli A, Roncarati D et al. Dual control of Helicobacter
pylori heat shock gene transcription by HspR and HrcA. J
Bacteriol 2004;186:2956–65.
Spohn G, Scarlato V. The autoregulatory HspR repressor protein
governs chaperone gene transcription in Helicobacter pylori.
Mol Microbiol 1999;34:663–74.
Stewart G, Wernisch L, Stabler R et al. Dissection of the heatshock
response in Mycobacterium tuberculosis using mutants
and microarrays. Microbiology 2002;148:3129–38.
Straus D, Walter W, Gross CA. DnaK, DnaJ, and GrpE heat shock
proteins negatively regulate heat shock gene expression by
controlling the synthesis and stability of sigma 32. Gene Dev
1990;4:2202–9.
Straus DB, Walter WA, Gross CA. The heat shock response of E.
coli is regulated by changes in the concentration of sigma 32.
Nature 1987;329:348–51.
Suokko A, Poutanen M, Savijoki K et al. ClpL is essential for
induction of thermotolerance and is potentially part of
the HrcA regulon in Lactobacillus gasseri. Proteomics 2008;8:
1029–41.
Suokko A, Savijoki K, Malinen E et al. Characterization of a mobile
clpL gene from Lactobacillus rhamnosus. Appl Environ Microb
2005;71:2061–9.
Susin MF, Perez HR, Baldini RL et al. Functional and structural
analysis of HrcA repressor protein from Caulobacter crescentus.
J Bacteriol 2004;186:6759–67.
Suzuki H, Ikeda A, Tsuchimoto S et al. Synergistic binding of DnaJ
and DnaK chaperones to heat shock transcription factor σ32
ensures its characteristic high metabolic instability: implications
for heat shock protein 70 (Hsp70)-Hsp40 mode of function.
J Biol Chem 2012;287:19275–83.
Takano T, Kakefuda T. Involvement of a bacterial factor
in morphogenesis of bacteriophage capsid. Nat New Biol
1972;239:34–7.
Thieffry D, Huerta AM, P´erez-Rueda E et al. From specific
gene regulation to genomic networks: a global analysis
of transcriptional regulation in Escherichia coli. Bioessays
1998;20:433–40.
Tomoyasu T, Gamer J, Bukau B et al. Escherichia coli FtsH is
a membrane-bound, ATP-dependent protease which degrades
the heat-shock transcription factor sigma 32. EMBO
J 1995;14:2551–60.
Tomoyasu T, Ogura T, Tatsuta T et al. Levels of DnaK and
DnaJ provide tight control of heat shock gene expression
and protein repair in Escherichia coli. Mol Microbiol 1998;30:
567–81.
Tuominen I, Pollari M, Aguirre von Wobeser E et al. Sigma factor
SigC is required for heat acclimation of the cyanobacterium
Synechocystis sp. strain PCC 6803. FEBS Lett 2008;582:
346–50.
Tuominen I, Pollari M, Tyystj¨arvi E et al. The SigB sigma factor
mediates high-temperature responses in the cyanobacterium
Synechocystis sp. PCC6803. FEBS Lett 2006;580:
319–23.
van Bokhorst-van de Veen H, Bongers RS, Wels M et al. Transcriptome
signatures of class I and III stress response deregulation
in Lactobacillus plantarum reveal pleiotropic adaptation.
Microb Cell Fact 2013;12:112.
van der Veen S, Hain T, Wouters JA et al. The heat-shock response
of Listeria monocytogenes comprises genes involved in
heat shock, cell division, cell wall synthesis, and the SOS response.
Microbiology 2007;153:3593–607.
Varmanen P, Ingmer H, Vogensen FK. ctsR of Lactococcus lactis encodes
a negative regulator of clp gene expression. Microbiology
2000;146:1447–55.
Wade JT, Castro Roa D, Grainger DC et al. Extensive functional
overlap between sigma factors in Escherichia coli. Nat Struct
Mol Biol 2006;13:806–14.
Waldminghaus T, Heidrich N, Brantl S et al. FourU: a novel type
of RNA thermometer in Salmonella. Mol Microbiol 2007;65:
413–24.
Waldminghaus T, Gaubig LC, Klinkert B et al. The Escherichia coli
ibpA thermometer is comprised of stable and unstable structural
elements. RNA Biol 2009;6:455–63.
Walsh NP, Alba BM, Bose B et al. OMP peptide signals initiate
the envelope-stress response by activating DegS protease
via relief of inhibition mediated by its PDZ domain. Cell
2003;113:61–71.
Wang Q, Kaguni JM. A novel sigma factor is involved in expression
of the rpoH gene of Escherichia coli. J Bacteriol
1989a;171:4248–53.
Wang QP, Kaguni JM. dnaA protein regulates transcriptions of the
rpoH gene of Escherichia coli. J Biol Chem 1989b;264:7338–44.
Wawrzynow A, Banecki B, Zylicz M. The Clp ATPases define
a novel class of molecular chaperones. Mol Microbiol
1996;21:895–9.
Wetzstein M, Volker U, Dedio J et al. Cloning, sequencing and
molecular analysis of the dnaK locus from Bacillus subtilis. J
Bacteriol 1992;174:3300–10.
Wiegert T, Schumann W. Analysis of a DNA-binding motif of
the Bacillus subtilis HrcA repressor protein. FEMS Microbiol Lett
2003;223:101–6.
Wilson AC, Tan M. Functional analysis of the heat shock regulator
HrcA of Chlamydia trachomatis. J Bacteriol 2002;184:
6566–71.
Wilson AC, Tan M. Stress response gene regulation in Chlamydia
is dependent on HrcA-CIRCE interactions. J Bacteriol
2004;186:3384–91.
Wilson AC, Wu CC, IIIYates JR et al. Chlamydial GroEL autoregulates
its own expression through direct interactions with the
HrcA repressor protein. J Bacteriol 2005;187:7535–42.
Xu Z, Horwich AL, Sigler PB. The crystal structure of the
asymmetric GroEL-GroES-(ADP)7 chaperonin complex.
Nature 1997;388:741–50.
Young DB, Garbe TR. Heat shock proteins and antigens of Mycobacterium
tuberculosis. Infect Immun 1991;59:3086–93.
Yuan G, Wong SL. Regulation of groE expression in Bacillus
subtilis: the involvement of the sA-like promoter and the
roles of the inverted repeat sequence (CIRCE). J Bacteriol
1995a;177:5427–33.
Yuan G, Wong SL. Isolation and characterization of Bacillus
subtilis groE regulatory mutants: evidence for orf39 in
the dnaK operon as a repressor gene in regulating the
expression of both groE and dnaK. J Bacteriol 1995b;177:
6462–8.
Yuewei H, Haley FO, Raengpradub S et al. Transcriptomic and
phenotypic analyses suggest a network between the transcriptional
regulators HrcA and sigmaB in Listeria monocytogenes.
Appl Environ Microb 2007;73:7981–91.
Yura T, Guisbert E, Poritz M et al. Analysis of sigma32 mutants
defective in chaperone-mediated feedback control reveals
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021
574 FEMS Microbiology Reviews, 2017, Vol. 41, No. 4
unexpected complexity of the heat shock response. P Natl
Acad Sci USA 2007;104:17638–43.
Yuzawa H, Nagai H, Mori H et al. Heat induction of sigma 32
synthesis mediated by mRNA secondary structure: a primary
step of the heat shock response in Escherichia coli. Nucleic Acids
Res 1993;21:5449–55.
Zhao K, Liu M, Burgess RR. The global transcriptional response of
Escherichia coli to induced sigma 32 protein involves sigma 32
regulon activation followed by inactivation and degradation
of sigma 32 in vivo. J Biol Chem 2005;280:17758–68.
Zomer A, Fernandez M, Kearney B et al. An interactive regulatory
network controls stress response in Bifidobacterium breve
UCC2003. J Bacteriol 2009;191:7039–49.
Zomer A, van Sinderen D. Intertwinement of stress response
regulons in Bifidobacterium breve UCC2003. Gut Microbes
2010;1:100–2.
Zuber U, Schumann W. CIRCE, a novel heat shock
element involved in regulation of heat shock
operon dnaK of Bacillus subtilis. J Bacteriol 1994;176:
1359–63.
Downloadedfromhttps://academic.oup.com/femsre/article/41/4/549/3574666bygueston14September2021