FEBS Letters 588 (2014) 1961–1966

journal homepage: www.FEBSLetters.org

A cis-encoded sRNA controls the expression of fabH2 in Yersinia

Pei Lu a, Yong Zhang a, Yangbo Hu a, Matthew S. Francis b, Shiyun Chen a,⇑
a
Key Laboratory of Etiology and Biosafety for Emerging Infectious Diseases, Wuhan Institute of Virology, Chinese Academy of Sciences, Wuhan 430071, China
b
Department of Molecular Biology and Umeå Centre for Microbial Research (UCMR), Umeå University, SE-901 87 Umeå, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: YsrH is a novel cis-encoded sRNA located on the opposite strand to fabH2, which is essential for fatty
Received 10 February 2014 acid biosynthesis in bacteria. In this study, YsrH-mediated regulation of fabH2 expression was inves-
Revised 10 March 2014 tigated in Yersinia pseudotuberculosis. Constitutive and inducible over-expression of YsrH decreased
Accepted 3 April 2014
the mRNA level of fabH2, while expression of downstream fabD and fabG remained unaffected. Poly-
Available online 13 April 2014
nucleotide phosphorylase (PNPase) also played an important role in this regulation process by medi-
Edited by Renee Tsolis ating YsrH decay in the exponential phase. Thus, our data defines a cis-encoded sRNA that regulates
fatty acid synthesis via a regulatory mechanism also involving PNPase.
Keywords:
Ó 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
3-oxoacyl-ACP synthase (fabH2)
Yersinia sRNA regulator of fabH2 (YsrH)
cis-Encoded sRNA
PNPase
Yersinia

1. Introduction post-transcriptional regulation of mRNAs via base-pairing interac-

tions [9,10]. These sRNAs can be divided into two categories that
Fatty acids are essential components of cell membranes and are are referred to as cis-encoded and trans-encoded sRNAs [9,10].
also important sources of metabolic energy in all organisms [1]. To date the majority of sRNAs are trans-encoded, and these reg-
Thus, fatty acid biosynthesis and degradation pathways are gener- ulate target mRNAs via imperfect base-pairing [10–12]. Most trans-
ally highly conserved and coordinately regulated to maintain mem- encoded sRNAs repress their target mRNAs through blocking the
brane lipid homeostasis [1]. In bacteria, fatty acid biosynthesis is a ribosome binding site (RBS) to inhibit translation initiation and/
type II, or disassociated-enzyme system, where each of the reac- or reduce target mRNA stability [10,13,14]. Additionally, the RNA
tions of the pathway is catalyzed by a discrete cytoplasmic enzyme chaperone Hfq is often involved in trans-encoded sRNA-mediated
[1,2]. Hence, the pathway consists of a collection of individual pro- regulation in Gram-negative bacteria [10]. In contrast, cis-encoded
teins encoded by unique genes [3]. Studies have revealed that fatty sRNAs are transcribed from the opposite DNA strand to their target
acid synthesis is initiated by b-ketoacyl-ACP synthase III, a product mRNAs [15]. Consequently, they share extended regions of com-
of the fabH gene, which is universally found in both Gram-negative plete complementarity with their targets [15]. The known regula-
and Gram-positive bacteria [3,4]. Furthermore, some chemical tory mechanisms employed by cis-encoded antisense sRNAs
compounds have been shown to inhibit FabH from diverse microor- include transcription attenuation, translation inhibition, inhibition
ganisms, including multi-drug resistant bacterial pathogens [5]. of primer maturation, and promotion or inhibition of mRNA degra-
This suggests that FabH represents an effective molecular target dation [15,16].
for the development of new antimicrobial agents [5]. Duplexes of trans-encoded sRNA-mRNA are often degraded by
During the past decade, it has become evident that small non- endoribonucleases (endoRNases) such as RNase E or RNase III
coding RNAs (sRNAs) serve essential regulatory mechanisms in [13,17]. RNase E is an important endoRNase responsible for the pro-
both eukaryotic and prokaryotic cells. In bacteria, sRNA-mediated cessing and degradation of RNAs. Specifically, its C-terminal region
regulation impacts on a wide range of physiological processes, binds to polynucleotide phosphorylase (PNPase), RNA helicase B
including outer membrane biogenesis [6], sugar stress and metab- and enolase to form the RNA degradosome that is responsible for
olism [7] and pathogenesis [8]. Most identified sRNAs act through degradation of mRNAs [18,19]. RNase E also forms ribonucleopro-
tein complexes with Hfq/sRNAs through its C-terminal scaffold
region, which could actively recruit sRNAs to their target mRNAs
⇑ Corresponding author. Fax: +86 27 87199354.
[13,19]. RNase III is another endoRNase involved both in
E-mail address: sychen@wh.iov.cn (S. Chen).

http://dx.doi.org/10.1016/j.febslet.2014.04.005
0014-5793/Ó 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1962 P. Lu et al. / FEBS Letters 588 (2014) 1961–1966

trans-encoded and cis-encoded sRNA-mediated target gene regula- were constructed by the k Red recombinase one-step inactivation
tion [15,20], although the detailed mechanisms for this have not yet method using pKD4 [27] as template DNA in combination with
been elucidated. the respective primer pairs: PLO-52/PLO-53, PLO-54/PLO-55 and
Several exoribonucleases are also responsible for trans-encoded PLO-48/PLO-49 (Table S2). In strain rne701, the nucleotides
sRNA-mediated target mRNAs instability. For example, PNPase is a 2103–3651 of the rne gene were replaced by the kanamycin resis-
30 –50 exoribonuclease which degrades single-stranded RNA and tance cassette. All mutants were confirmed by a combination of
directly binds to Hfq [19,21]. Although PNPase binds to the PCR and sequencing.
C-terminal scaffold region of RNase E to form RNA degradosome to
induce RNA decay, it can also influence the stability of sRNAs such 2.4. RNA preparation and Northern blot
as MicA and RybB in a degradosome-independent manner [22].
PNPase is also responsible for trans-encoded sRNA turn-over for RNA preparation was performed essentially as previously
which the mechanism is still unclear [23]. described [25], but with some slight modifications. Overnight cul-
Using a cDNA cloning approach, a previous study identified 25 tures of the different YPIII strains were diluted (1:100) in fresh
cis-encoded sRNAs in the plague bacilli Yersinia pestis [24]. Among medium and grown to the indicated cell densities (OD600). Culture
these sRNAs, Yp-sR16 (Yersinia sRNA regulator of fabH2, YsrH) is a aliquots were removed and mixed with 0.2 volume of the stop
cis-encoded sRNA conserved in all Yersinia species [24]. YsrH is 64 solution (95% ethanol, 5% water-saturated phenol) and immedi-
nucleotides in length and is transcribed on the opposite strand ately frozen in liquid nitrogen. RNA was extracted using TRIzol
from fabH2, which encodes b-ketoacyl-acyl carrier protein reagent (Invitrogen).
synthase III that is essential for bacterial fatty acid synthesis [24]. Following denaturation at 95 °C for 5 min followed by a further
In this study, we investigated the regulatory role of YsrH on the 5 min on ice, the RNA samples (10 lg) were loaded onto 7 M urea/
expression of fabH2 in the food-borne pathogen Yersinia 6% polyacrylamide gels and transferred to positively charged Nylon
pseudotuberculosis. membranes (Roche). After UV-crosslinking and pre-hybridization of
the membranes in PerfectHyb™ plus Hybridization Buffer (Sigma)
2. Materials and methods for 30 min at 42 °C, biotinylated oligonucleotide probes were added
and allowed to hybridize at 42 °C overnight. Membranes were then
2.1. Bacterial strains, plasmids, growth media and oligonucleotides washed at 42 °C in three successive steps of 15 min in respective
solutions of 5, 1 and 0.5 SSC containing 0.1% SDS. Hybridized
Bacterial strains and plasmids used in this study are summa- probe was detected by a Biodetect detection kit (Beyotime).
rized in Table S1. Escherichia coli strains were routinely grown in
Luria Bertani (LB) broth at 37 °C. Y. pseudotuberculosis YPIII strains 2.5. Quantitative RT-PCR
were grown at 28 °C in Yersinia–Luria–Bertani (YLB) medium (1%
tryptone, 0.5% yeast extract, 0.5% NaCl). When appropriate, antibi- To obtain appropriate cDNA templates for PCR, 2 lg of RNase-
otics were added at the following concentrations: 100 lg ml1 free DNase I (Promega) treated RNA samples were reverse tran-
ampicillin, 50 lg ml1 kanamycin, 30 lg ml1 chloramphenicol scribed using Random 9 mers (TaKaRa) or the specific ysrH gene
and 15 lg ml1 nalidixic acid. Sequences of all oligonucleotides primer PLO-246. The PCR was then performed with the oligos
used in this study are listed in Table S2. PLO-245/PLO-246 to detect YsrH-specific RNA transcript. The rela-
tive amount of target mRNA was analyzed by quantitative RT-PCR
2.2. Constitutive and inducible expression of YsrH (qRT-PCR) using iTaq™ Universal SYBRÒ Green Supermix (Bio-Rad)
following the manufacturer’s instructions.
To constitutively over-express YsrH, a plasmid named pRO-
YsrH was constructed with an approach described previously 2.6. Statistical analysis
[25]. Briefly, the ysrH gene encompassing both its transcriptional
start site and 30 bp downstream of the transcriptional terminator All results were presented as mean ± standard deviation based
identified by a stretch of T’s was PCR-amplified from YPIII-derived on at least three separate experiments, and the data analyzed using
genomic DNA using primers PLO-83/PLO-84 (Table S2). Similarly, the student’s t-test.
the primers PLO-163/PLO-194 were used to amplify a DNA frag-
ment representing anti-antisense ysrH from its terminator T- 3. Results
stretch to the start site, this plasmid was named pRO-YsrHas. A
control vector named pRO100 was constructed based on the 3.1. Expression pattern of YsrH in Y. pseudotuberculosis
pMD 18-T Vector (TaKaRa) as described previously [25]. The vector
backbone containing the PLlacO promoter (from the position-1), The ysrH gene was identified downstream of the plsX gene on
ampicillin resistance cassette, pUC replicon and a strong rrnB ter- the minus strand of the chromosome in Y. pestis and this sequence
minator was PCR-amplified with primers pRO-514EF/pRO-146R and arrangement is conserved in other Yersinia species, including Y.
(Table S2), digested with EcoRI and self-ligated to obtain pRO100. pseudotuberculosis [24]. To test the expression pattern of YsrH in Y.
The ysrH gene amplified with primers PLO-144/PLO-145 was pseudotuberculosis, RT-PCR was performed with RNA samples iso-
cloned into pBAD22 to give the L-arabinose inducible YsrH con- lated from bacteria in different growth phases. We found that YsrH
struct pBAD-YsrH. L-Arabinose was added at a final concentration is already expressed in the early-exponential growth phase and
of 0.2% (w/v) to induce YsrH expression. this expression is maintained at the same level in later stages of
growth with that in the early-exponential growth phase (picture
2.3. Mutant construction not shown).

The pnp deletion mutant (Dpnp) was constructed as described 3.2. YsrH inhibits fabH2 mRNA level
previously [26]. Primers PLO-39/PLO-40 and PLO-41/PLO-17
(Table S2) were used to amplify the upstream and downstream Since the ysrH gene is located within the coding region of fabH2,
fragments of the pnp gene. The mutants Drnb, Drnr and rne701 deletion of ysrH will also disrupt the expression of fabH2. Hence, to
 P. Lu et al. / FEBS Letters 588 (2014) 1961–1966 1963

determine the role of YsrH in Y. pseudotuberculosis we constructed a establishing an inducible YsrH expression system where ysrH
constitutive over-expression plasmid named pRO-YsrH by cloning was placed under the control of an arabinose-inducible PBAD pro-
the ysrH gene downstream of a constitutive PLlacO promoter. This moter. Since we observed that the parental expression level of
and the control plasmid were respectively transformed into Y. pseu- YsrH was consistent during the growth stages, and fabH2 mRNA
dotuberculosis wild type strain, and the fabH2 mRNA levels in differ- level reached the peak during the early-exponential phase
ent growth phases were analyzed by qRT-PCR. As shown in Fig. 1a, (Fig. 1a), we induced YsrH expression in the early-exponential
fabH2 mRNA levels were highest in the early-exponential phase phase to analyze YsrH-mediated fabH2 mRNA regulation, and the
and then decreased when bacteria were grown to stationary phase. results showed that YsrH expression was strongly induced by arab-
Over-expression of YsrH severely reduced the fabH2 mRNA level by inose (Fig. S1). Pre-induction of YsrH with 0.2% arabinose at early-
10-fold at early-exponential phase and repressed 7-fold and 3- exponential phase for as little as 5 min caused a rapid 34% decrease
fold at late-exponential and stationary stage, respectively (Fig. 1a). of fabH2 mRNA (Fig. 1b). Moreover, a 10 min exposure to YsrH
As a specificity control, we also constructed the pRO-YsrHas resulted in even greater repression that was in the vicinity of
plasmid that expresses under the control of the constitutive PLlacO 75%. In contrast, fabH2 mRNA level was not changed after 10 min
promoter a RNA product with sequence that is complementary to induction in control cells harbouring empty vector (Fig. 1b). Collec-
YsrH. This antisense product is designed to interfere with endoge- tively, our data suggest that YsrH is a sRNA that restricts fabH2
nously produced YsrH, which would be expected to limit its ability mRNA levels by directly duplexing with complimentary sequence
to base-pair to the fabH2 target mRNA. Indeed, when the anti-YsrH in this specific target.
RNA was expressed in the early-exponential phase, the fabH2
mRNA level was even higher than that in the wild type cells 3.3. YsrH has no regulatory effect on fabDG in the fabHDG operon
(Fig. 1a). The most obvious explanation for this is that the inhibi-
tory role of endogenous YsrH is negated by duplex formation fabH is transcribed in the same operon with fabD and fabG in
between YsrH and antisense YsrH. Over-expression of anti-YsrH E. coli and Salmonella [2,28]. Analysis of the YPIII genome indicates
also de-repressed fabH2 mRNA levels in both late-exponential that this operon architecture is retained in Y. pseudotuberculosis
and stationary phase (Fig. 1a). These data indicate that the repres- (data from BIOCYC). To identify whether YsrH also exerts a regula-
sive effect of YsrH is specific. tory role on fabD and fabG in Yersinia, we compared mRNA levels of
To further validate the direct regulation of YsrH on fabH2, we fabD and fabG in strains constitutively over-expressing YsrH or
next used qRT-PCR to examine changes to fabH2 mRNA levels in anti-YsrH as well as the control strain with the empty vector. With
response to short-term exposure to YsrH. This was achieved by respect to fabD mRNA levels, no significant difference at any phase
of growth was observed among the three different strains (Fig. 2a).
Concerning fabG mRNA levels, a small but significant elevation was
observed specifically in the strain over-expressing anti-YsrH and
grown to early exponential phase, but no significant difference was
observed when any of the strains were grown to late-exponential
and stationary phase (Fig. 2b). To further validate these results,
fabD and fabG mRNA levels were analyzed using the arabinose
inducible system. As shown in Fig. 2c, a short burst of YsrH
(5 min) caused an elevation in fabD and fabG mRNA levels whereas
a longer burst (10 min) resulted in significant reductions of fabD
and fabG mRNA levels. Given these alterations, and compared with
the regulation effects on fabH2, we suggest that YsrH has no regu-
latory effect on fabD and fabG in the fabHDG operon. Rather, the
dramatic reduction in fabH2 mRNA levels caused by YsrH (between
3-fold and 10-fold reduction across the growth curve; see
Fig. 1) leaves no doubt that this particular allele is the bona fide
target of YsrH.

3.4. PNPase influences YsrH-mediated fabH2 regulation

RNase E is the major endoRNase involved in sRNA-induced

mRNA decay. In particular, the C-terminal scaffold region com-
plexes with Hfq or other proteins to form the RNA degradosome
to participate in the decay of mRNAs targeted by multiple trans-
encoded sRNAs [18,29]. Since YsrH induced the specific decay of
fabH2 mRNA, the role of RNases in this process was investigated.
The first mutant constructed is deficient in RNA degradosome for-
mation because it harbors the rne701 allele encoding a truncated
RNase E unable to bind Hfq, PNPase, RNA helicase B or enolase
Fig. 1. YsrH represses fabH2 mRNA levels. (a) Quantitative RT-PCR analysis of the [13]. We also constructed three additional mutants lacking the
regulatory effect that constitutive expression of YsrH (pRO-YsrH) or anti-YsrH
ability to produce PNPase (Dpnp), RNase II (Drnb) and RNase R
(antisense of YsrH; pRO-YsrHas) has on fabH2 mRNA levels at different growth
phases. The fabH2 mRNA level in control strain was set 1. (b) Quantitative RT-PCR (Drnr), respectively. In all these strains we then over-expressed
analysis of the regulatory effect that inducible YsrH has on fabH2 mRNA levels. Y. YsrH to investigate whether the RNA degradosome or RNases is
pseudotuberculosis carrying pBAD22 control vector or pBAD-YsrH was induced by involved in YsrH mediated fabH2 regulation. This analysis revealed
0.2% arabinose at an OD600 of 0.5 and samples were collected after cells were two notable observations. Firstly, compared to parental bacteria
induced for 0, 5 and 10 min, respectively. The fabH2 mRNA level obtained at 0 min
in control strain was set 1. All experiments were performed in triplicate, and the
the basal fabH2 mRNA levels in the presence of endogenous YsrH
levels of fabH2 expression in each sample were calculated after normalization to were significantly reduced in the rne701 and Dpnp mutants, but
16S rRNA levels. Error bars represent standard deviation (⁄⁄P < 0.01, ⁄P < 0.05). not in the Drnb and Drnr mutants (Fig. 3a). Secondly, trans
1964 P. Lu et al. / FEBS Letters 588 (2014) 1961–1966

2-fold and 3-fold in most other strains except for the rne701
mutant that showed even less fabH2 mRNA repression (Fig. 3b).
However, since the kinetics of mRNA decay in parent, Drnr and
Drnb bacteria were essentially identical, this argues against our
initial suggestion that RNase II is involved in fabH2 mRNA control
(see Fig. 3a). Despite this, both data sets consistently demonstrated
an essential role for PNPase activity in YsrH-induced fabH2 mRNA
destabilization, and that the RNA degradosome is involved in this
regulatory process.

3.5. PNPase mediates YsrH decay

Since PNPase contributed to YsrH-mediated fabH2 mRNA degra-

dation, we compared the YsrH level in both parental and Dpnp bac-
teria. Analysis by RT-PCR revealed relatively constant expression of
endogenous YsrH in parental bacteria at different phases of
growth, but YsrH was slightly elevated in the Dpnp mutant during
exponential growth phase (picture not shown). Furthermore, com-
bining the constitutive ectopic expression system with Northern
blotting revealed that YsrH accumulated during entry into station-
ary phase, but this was much more pronounced in the Dpnp
mutant (Fig. 4a). To better appreciate the basis for YsrH accumula-
tion in the Dpnp mutant, we performed an YsrH stability assay in
the presence of the transcription inhibitor rifampicin, which was
supplemented to growing bacterial cultures after entry into
early-exponential phase. As shown in Fig. 4b, the turnover of YsrH
after rifampicin treatment of parental bacteria was rapid, with lost
signal occurring within 2 min (Fig. 4b). In contrast, YsrH turnover
was delayed in the Dpnp background (Fig. 4b). Additionally, North-
ern blotting revealed that there was one more upper band in the
Dpnp mutant, which was not observed in the wild type (Fig. 4).
This upper band may be the full-length YsrH, which was unstable
in the presence of PNPase and was processed into smaller products
immediately after transcription. Taken together, these results sug-
gest that PNPase contributes to fabH2 control through specific tar-
geting of YsrH for rapid decay.

4. Discussion

Fig. 2. YsrH has no effect on fabDG mRNA expression. Quantitative RT-PCR analysis The fatty acid biosynthesis pathway has long been targeted for
of the regulatory effect of YsrH on fabD mRNA (a and c) and fabG mRNA (b and c) at the development of new antibacterial agents. For example, the
different growth phases using either the constitutive expression system (a and b) or
products of fabD, fabG and fabH2 have all been successfully used
in the inducible expression system (c). In a and b, total RNA was prepared from
early-exponential phase (OD600 of 0.4), late-exponential phase (OD600 of 0.9), and as targets for the development of potent antibacterial agents [5].
stationary phase (OD600 of 1.1) cells. The fabD and fabG mRNA levels in control FabH is a b-ketoacyl-acyl carrier protein synthase III and is respon-
strain were set 1, respectively. In part c, cells harboring pBAD-YsrH or pBAD control sible for the first reaction in fatty acid biosynthesis – an essential
plasmid were treated with 0.2% arabinose at OD600 of 0.5. The fabD and fabG mRNA enzymatic reaction in bacteria [5,30]. In E. coli and Salmonella, half
level obtained at 0 min in control strain were set 1, respectively. All the
of the fab genes are transcribed in the cluster plsX, fabH, fabD, fabG,
experiments were performed in triplicate, and the levels of fabD and fabG
expression in every sample were calculated after normalization to 16S rRNA levels. acpP, and fabF [28]. It is not yet clear if these genes are transcribed
Error bars represent standard deviation (⁄⁄P < 0.01, ⁄P < 0.05). as a polycistronic unit or as monocistronic units. It follows that the
molecular mechanism governing fabH2 regulation is still elusive.
over-expression of YsrH dramatically reduced fabH2 mRNA in the Interestingly, similar fab gene clusters are reported in other bacte-
Dpnp mutant by 36-fold compared to just 4-fold in parental ria [2], suggesting that conserved regulatory mechanisms are pos-
bacteria (Fig. 3a). These data suggest that the PNPase plays a major sible. In this study, we demonstrate the regulatory role of a cis-
role in YsrH-mediated regulation of fabH2 mRNA. Given that YsrH encoded sRNA, YsrH, on the expression of fabH2.
over-expression also suppressed fabH2 mRNA by 8-fold in the Loss of FabH results in altered fatty acid composition with small-
Drnb background, perhaps RNase II could contribute a minor role colony, reduced cell size and slow growth rate phenotypes in E. coli
in YsrH induced fabH2 mRNA decay. [30]. To investigate whether over-expression of YsrH alters fatty
In parallel, we transformed the plasmid pBAD-YsrH encoding acid biosynthesis and composition in Yersinia, we compared the col-
arabinose inducible YsrH and the control vector pBAD into paren- ony phenotype and growth rate between the control and YsrH over-
tal, rne701, Dpnp, Drnb and Drnr strains and the expression of YsrH expressed strains, but no differences were observed (picture not
in these strains were detected by Northern blotting (Fig. S2). Again shown). This may be explained by the fact that there is complemen-
we used qRT-PCR to compare the fabH2 mRNA levels in the five tation pathway(s) for bacterial fatty acid biosynthesis [30].
backgrounds. As shown in Fig. 3b, fabH2 mRNA was repressed by First identified in an avirulent Y. pestis strain, the ysrH gene
5-fold and 7-fold in the Dpnp background that had expressed transcribes divergently from fabH2 and is completely complemen-
YsrH for 5 and 10 min, respectively. This compared to between tary with the coding region of fabH2 mRNA, and is highly
 P. Lu et al. / FEBS Letters 588 (2014) 1961–1966 1965

Fig. 3. Influence of RNase activity on YsrH-mediated repression of fabH2 expression. Quantitative RT-PCR analysis of down-regulation of fabH2 mRNA upon constitutive
expression of YsrH (a) or induction of YsrH from pBAD-YsrH plasmid (b) in parental, rne701, Dpnp, Drnb and Drnr strains. Cells were grown to OD600 of 0.5, and then either
removed (a) or treated with 0.2% arabinose for 0, 5 or 10 min (b). The levels of fabH2 expression in wild type (WT) and the various RNase mutants were calculated after
normalization to 16S rRNA levels. In panel a, the level of fabH2 mRNA in parental strain and RNase mutants harboring control plasmid was set 1, respectively. In panel b, the
fabH2 mRNA levels in parental strain and RNase mutants harboring the YsrH over-expression vector obtained at 0 min were all set to 100%. The experiments were performed
in triplicate. Error bars represent standard deviation (⁄⁄P < 0.01, ⁄P < 0.05).

Almost all known cis-encoded antisense RNAs from the bacterial

chromosome seem to affect either translation or mRNA stability of
their target, although the mechanistic details of these regulatory
process are seldom well-defined [15]. In E. coli, the cis-encoded
sRNA GadY positively regulates gadX mRNA level, which is
co-transcribed in an operon with gadW [31]. Despite this, the
full-length gadX–gadW polycistronic mRNA transcript does not
accumulate in bacteria when GadY is over-expressed [31]. In this
case, it is thought that the interaction between GadY and the inter-
genic region of the gadX–gadW mRNA results in directed processing
of this fully complementary RNA duplex by the double-strand
RNA-specific endoRNase RNase III [31]. However, whether other
RNases participate in the cis-encoded sRNA mediated regulation
remains unknown. We have now demonstrated that at least one
other RNase – the PNPase – can also mediate decay of cis-encoded
sRNA molecules exemplified by our studies of YsrH. Previously,
PNPase has been described as a protector of some trans-encoded
sRNAs in exponentially growing cells [23]. Moreover, trans-encoded
sRNAs in their Hfq-free state are rapidly degraded by PNPase,
particularly in the stationary phase of growth [20]. Clearly there-
Fig. 4. The effect of PNPase on YsrH sRNA levels. Northern blotting analysis was
used to determine the expression level and the decay rates of the YsrH sRNA in Y. fore, PNPase activity plays an integral role in sRNA regulatory
pseudotuberculosis parent (WT) and Dpnp strains. (a) RNA was extracted at different processes. However, the mechanisms by which PNPase activity
growth phases and Northern hybridization was performed with 50 -biotinylated impacts on cis-encoded and trans-encoded sRNAs must be further
oligonucleotide probes. (b) Wild type or Dpnp strain harboring pRO-YsrH plasmid elucidated in order to better understand these important regulatory
was treated with 500 lg/mL Rifampicin (+Rif), and samples were collected at the
processes.
time points indicated. The experiments were performed in duplicate.
Regarding trans-encoded sRNA-mediated target mRNA
degradation, RNase E also plays a significant role in cleaving
sRNA–mRNA duplexes in a process often involving Hfq [13,17]. In
conserved in Y. pestis and Y. pseudotuberculosis [24]. In our results, this context, RNase E is the core component of a large multimeric
constitutively over-expression of YsrH only has influence on the complex known as the RNA degradosome [18]. Our results demon-
fabH2 mRNA and not mRNA derived from the downstream fabD strate that the C-terminal region of RNase E acting as a scaffold for
or fabG genes. In contrast, inducible expression of YsrH causes RNA degradosome assembly facilitates YsrH-mediated fabH2
fabH2 mRNA reduction, while a short burst of YsrH increases fabD mRNA decay when our arabinose induction experimental set-up
and fabG mRNA, whereas a longer exposure to YsrH reduced fabD was used (see Fig. 3b). This suggests that the RNA degradosome
and fabG mRNA levels significantly (see Figs. 1b and 2c). Due to has a role in YsrH-induced fabH2 mRNA degradation. Given that
the fact that the operon architecture for this biosynthesis pathway PNPase mediates YsrH decay, it is possible that YsrH dependent
has not been elucidated, one possible explanation of our data is regulation on fabH2 mRNA involves a cascade of events.
that fabD and fabG are transcribed from their own promoters or In conclusion, our study is the first to demonstrate that a
are co-transcribed with other downstream genes. Combining the cis-encoded antisense sRNA, YsrH, regulates the fatty acid synthe-
results in Fig. 2, it is possible that YsrH has some indirect regula- sis pathway in bacteria. YsrH operates at the post-transcriptional
tory effects on fabD and fabG, which worth further investigation. level by specifically targeting fabH2 monocistronic mRNA
1966 P. Lu et al. / FEBS Letters 588 (2014) 1961–1966

transcripts for degradation via PNPase- and RNase E-associated [14] Morita, T., Mochizuki, Y. and Aiba, H. (2006) Translational repression is
sufficient for gene silencing by bacterial small noncoding RNAs in the absence
processing pathways.
of mRNA destruction. Proc. Natl. Acad. Sci. USA 103, 4858–4863.
[15] Brantl, S. (2007) Regulatory mechanisms employed by cis-encoded antisense
Acknowledgement RNAs. Curr. Opin. Microbiol. 10, 102–109.
[16] Wen, Y., Feng, J. and Sachs, G. (2013) Helicobacter pylori 5’ureB-sRNA, a cis-
encoded antisense small RNA, negatively regulates ureAB expression by
This research was supported by a Grant from the National transcription termination. J. Bacteriol. 195, 444–452.
Natural Science Foundation of China (No. 31100051). [17] Afonyushkin, T., Vecerek, B., Moll, I., Blasi, U. and Kaberdin, V.R. (2005) Both
RNase E and RNase III control the stability of sodB mRNA upon translational
inhibition by the small regulatory RNA RyhB. Nucleic Acids Res. 33, 1678–
Appendix A. Supplementary data 1689.
[18] Carpousis, A.J. (2007) The RNA degradosome of Escherichia coli: an mRNA-
degrading machine assembled on RNase E. Annu. Rev. Microbiol. 61, 71–87.
Supplementary data associated with this article can be found, in
[19] Morita, T., Maki, K. and Aiba, H. (2005) RNase E-based ribonucleoprotein
the online version, at http://dx.doi.org/10.1016/j.febslet.2014.04. complexes: mechanical basis of mRNA destabilization mediated by bacterial
005. noncoding RNAs. Genes Dev. 19, 2176–2186.
[20] Andrade, J.M., Pobre, V., Matos, A.M. and Arraiano, C.M. (2012) The crucial role
of PNPase in the degradation of small RNAs that are not associated with Hfq.
References RNA 18, 844–855.
[21] Mohanty, B.K., Maples, V.F. and Kushner, S.R. (2004) The Sm-like protein Hfq
[1] Fujita, Y., Matsuoka, H. and Hirooka, K. (2007) Regulation of fatty acid regulates polyadenylation dependent mRNA decay in Escherichia coli. Mol.
metabolism in bacteria. Mol. Microbiol. 66, 829–839. Microbiol. 54, 905–920.
[2] Zhang, Y. and Cronan Jr., J.E. (1998) Transcriptional analysis of essential genes [22] Andrade, J.M. and Arraiano, C.M. (2008) PNPase is a key player in the
of the Escherichia coli fatty acid biosynthesis gene cluster by functional regulation of small RNAs that control the expression of outer membrane
replacement with the analogous Salmonella typhimurium gene cluster. J. proteins. RNA 14, 543–551.
Bacteriol. 180, 3295–3303. [23] De Lay, N. and Gottesman, S. (2011) Role of polynucleotide phosphorylase in
[3] Choi, K.H., Heath, R.J. and Rock, C.O. (2000) Beta-ketoacyl-acyl carrier protein sRNA function in Escherichia coli. RNA 17, 1172–1189.
synthase III (FabH) is a determining factor in branched-chain fatty acid [24] Qu, Y., Bi, L., Ji, X., Deng, Z., Zhang, H., Yan, Y., Wang, M., Li, A., Huang, X., Yang,
biosynthesis. J. Bacteriol. 182, 365–370. R. and Han, Y. (2012) Identification by cDNA cloning of abundant sRNAs in a
[4] Tsay, J.T., Oh, W., Larson, T.J., Jackowski, S. and Rock, C.O. (1992) Isolation and human-avirulent Yersinia pestis strain grown under five different growth
characterization of the beta-ketoacyl-acyl carrier protein synthase III gene conditions. Future Microbiol. 7, 535–547.
(fabH) from Escherichia coli K-12. J. Biol. Chem. 267, 6807–6814. [25] Lu, P., Zhang, Y., Li, L., Hu, Y., Huang, L., Li, Y., Rayner, S. and Chen, S. (2012)
[5] Castillo, Y.P. and Perez, M.A. (2008) Bacterial beta-ketoacyl-acyl carrier protein Small non-coding RNA SraG regulates the operon YPK_1206-1205 in Yersinia
synthase III (FabH): an attractive target for the design of new broad-spectrum pseudotuberculosis. FEMS Microbiol. Lett. 331, 37–43.
antimicrobial agents. Mini Rev. Med. Chem. 8, 36–45. [26] Hu, Y., Lu, P., Wang, Y., Ding, L., Atkinson, S. and Chen, S. (2009) OmpR
[6] Guillier, M., Gottesman, S. and Storz, G. (2006) Modulating the outer positively regulates urease expression to enhance acid survival of Yersinia
membrane with small RNAs. Genes Dev. 20, 2338–2348. pseudotuberculosis. Microbiology 155, 2522–2531.
[7] Gorke, B. and Vogel, J. (2008) Noncoding RNA control of the making and [27] Datsenko, K.A. and Wanner, B.L. (2000) One-step inactivation of chromosomal
breaking of sugars. Genes Dev. 22, 2914–2925. genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97,
[8] Romby, P., Vandenesch, F. and Wagner, E.G. (2006) The role of RNAs in the 6640–6645.
regulation of virulence-gene expression. Curr. Opin. Microbiol. 9, 229–236. [28] Zhang, Y. and Cronan Jr., J.E. (1996) Polar allele duplication for transcriptional
[9] Brantl, S. (2009) Bacterial chromosome-encoded small regulatory RNAs. analysis of consecutive essential genes: application to a cluster of Escherichia
Future Microbiol. 4, 85–103. coli fatty acid biosynthetic genes. J. Bacteriol. 178, 3614–3620.
[10] Waters, L.S. and Storz, G. (2009) Regulatory RNAs in bacteria. Cell 136, 615– [29] Viegas, S.C., Pfeiffer, V., Sittka, A., Silva, I.J., Vogel, J. and Arraiano, C.M. (2007)
628. Characterization of the role of ribonucleases in Salmonella small RNA decay.
[11] Vogel, J. (2009) A rough guide to the non-coding RNA world of Salmonella. Mol. Nucleic Acids Res. 35, 7651–7664.
Microbiol. 71, 1–11. [30] Yao, Z., Davis, R.M., Kishony, R., Kahne, D. and Ruiz, N. (2012) Regulation of cell
[12] Papenfort, K. and Vogel, J. (2010) Regulatory RNA in bacterial pathogens. Cell size in response to nutrient availability by fatty acid biosynthesis in
Host Microbe 8, 116–127. Escherichia coli. Proc. Natl. Acad. Sci. USA 109, E2561–E2568.
[13] Pfeiffer, V., Papenfort, K., Lucchini, S., Hinton, J.C. and Vogel, J. (2009) Coding [31] Opdyke, J.A., Fozo, E.M., Hemm, M.R. and Storz, G. (2011) RNase III participates
sequence targeting by MicC RNA reveals bacterial mRNA silencing in GadY-dependent cleavage of the gadX-gadW mRNA. J. Mol. Biol. 406, 29–43.
downstream of translational initiation. Nat. Struct. Mol. Biol. 16, 840–846.