Microbiology (2012), 158, 1196–1205
DOI 10.1099/mic.0.056309-0
Involvement of pnp in survival of UV radiation in
Escherichia coli K-12
Devashish Rath,3 Suhas H. Mangoli, Amruta R. Pagedar
and Narendra Jawali
Correspondence
Molecular Biology Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India
Narendra Jawali
[email protected]
Received 14 November 2011
Revised
17 January 2012
Accepted 3 February 2012
Polynucleotide phosphorylase (PNPase), a multifunctional protein, is a 39A59 exoribonuclease or
exoDNase in the presence of inorganic phosphate (Pi), and extends a 39-OH of RNA or ssDNA in
the presence of ADP or dADP. In Escherichia coli, PNPase is known to protect against H2O2and mitomycin C-induced damage. Recent reports show that Bacillus subtilis PNPase is required
for repair of H2O2-induced double-strand breaks. Here we show that absence of PNPase makes
E. coli cells sensitive to UV, indicating that PNPase has a role in survival of UV radiation damage.
Analyses of various DNA repair pathways show that in the absence of nucleotide excision repair,
survival of UV radiation depends critically on PNPase function. Consequently, uvrA pnp, uvrB pnp
and uvrC pnp strains show hypersensitivity to UV radiation. Whereas the pnp mutation is nonepistatic to recJ, recQ and recG mutations with respect to the UV-sensitivity phenotype, it is
epistatic to uvrD, recB and ruvA mutations, implicating it in the recombinational repair process.
INTRODUCTION
The survival of bacterial cells following exposure to UV
radiation depends on their ability to repair or remove the
damaged molecules. Irradiation of cells with UVA (320–
400 nm) leads to indirect damage to DNA through sensitizer
radicals and reactive oxygen species (Griffiths et al., 1998).
UVA-excited photosensitizers promote the formation of 8hydroxydeoxyguanosine, hydroxyhydroperoxides and lower
levels of pyrimidine dimers (Griffiths et al., 1998). UVA has
also been implicated in base loss, formation of DNA–protein
cross-links and production of single-strand breaks (Griffiths
et al., 1998). Irradiation at shorter wavelengths (UVB, 290–
320 nm; UVC, 100–290 nm) leads to three major base
modifications: (1) cyclobutane-type pyrimidine dimers, (2)
the (6,4) photoproduct (and corresponding Dewar isomer)
and (3) thymine glycols. Pyrimidine dimers are the most
prevalent (though limited) strand breaks, and DNA–protein
cross-links are also formed (Griffiths et al., 1998).
The majority of these lesions, such as pyrimidine dimers
and 6,4 photoproducts, are repaired by the nucleotide
excision repair (NER) system (Friedberg et al., 2006).
However, replication through these lesions can result in the
3Present address: Department of Cell and Molecular Biology,
Biomedical Center, Uppsala University, Box 596, SE-751 24 Uppsala,
Sweden.
Abbreviations: DSB, double-strand break; Nal, nalidixic acid; NER,
nucleotide excision repair; PNPase, polynucleotide phosphorylase; WT,
wild-type.
Six supplementary figures and two supplementary tables are available
with the online version of this paper.
1196
formation of gaps and breaks that are primarily repaired by
the recombinational repair mechanisms (Kuzminov, 1995).
In Escherichia coli, there are two pathways for recombinational DNA repair: the RecBCD and RecF pathways. The
RecBCD pathway is responsible for repair of double-strand
breaks (DSBs) in DNA, whereas the RecF pathway is largely
responsible for repair of single-strand DNA gaps (Cox, 1998;
Kowalczykowski et al., 1994). Both of these pathways involve
extensive processing of the broken ends before homologous
pairing and joint molecule formation can occur (Handa
et al., 2009). In many bacteria, UV irradiation also induces
the SOS pathway, leading to activation of many genes
involved in DNA repair, including translesion polymerases.
Translesion polymerases can synthesize DNA beyond these
lesions by an error-prone mechanism, leading to UV
mutagenesis (Perry et al., 1985).
Polynucleotide phosphorylase (PNPase), a multifunctional
protein, is widely conserved in eubacteria (Bermúdez-Cruz
et al., 2005). Traditionally, it has been reported to be
involved in RNA metabolism, as it is responsible for Mg2+
and inorganic phosphate (Pi)-dependent 39A59 processive
phosphorolytic degradation of RNA as well as templateindependent polymerization of rNDPs into RNA (Nierlich
& Murakawa, 1996; Coburn et al., 1999; Grunberg-Manago,
1999). It is associated with the RNA degradosome, a multiprotein complex comprising RNase E, enolase and RNA
helicase RhlB (Carpousis, 2002; Py et al., 1994). In E. coli
mutants lacking the main polyadenylating enzyme PAP I,
PNPase is responsible for the residual RNA tailing observed
(Mohanty & Kushner, 2000). Although in vitro analyses have
suggested that PNPase may catalyse template-independent
polymerization of dNDPs into ssDNA in the presence of
Downloaded from www.microbiologyresearch.org by
056309 G 2012 SGM
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
Printed in Great Britain
E. coli pnp and UV repair
either Mn2+ or Fe3+ ions (Gillam & Smith, 1974; Beljanski,
1996), recently there has been growing evidence of the
involvement of PNPase in DNA metabolism (Cardenas
et al., 2009, 2011). Whereas E. coli PNPase (PNPaseEco) copurifies with RecA (Register & Griffith, 1985), Bacillus
subtilis PNPase (PNPaseBsu) has been shown to co-purify
with RecN, a DNA repair protein used in recombinational
repair of DSBs (Cardenas et al., 2009). In vivo analyses in B.
subtilis have revealed that a null pnpA (DpnpA) mutation is
epistatic to DrecN or DykoV mutations, two genes involved
in homologous recombination repair and non-homologous
end joining, respectively (Cardenas et al., 2009). In vitro
PNPaseBsu and PNPaseEco show Mn2+-dependent 39A59
exoDNase activity. In the presence of Mn2+ and dADP, both
of these enzymes catalyse template-independent ssDNA
polymerization on both unstructured oligo-A56 and structured oligo-A ssDNA (Cardenas et al., 2011).
Here we analyse the effect of the absence of PNPase on
survival of E. coli K-12 strains following exposure to UV
radiation. The results show that the absence of PNPase
decreases the survival of cells exposed to UV radiation,
suggesting a role for PNPase in survival of UV radiation in
E. coli. Genetic analysis of various DNA repair pathway
mutants suggests an interaction between pnp and the
recombination repair pathways.
METHODS
Bacterial strains and plasmids. All bacterial strains used were
derivatives of E. coli K-12 and are described in Table 1. Bacterial cultures were grown at 37 uC in Lysogeny broth (LB) and supplemented
with kanamycin (30 mg ml21), where required. pAZ101 (Regonesi
et al., 2004) carrying the wild-type (WT) pnp was a kind gift from Dr
G. Dehò, Università degli Studi di Milano.
Construction of strains. For newly constructed strains, P1-mediated
transduction as described by Miller (1992) was performed, with the
donor and recipient designated P1(donor)6recipient in Table 1.
P1vir grown on C5641 was used to transfer the pnp : : Tn5 allele to
MG1655 with kanamycin selection to construct DR2001. Transfer of
the pnp mutation was followed by analysing transfer of cold sensitivity
and confirmed by reversal to cold resistance in the presence of a
plasmid carrying the pnp gene (Fig. S1, available with the online
version of this paper). Plasmid pCP20 was used to remove the
kanamycin cassette from JW5851, as described by Datsenko &
Wanner (2000), to construct a markerless pnp deletion mutant
DL5851. DL5851 served as a recipient in transductions with P1vir
grown on appropriate Keio collection mutants (Baba et al., 2006) with
kanamycin selection to construct the double mutants (Table 1).
UV and gamma radiation survival assays. Overnight cultures were
diluted in fresh LB medium (1 : 100) and grown at 37 uC in an
incubator shaker to mid-exponential phase (1–26108 c.f.u. ml21).
These cells were then centrifuged at 4000 g for 10 min and
resuspended in an equal volume of 0.9 % NaCl. An aliquot (4.5 ml)
of this suspension was taken in a sterile Petri dish (diameter 5 cm)
and exposed to UV [254 nm, 5 J m22 s21: the dose was calibrated
using a model VLX-3W radiometer (UVI tec)] for different time
periods. Different dilutions were then plated on LB agar plates and
c.f.u. counts were determined after 24 h incubation at 37 uC in the
dark. Appropriate dilutions of control cells not exposed to UV were
http://mic.sgmjournals.org
plated to determine initial counts. For gamma radiation survival
assays, overnight cultures were diluted in fresh LB medium and
grown to mid-exponential phase as above in an incubator shaker.
These cells were then exposed to gamma radiation [Gamma Chamber
5000; Board of Radiation & Isotope Technology (BRIT), DAE,
Mumbai, India; dose 1.5 Gy s21, source cobalt-60] for different time
periods. Different dilutions of exposed and control cells were then
plated on LB agar plates and c.f.u. counts were determined after 24 h
incubation at 37 uC.
Spontaneous mutation frequency assay. Spontaneous mutation
to nalidixic acid (Nal) resistance was used to determine the
spontaneous mutation frequency, as described by Viswanathan &
Lovett (1998). Briefly, mutation to Nal resistance was determined for
eight independent cultures in 2 ml LB medium after overnight
growth. For selection of Nal-resistant cells, the overnight cultures
were washed twice in saline, and subsequently plated onto LB-Nal
plates (20 mg ml21). Colonies were counted after overnight growth at
37 uC. Mutation rates were calculated according to the median
method of Lea & Coulson (1949) using the following formula:
mutation rate5M/N, where M is the calculated number of mutation
events and N is the mean number of viable cells in the culture. M is
solved by interpolation from experimental determination of r0, the
median number of mutant cells determined among the cultures, by
the formula r05M(1.24+lnM).
Recovery of DNA synthesis. Fresh overnight cultures were diluted
1 : 100 and grown in M9 medium to OD600 0.3, at which point half of
the culture received an incident UV dose of 15 J m22, whereas the
other half of the culture was mock-irradiated. At the times indicated,
duplicate 0.1 ml aliquots of culture were pulse-labelled with 1 mCi
[3H]thymidine ml21 (3.76104 Bq ml21) for 2 min at 37 uC. Cells
were spotted on Whatman 3 mm filter discs and air-dried. DNA was
precipitated by treating the discs serially with cold 10 % TCA for
30 min and 5 % TCA for 10 min at room temperature. The discs were
then washed serially with 70 % ethanol and 100 % acetone. After air
drying, the amount of 3H in each sample was determined by liquid
scintillation counting.
Conjugal recombination. AB1157 and its pnp : : Tn5 mutant
derivative DR3001 were used as recipients in conjugal crosses with
HfrC (MD2223) as donor, as described by Miller (1992). Thereafter,
pro+ and leu+ recombinants were scored for in these crosses.
Recovery of the individual markers was expressed as a percentage of
the total recipient numbers.
RESULTS
Mutation in pnp enhances UV sensitivity
To assess whether a mutation in pnp affects UV sensitivity,
MG1655 and its pnp : : Tn5 derivative DR2001 were
exposed to different doses of UV radiation and survival
was analysed. DR2001 was more sensitive than MG1655 to
UV radiation. Overall survival of DR2001 was two- to 20fold lower than MG1655 depending upon the UV dose
employed (Fig. 1), suggesting that a functional pnp gene
was essential for optimum survival after UV exposure. In
the presence of pAZ101, carrying WT pnp, the UV
sensitivity of DR2001 was restored to nearly the same
level as that of MG1655 (Fig. 1). This suggested that the
observed increase in UV sensitivity of DR2001 was indeed
due to the absence of PNPase.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
1197
D. Rath and others
Table 1. List of strains used in this study
Strain
MG1655
C5641
DR2001
AB1157
DR3001
MD2223
W3110
GW11
MG1693
SK5665
BW25113
JW5851
DL5851
JW4019
JW0762
JW1898
JW3786
JW1850
JW3627
JW2860
JW5855
JW2788
DR9001
DR9002
DR9003
DR9004
DR9005
DR9008
DR9009
DR9011
DR1015
JW1644
JW1738
JW2493
JW1993
JW0204
DR9033
DR9034
DR9035
DR9036
Genotype
Source or reference
F l rph-I
C-1a, pnp : : Tn5
MG1655, pnp : : Tn5
F2 thi-1 thr-1 leuB6 D(gpt-proA)62 his4 argE3 ara-14 lacY1 galK2 xyl-5 mtl-1
rpsL31kdg K-51 tsx-33 supE44
AB1157, pnp : : Tn5
HfrC relA1 tonA22 T2rcspC-IS150
F2, l2, IN(rrnD-rrnE)1, rph-1
W3110, zce-726 : : Tn10 rng : : cat
thyA715
thyA715 rne-1 (TS)
rrnB3 DlacZ4787 hsdR514 D(araBAD)567 D(rhaBAD)568 rph-1
BW25113 Dpnp776 : : kan
JW5851 Dkan
BW25113 DuvrA753 : : kan
BW25113 DuvrB751 : : kan
BW25113 DuvrC759 : : kan
BW25113 DuvrD769 : : kan
BW25113 DruvA786 : : kan
BW25113 DrecG756 : : kan
BW25113 DrecJ743 : : kan
BW25113 DrecQ767 : : kan
BW25113 DrecB745 : : kan
Dpnp DuvrA753 : : kan
Dpnp DuvrB751 : : kan
Dpnp DuvrC759 : : kan
Dpnp DuvrD769 : : kan
Dpnp DruvA786 : : kan
Dpnp DrecG756 : : kan
Dpnp DrecJ743 : : kan
Dpnp DrecB745 : : kan
Dpnp DrecQ767 : : kan
Drnt-730 : : kan
DxthA747 : : kan
DxseA758 : : kan
DsbcB780 : : kan
DrnhA733 : : kan
Dpnp DxthA747 : : kan
Dpnp DsbcB780 : : kan
Dpnp DxseA758 : : kan
Dpnp Drnh733 : : kan
E. coli Genetic Stock Centre
Piazza et al. (1996)
P1(C5641)6MG1655 to KmR
E. coli Genetic Stock Centre
2
2
P1(C5641)6AB1157 to KmR
Laboratory collection
Laboratory collection
Umitsuki et al. (2001)
E. coli Genetic Stock Centre
Arraiano et al. (1988)
Baba et al. (2006)
Baba et al. (2006)
This work
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
P1(JW4019)6DL5851 to KmR
P1(JW0762)6DL5851 to KmR
P1(JW1898)6DL5851 to KmR
P1(JW3786)6DL5851 to KmR
P1(JW1850)6DL5851 to KmR
P1(JW3627)6DL5851 to KmR
P1(JW2860)6DL5851 to KmR
P1(JW2788)6DL5851 to KmR
P1(JW5855)6DL5851 to KmR
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
Baba et al. (2006)
P1(JW1738)6DL5851 to KmR
P1(JW1993)6DL5851 to KmR
P1(JW2493)6DL5851 to KmR
P1(JW0204)6DL5851 to KmR
Mutations in other RNases and DNases do not
confer UV sensitivity
BW25113, this indicated that the effect of pnp is independent of the genetic background.
As PNPase is involved in RNA turnover we analysed
whether a general reduction in RNA degradative activity in
the cell results in UV sensitivity. To facilitate genetic analysis, isogenic mutant strains from the Keio collection
(Baba et al., 2006) were used. A markerless Dpnp strain
DL5851 was constructed and utilized as a recipient in
transduction crosses to transfer individual mutations to
construct double mutants. As observed with DR2001,
DL5851 (Dpnp) was found to be more UV-sensitive than
the WT parent BW25113 (Fig. S2). As the absence of pnp
conferred decreased UV resistance to MG1655 as well as
Cells lacking RNaseG (GW11), or RNaseT (Drnt) or
RNaseH (DrnhA), did not show enhanced UV sensitivity
over the WT (data not shown). As the rne null mutant is
unviable, we tested a temperature-sensitive rne mutant
(SK5665), and this also did not show increased UV sensitivity
(data not shown). An increase in activity of RNaseH by
expressing it from a multicopy plasmid has been shown to
make E. coli B/r strains sensitive to UV radiation (Bockrath
et al., 1987). We analysed an rnhA null mutant (JW0204) and
found it to be as UV-sensitive as the WT. The level of UV
sensitivity in an rnhA pnp double mutant was similar to the
1198
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
Microbiology 158
E. coli pnp and UV repair
Fig. 1. Survival of the WT (MG1655), its pnp mutant derivative
(DR2001) and the mutant complemented with pAZ101 (pnp+) of
different doses of UV. Each data point represents the mean±SEM
of at least three replicates.
pnp strain (data not shown), suggesting that RNaseH did not
contribute to UV survival.
In a recent in vitro study, E. coli PNPase has been shown to
degrade ssDNA as well as to process the ends of duplex DNA
with overhangs (Cardenas et al., 2011), suggesting that it has
significant exoDNase activity. Analysis of knockout mutants
for the major exonucleases, namely exonuclease III (Dxth),
large subunit of exonuclease VII (DxseA) or exonuclease I
(DsbcB), did not reveal UV sensitivity in these strains (Fig.
2). Further deletion of pnp in these strains (DxthDpnp,
DxseADpnp and DsbcBDpnp, respectively) failed to decrease
their UV resistance, with the double mutants showing a
similar level of UV sensitivity to the Dpnp strain (Fig. 2).
Together these results suggest that PNPase has a specific
role in UV survival, and that UV sensitivity in the absence
of PNPase is not due to a general depression of RNase or
DNase function in the cell.
pnp-mediated UV sensitivity is independent of the
NER pathway
NER is considered the major repair pathway among the
several pathways implicated in repair of UV damage to
DNA. We analysed the possibility of UV sensitivity of pnp
being mediated through the NER pathway. For this, uvrA,
pnp and uvrA pnp mutants were screened for UV sensitivity
at different doses. As shown in Fig. 3, at 25, 50 and
100 J m22 survival of the DuvrA strain decreased by 250-,
1187- and 6554-fold, respectively, compared with the WT.
At identical doses, the absence of pnp synergistically
increased the sensitivity of the DuvrA strain to 839-, 8719and 85 004-fold, respectively, over the WT (see Table S2 for
D10 values). Similar results were observed for uvrB and uvrC
strains, which showed an increase in UV sensitivity by about
http://mic.sgmjournals.org
Fig. 2. Survival of exonuclease null mutants, in the presence or
absence of PNPase, exposed to UV. Each data point represents
the mean±SEM of at least three separate experiments.
10 000- and 830-fold, respectively, compared with the WT at
100 J m22 (Fig. S3). Further deletion of pnp in DuvrB and
DuvrC strains enhanced their UV sensitivity to 83 333- and
12 738-fold, respectively (Fig. S3). The results indicated that
the decreased UV resistance seen in the Dpnp background
was not mediated through the NER pathway. Surprisingly,
DpnpDuvrD double mutants did not show any change in UV
sensitivity compared with the DuvrD strain (Fig. S4),
suggesting a genetic interaction between uvrD and pnp.
Double pnp uvrD mutants display altered
mismatch repair capacity
UvrD is involved in a number of pathways, including the
NER and mismatch repair pathways. The mutation rate is
highly elevated in strains carrying mutations in genes such
as uvrD, mutH and mutS, which impair mismatch repair
capacity (Horst et al., 1999). Spontaneous development of
resistance against antibiotics such as Nal has been used to
measure spontaneous mutation frequency (Viswanathan &
Lovett, 1998). As genetic analysis indicated an interaction
between uvrD and pnp we used Nal resistance to measure
the mutation frequency in WT, pnp, uvrD and uvrD pnp
strains. The absence of pnp did not significantly alter the
spontaneous mutation frequency in the WT background;
mean values obtained for BW25113 and DL5851 were
5.6661029 and 7.6461029, respectively (Fig. 4). The mutation frequency was elevated 35-fold in the DuvrD strain as
compared with the WT strain (Fig. 4). Contrary to the above,
deletion of pnp in the DuvrD background decreased the
spontaneous mutation frequency to 50.7661029, a roughly
fourfold decrease compared with the DuvrD strain. This result
indicates that loss of UvrD function is partially complemented by the absence of PNPase.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
1199
D. Rath and others
Fig. 3. Survival of the uvrA mutant, in the
presence or absence of PNPase, exposed to UV.
Each data point represents the mean±SEM of at
least three separate experiments. A significant
increase in sensitivity in the double mutant
compared with the corresponding single mutant
parent was observed at all doses (P,0.05;
Table S1).
Genetic analysis indicates that pnp interacts with
the recBCD pathway
UV radiation induces DNA lesions that block the progression
of the replication machinery. Progression of the fork past
UV-induced lesions can lead to formation of gaps and breakages (Kuzminov, 1995). In such situations recombinational
DNA repair mechanisms, namely the RecBCD-dependent
DSB repair pathway and the RecJQFOR-dependent postreplication gap repair pathway, serve to maintain the
genome integrity and allow restoration of DNA replication (Cox, 1998; Kowalczykowski et al., 1994). Irrespective
of the initial pathway utilized for processing of DNA ends,
formation of Holliday junctions is a common step, and
the resolution of these structures is crucial to survival. The
Fig. 4. Spontaneous mutation frequency (SMF) of WT, pnp, uvrD
and uvrD pnp strains. Each data point represents the mean±SEM
of at least three independent experiments. The asterisk marks a
significant decrease in SMF of the uvrD pnp strain compared with
the uvrD strain (P,0.05; Table S1).
1200
junction intermediates are resolved either by RuvABC
helicase–nuclease complex or by RecG helicase (Kuzminov,
1995).
To gain insight into the functional relationship between
pnp and recombination repair, we analysed a series of
mutants in the recombination repair pathways, individually and in combination with pnp deletion, for UV survival.
The results (Fig. 5a) show that deletion of recJ, the product
of which acts in the early step of the RecF pathway,
conferred a low level of UV sensitivity, in agreement with
the results of Courcelle et al. (2006). Absence of pnp led to
a significant increase in UV sensitivity of the DrecJ strain
and similar results were obtained with the DrecQDpnp
double mutant (Fig. S5). In contrast to recJ or recQ, the
DrecB strain showed higher sensitivity to UV. Whereas the
DrecBDpnp double mutant showed numerically slightly
higher sensitivity than DrecB, this was not statistically
significant (Fig. 5b, Table S1). These results suggest a
genetic interaction between pnp and the RecBCD pathway.
The genetic interaction between pnp and resolvase activities
(RuvABC and RecG) was studied by analysing survival of
UV radiation in DruvA, DrecG and the corresponding pnp
double mutants. The DruvA strain showed enhanced UV
sensitivity compared with the WT and the highest difference
in sensitivity was observed at low dose (25 J m22; Fig. 6a).
Additional deletion of pnp in this strain failed to show higher
UV sensitivity, suggesting a genetic interaction between pnp
and ruvA. The DrecG strain showed two- to threefold higher
UV sensitivity than the WT strain up to 50 J m22. In
contrast to DruvA, deletion of pnp in the DrecG strain led to
a sharp decline in survival. A synergistic increase in UV
sensitivity by about 50-fold over the WT was observed
except at a dose of 100 J m22, where a saturation killing
effect was seen (Fig. 6b). Overall these results suggest that
whereas RecG functions independently, PNPase functions in
a pathway common to RecB and RuvA.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
Microbiology 158
E. coli pnp and UV repair
Fig. 5. Survival analysis of (a) RecF and (b) RecBCD pathway
mutants, exposed to UV, in the presence or absence of PNPase.
Each data point represents the mean±SEM of at least three
independent experiments. For statistical analysis see Table S1.
Conjugational recombination is unaffected by
pnp mutation
The post-synaptic function of RuvABC or RecG, i.e.
resolution of recombination intermediates, is crucial for
recovery of recombinants in conjugational crosses (Lloyd,
1991; Lloyd & Buckman, 1991). The recovery of recombinants (leu+ and pro+) was assessed using an HfrC strain as
the donor and AB1157 and its pnp derivative DR3001 as
recipient in conjugation crosses. The recovery of recombinants was 1–2 % of the recipient number for the two
http://mic.sgmjournals.org
Fig. 6. Survival analysis of resolvase mutants, exposed to UV, in
the presence or absence of PNPase: (a) ruvA, (b) recG. Each data
point represents the mean±SEM of at least three independent
experiments.
markers, and no significant difference between the pnp
mutant and the WT was observed.
Absence of pnp does not confer sensitivity to
gamma irradiation
The gamma radiation sensitivity of a pnp mutant (DR2001)
and its parent was assessed by exposing cells to various
doses (Fig. S6). The mutant showed a small increase (1.5to 2.5-fold) in survival compared with the WT parent
depending upon the dose.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
1201
D. Rath and others
Fig. 7. Assay for kinetics of DNA synthesis
following UV irradiation. The cultures were
pulse-labelled with [3H]thymidine for 2 min at
the indicated times following either 15 J m”2
of UV irradiation or mock irradiation. The
relative amount of 3H incorporated is plotted
over time. Cultures were irradiated at time 0.
Each data point represents the mean±SEM of
three experiments. U, unexposed; E, exposed
to UV.
The pnp mutant shows higher replication arrest
events immediately after UV irradiation
DNA lesions formed after irradiation of cells with UV result
in stalling of replication forks and arrest of the replication
machinery (Setlow et al., 1963). Many recombination repair
genes have been implicated in fork processing (Cox, 2001,
and references therein). Mutations in some of these genes
result in altered kinetics of recovery of DNA synthesis
following irradiation with UV (Courcelle et al., 2003, 2006).
As our data indicated an interaction between pnp and
components (recB and ruvA) of recombination repair we
followed the kinetics of nascent DNA synthesis in the WT
and pnp deletion strain. A low dose (15 J m22) was chosen,
as no significant reduction in the survival of WT or pnp cells
occurred at this dose. In the WT cells this UV dose initially
reduced the rate of DNA synthesis by about 90 %. Recovery
of DNA synthesis was observed after about 15–20 min (Fig.
7), consistent with earlier reports (Courcelle et al., 2003,
2006). For the pnp deletion strain a greater decrease in the
inhibition of rate of synthesis was observed up to 10–15 min
after UV exposure, indicating a higher number of replication
arrest events. However, no significant difference in the
kinetics of resumption of DNA synthesis between the WT
cells and the pnp deletion strain was observed (Fig. 7).
DISCUSSION
In spite of the growing recognition of RNases as modulators of various cellular processes, few studies specifically
on the role of RNases in survival of radiation stress
in E. coli have been reported (Brégeon & Sarasin, 2005;
Viswanathan et al., 1999; Bockrath et al., 1987). PNPase,
although initially reported to be involved only in RNA
metabolism, is now shown to be a multi-functional enzyme,
involved in ssDNA degradation as well as 39 DNA end
resection in addition to RNase function (Cardenas et al.,
1202
2011). Here we have shown that the absence of PNPase
enhanced the sensitivity of E. coli cells to UV radiation. To
understand the interaction between pnp and DNA repair
pathways we analysed the UV sensitivity of DNA repair
mutants individually and in combination with the pnp
mutation. If the pnp effect were mediated through a gene/
pathway, combining the pnp mutation with a mutation of
that pathway would not show an additional effect (interacting genes). However, if the double mutant showed
additional or synergistic effects then the pnp effect is
mediated independently of that pathway (non-interacting
genes). Such epistasis analysis has been extensively used to
define interactions or non-interactions at the genetic level
(Mahdi & Lloyd, 1989; Howard-Flanders et al., 1966;
Cardenas et al., 2009; Lloyd et al., 1988; Li & Waters,
1998; Mandal et al., 1993).
Deletion of pnp in strains lacking uvrA, uvrB or uvrC made
the cells hypersensitive to UV, while it had no effect in a
uvrD strain. UvrD, in addition to NER, also functions in
recombination repair (Centore & Sandler, 2007; Lestini &
Michel, 2007) and mismatch repair (Horst et al., 1999;
Matson & Robertson, 2006), and the above results suggested
an interaction of pnp with uvrD in a pathway other than
NER. Our analysis showed that while the spontaneous
mutation frequency increased 35-fold in a uvrD strain it was
not significantly altered in pnp cells, indicating that the
absence of pnp, unlike uvrD, did not affect the mismatch
repair pathway (Fig. 4). However, deletion of pnp in a uvrD
strain resulted in a fourfold decrease in mutation frequency
(Fig. 4). This result shows that a partial complementation
for UvrD activity can be obtained in the absence of PNPase,
implying that pnp may negatively regulate a functional
homologue of uvrD. It is also possible that in the absence of
PNPase the activity of this homologue is masked by the
predominant UvrD activity; however, in the absence of both
PNPase and UvrD, the activity is unmasked. Although the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
Microbiology 158
E. coli pnp and UV repair
above results confirm a genetic interaction between PNPase
and UvrD, because mismatch repair does not play an
important role in UV tolerance it does not explain the lack
of additional UV sensitivity in a uvrD strain upon deletion of
pnp. It is interesting to note that while PNPaseEco co-purifies
with RecA (Register & Griffith, 1985), RecA in turn is known
to interact with UvrD (Lestini & Michel, 2007). RecA in
addition to its essential role in recombinational repair is
known to be involved in translesion synthesis both directly
(Pham et al., 2002) and indirectly through SOS activation of
translesion polymerase (Steinborn, 1979). However, the
possibility of pnp–uvrD–recA interaction and the impact of
this interaction on recombination repair and/or on translesion synthesis needs to be verified experimentally.
There is general agreement that in UV-irradiated cells
single-stranded gaps generated due to incomplete replication are repaired by RecJQFOR-dependent homologous
recombination (Wang & Smith, 1983; Tseng et al., 1994).
As pnp deletion enhances UV sensitivity in recJ and recQ
strains it appears that the effect of pnp on UV survival is
independent of the RecF pathway. On the other hand, pnp
deletion is epistatic with recB deletion as well as ruvA
deletion, suggesting that pnp interacts with the recBCD
pathway. This conclusion is supported by our data, which
show a large increase in UV sensitivity of the pnp recG
double mutant, as it is known that additional mutation of
recG in a RecBCD pathway mutant (e.g. recB recG or ruvA
recG) leads to hypersensitivity to UV (Lloyd, 1991; Lloyd &
Buckman, 1991). While the UV sensitivity of the recG
strain was consistent with earlier reports (Lombardo &
Rosenberg, 2000) the UV sensitivity of the ‘ruv’ strain used
in our study (JW1850) was different from the values
reported for other ‘ruv’ mutants (Le Masson et al., 2008; Li
& Waters, 1998; Zahradka et al., 2002). A literature survey
revealed an up to 100-fold difference in the survival values
reported for ruvA alleles (Le Masson et al., 2008; Lombardo
& Rosenberg, 2000; Li & Waters, 1998; Zahradka et al.,
2002), and our values are in agreement with those reported
by Lombardo & Rosenberg (2000). These variations could
be due to differences in the alleles used as well as
differences in the genetic background.
It has been shown that PNPaseEco is essential for survival in
cells exposed to H2O2 (Wu et al., 2009). H2O2 induces a
wide variety of lesions in DNA, including single-strand
breaks, DSBs, base modifications, abasic sites and sugar
modifications (Friedberg et al., 2006). Consequently, both
base excision repair and recombination repair are required
for H2O2 tolerance (Friedberg et al., 2006). UV radiation at
254 nm predominantly generates pyrimidine dimers and
6,4 photoproducts but can also result in DNA–protein
cross-links (Griffiths et al., 1998). NER is the major repair
pathway for these lesions. However, a replication fork
encountering such lesions leads to gaps and breaks, which
are repaired by recombinational mechanisms (Kuzminov,
1995). A study in B. subtilis showed the involvement of
PNPase in homologous DNA recombination and raised the
possibility of PNPaseBsu directly participating in DSB repair
http://mic.sgmjournals.org
by virtue of its limited 39 to 59 exonuclease activity
(Cardenas et al., 2009). Although the effect of UV-induced
DNA damage was not explored in that study, our results
showing that pnp is epistatic with recB and ruvA further
strengthen this possibility. In vitro analysis showed a
more potent 39 to 59 exonuclease activity associated with
PNPaseEco compared with PNPaseBsu (Cardenas et al.,
2011). A comparison of PNPase function in evolutionarily
distant E. coli and B. subtilis shows many similarities as well
as differences. The two proteins are involved in the homologous recombination pathway (this study; Cardenas et al.,
2009), but show pleiotropic effects with respect to DNAdamaging agents. PNPaseBsu is required for repair of H2O2induced DSB repair but is dispensable for mitomycin
C-induced DSB repair, whereas PNPaseEco is indispensable
for repair of damage by both of these genotoxic agents
(Cardenas et al., 2009). A pnp null strain of B. subtilis is
non-epistatic to addAB (equivalent of recBCD) (Cardenas
et al., 2009). In contrast, the Dpnp strain of E. coli is
epistatic to recB but does not show sensitivity to gamma
radiation (this study). Unlike B. subtilis, which lacks many
single-stranded exonucleases, E. coli harbours multiple
exonucleases. It is possible that degradation of 39 ends by
PNPaseEco contributes to processing of one-ended DSBs,
unlike PNPaseBsu, which is hypothesized to process the
two-ended DSBs (Cardenas et al., 2011). Overall, the
genetic data presented in this paper are in good agreement
with the replication fork reversal model (Michel et al.,
2007), which places RuvAB at very early stages (processing
of ends) and before RecBCD, and with in vitro reconstitution of the end-processing stage by RecJ as an alternative
pathway for DSB repair that is independent of the RecF
pathway (Handa et al., 2009).
Further biochemical characterization of the end-processing
activity of PNPaseEco is required to clarify its role in oneended DSB repair vis-à-vis two-ended DSB repair. It is
known that NER mutants show altered kinetics of nascent
DNA synthesis compared with the WT, and that DNA
synthesis following UV irradiation in such mutants is never
restored to WT levels (Courcelle et al., 2006). Investigating
the kinetics of nascent DNA synthesis in pnp uvrA double
mutants and other repair pathway mutants that show high
UV sensitivity would shed more light on the role of PNPase
in reactivation of arrested replication forks and recovery of
DNA synthesis. Further biochemical assays of the induction and repair of UV-induced DNA lesions in the presence
and absence of PNPase are required to provide a better
understanding of the role of PNPase in UV radiation
tolerance.
ACKNOWLEDGEMENTS
We thank Dr S. K. Apte for his encouragement and support. We are
grateful to Dr G. Dehò for providing strain C5641 and plasmid
pAZ101. Keio collection clones were provided by NBRP (NIG, Japan).
We thank the members of the Molecular Markers and Gene
Expression section, Molecular Biology Division, Bhabha Atomic
Research Centre, for their help.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
1203
D. Rath and others
REFERENCES
Grunberg-Manago, M. (1999). Messenger RNA stability and its role
Arraiano, C. M., Yancey, S. D. & Kushner, S. R. (1988). Stabilization
in control of gene expression in bacteria and phages. Annu Rev Genet
33, 193–227.
of discrete mRNA breakdown products in ams pnp rnb multiple
mutants of Escherichia coli K-12. J Bacteriol 170, 4625–4633.
Handa, N., Morimatsu, K., Lovett, S. T. & Kowalczykowski, S. C.
(2009). Reconstitution of initial steps of dsDNA break repair by the
Baba, T., Ara, T., Hasegawa, M., Takai, Y., Okumura, Y., Baba, M.,
Datsenko, K. A., Tomita, M., Wanner, B. L. & Mori, H. (2006).
RecF pathway of E. coli. Genes Dev 23, 1234–1245.
Construction of Escherichia coli K-12 in-frame, single-gene knockout
mutants: the Keio collection. Mol Syst Biol 2, 2006.0008.
Beljanski, M. (1996). De novo synthesis of DNA-like molecules by
polynucleotide phosphorylase in vitro. J Mol Evol 42, 493–499.
Bermúdez-Cruz, R. M., Ramı́rez, F., Kameyama-Kawabe, L. &
Montañez, C. (2005). Conserved domains in polynucleotide phos-
phorylase among eubacteria. Biochimie 87, 737–745.
Horst, J. P., Wu, T. H. & Marinus, M. G. (1999). Escherichia coli mutator
genes. Trends Microbiol 7, 29–36.
Howard-Flanders, P., Boyce, R. P. & Theriot, L. (1966). Three loci in
Escherichia coli K-12 that control the excision of pyrimidine dimers
and certain other mutagen products from DNA. Genetics 53, 1119–
1136.
Kowalczykowski, S. C., Dixon, D. A., Eggleston, A. K., Lauder, S. D. &
Rehrauer, W. M. (1994). Biochemistry of homologous recombination
Bockrath, R., Wolff, L., Farr, A. & Crouch, R. J. (1987). Amplified
in Escherichia coli. Microbiol Rev 58, 401–465.
RNase H activity in Escherichia coli B/r increases sensitivity to
ultraviolet radiation. Genetics 115, 33–40.
Kuzminov, A. (1995). Collapse and repair of replication forks in
Escherichia coli. Mol Microbiol 16, 373–384.
Brégeon, D. & Sarasin, A. (2005). Hypothetical role of RNA damage
Le Masson, M., Baharoglu, Z. & Michel, B. (2008). ruvA and ruvB
avoidance in preventing human disease. Mutat Res 577, 293–302.
mutants specifically impaired for replication fork reversal. Mol
Microbiol 70, 537–548.
Cardenas, P. P., Carrasco, B., Sanchez, H., Deikus, G., Bechhofer,
D. H. & Alonso, J. C. (2009). Bacillus subtilis polynucleotide
phosphorylase 39-to-59 DNase activity is involved in DNA repair.
Nucleic Acids Res 37, 4157–4169.
Lea, D. E. & Coulson, C. A. (1949). The distribution of the number of
mutants in bacterial populations. J Genet 49, 264–285.
Lestini, R. & Michel, B. (2007). UvrD controls the access of
Cardenas, P. P., Carzaniga, T., Zangrossi, S., Briani, F., GarciaTirado, E., Dehò, G. & Alonso, J. C. (2011). Polynucleotide
recombination proteins to blocked replication forks. EMBO J 26,
3804–3814.
phosphorylase exonuclease and polymerase activities on singlestranded DNA ends are modulated by RecN, SsbA and RecA proteins.
Nucleic Acids Res 39, 9250–9261.
Li, S. & Waters, R. (1998). Escherichia coli strains lacking protein HU
Carpousis, A. J. (2002). The Escherichia coli RNA degradosome:
Lloyd, R. G. (1991). Conjugational recombination in resolvasedeficient ruvC mutants of Escherichia coli K-12 depends on recG.
J Bacteriol 173, 5414–5418.
structure, function and relationship in other ribonucleolytic multienzyme complexes. Biochem Soc Trans 30, 150–155.
are UV sensitive due to a role for HU in homologous recombination.
J Bacteriol 180, 3750–3756.
Centore, R. C. & Sandler, S. J. (2007). UvrD limits the number and
intensities of RecA-green fluorescent protein structures in Escherichia
coli K-12. J Bacteriol 189, 2915–2920.
Lloyd, R. G. & Buckman, C. (1991). Genetic analysis of the recG locus
Coburn, G. A., Miao, X., Briant, D. J. & Mackie, G. A. (1999).
Lloyd, R. G., Porton, M. C. & Buckman, C. (1988). Effect of recF, recJ,
Reconstitution of a minimal RNA degradosome demonstrates
functional coordination between a 39 exonuclease and a DEAD-box
RNA helicase. Genes Dev 13, 2594–2603.
recN, recO and ruv mutations on ultraviolet survival and genetic
recombination in a recD strain of Escherichia coli K12. Mol Gen Genet
212, 317–324.
Courcelle, J., Donaldson, J. R., Chow, K. H. & Courcelle, C. T. (2003).
Lombardo, M. J. & Rosenberg, S. M. (2000). radC102 of Escherichia
DNA damage-induced replication fork regression and processing in
Escherichia coli. Science 299, 1064–1067.
Courcelle, C. T., Chow, K. H., Casey, A. & Courcelle, J. (2006).
Nascent DNA processing by RecJ favors lesion repair over translesion
synthesis at arrested replication forks in Escherichia coli. Proc Natl
Acad Sci U S A 103, 9154–9159.
of Escherichia coli K-12 and of its role in recombination and DNA
repair. J Bacteriol 173, 1004–1011.
coli is an allele of recG. J Bacteriol 182, 6287–6291.
Mahdi, A. A. & Lloyd, R. G. (1989). Identification of the recR locus of
Escherichia coli K-12 and analysis of its role in recombination and
DNA repair. Mol Gen Genet 216, 503–510.
Mandal, T. N., Mahdi, A. A., Sharples, G. J. & Lloyd, R. G. (1993).
in bacteria. Genes Cells 3, 65–78.
Resolution of Holliday intermediates in recombination and DNA
repair: indirect suppression of ruvA, ruvB, and ruvC mutations.
J Bacteriol 175, 4325–4334.
Cox, M. M. (2001). Recombinational DNA repair of damaged replication
Matson, S. W. & Robertson, A. B. (2006). The UvrD helicase and its
forks in Escherichia coli: questions. Annu Rev Genet 35, 53–82.
Datsenko, K. A. & Wanner, B. L. (2000). One-step inactivation of
modulation by the mismatch repair protein MutL. Nucleic Acids Res
34, 4089–4097.
chromosomal genes in Escherichia coli K-12 using PCR products. Proc
Natl Acad Sci U S A 97, 6640–6645.
Michel, B., Boubakri, H., Baharoglu, Z., LeMasson, M. & Lestini, R.
(2007). Recombination proteins and rescue of arrested replication
Friedberg, E. C., Walker, G. C., Siede, W., Wood, R. D., Schultz, R. A.
& Ellenberger, T. (2006). DNA Repair and Mutagenesis, 2nd edn.
forks. DNA Repair (Amst) 6, 967–980.
Cox, M. M. (1998). A broadening view of recombinational DNA repair
Washington, DC: American Society for Microbiology.
Miller, J. H. (1992). A Short Course in Bacterial Genetics. Cold Spring
Harbor, NY: Cold Spring Harbor Laboratory Press.
Gillam, S. & Smith, M. (1974). Enzymatic synthesis of deoxyribo-
Mohanty, B. K. & Kushner, S. R. (2000). Polynucleotide phosphor-
oligonucleotides of defined sequence. Properties of the enzyme.
Nucleic Acids Res 1, 1631–1648.
ylase functions both as a 39A59 exonuclease and a poly(A)
polymerase in Escherichia coli. Proc Natl Acad Sci U S A 97, 11966–
11971.
Griffiths, H. R., Mistry, P., Herbert, K. E. & Lunec, J. (1998). Molecular
and cellular effects of ultraviolet light-induced genotoxicity. Crit Rev
Clin Lab Sci 35, 189–237.
1204
Nierlich, D. P. & Murakawa, G. J. (1996). The decay of bacterial
messenger RNA. Prog Nucleic Acid Res Mol Biol 52, 153–216.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
Microbiology 158
E. coli pnp and UV repair
Perry, K. L., Elledge, S. J., Mitchell, B. B., Marsh, L. & Walker, G. C.
(1985). umuDC and mucAB operons whose products are required for
UV light- and chemical-induced mutagenesis: UmuD, MucA, and
LexA proteins share homology. Proc Natl Acad Sci U S A 82, 4331–
4335.
Pham, P., Seitz, E. M., Saveliev, S., Shen, X., Woodgate, R., Cox,
M. M. & Goodman, M. F. (2002). Two distinct modes of RecA action
are required for DNA polymerase V-catalyzed translesion synthesis.
Proc Natl Acad Sci U S A 99, 11061–11066.
Piazza, F., Zappone, M., Sana, M., Briani, F. & Dehò, G. (1996).
Polynucleotide phosphorylase of Escherichia coli is required for the
establishment of bacteriophage P4 immunity. J Bacteriol 178, 5513–
5521.
Py, B., Causton, H., Mudd, E. A. & Higgins, C. F. (1994). A protein
Steinborn, G. (1979). Uvm mutants of Escherichia coli K12 deficient
in UV mutagenesis. II. Further evidence for a novel function in errorprone repair. Mol Gen Genet 175, 203–208.
Tseng, Y. C., Hung, J. L. & Wang, T. C. V. (1994). Involvement of RecF
pathway recombination genes in postreplication repair in UVirradiated Escherichia coli cells. Mutat Res 315, 1–9.
Umitsuki, G., Wachi, M., Takada, A., Hikichi, T. & Nagai, K. (2001).
Involvement of RNase G in in vivo mRNA metabolism in Escherichia
coli. Genes Cells 6, 403–410.
Viswanathan, M. & Lovett, S. T. (1998). Single-strand DNA-specific
exonucleases in Escherichia coli. Roles in repair and mutation
avoidance. Genetics 149, 7–16.
Viswanathan, M., Lanjuin, A. & Lovett, S. T. (1999). Identification of
complex mediating mRNA degradation in Escherichia coli. Mol
Microbiol 14, 717–729.
RNase T as a high-copy suppressor of the UV sensitivity associated
with single-strand DNA exonuclease deficiency in Escherichia coli.
Genetics 151, 929–934.
Register, J. C., III & Griffith, J. (1985). 10 nm RecA protein filaments
formed in the presence of Mg2+ and ATPcS may contain RNA. Mol
Wang, T. C. & Smith, K. C. (1983). Mechanisms for recF-dependent
Gen Genet 199, 415–420.
and recB-dependent pathways of postreplication repair in UVirradiated Escherichia coli uvrB. J Bacteriol 156, 1093–1098.
Regonesi, M. E., Briani, F., Ghetta, A., Zangrossi, S., Ghisotti, D.,
Tortora, P. & Dehò, G. (2004). A mutation in polynucleotide
Wu, J., Jiang, Z., Liu, M., Gong, X., Wu, S., Burns, C. M. & Li, Z. (2009).
phosphorylase from Escherichia coli impairing RNA binding and
degradosome stability. Nucleic Acids Res 32, 1006–1017.
Setlow, R. B., Swenson, P. A. & Carrier, W. L. (1963). Thymine dimers
and inhibition of DNA synthesis by ultraviolet irradiation of cells.
Science 142, 1464–1466.
Sinha, R. P. & Häder, D. P. (2002). UV-induced DNA damage and
repair: a review. Photochem Photobiol Sci 1, 225–236.
http://mic.sgmjournals.org
Polynucleotide phosphorylase protects Escherichia coli against oxidative stress. Biochemistry 48, 2012–2020.
Zahradka, D., Zahradka, K., Petranović, M., Dermić, D. & BrcićKostić, K. (2002). The RuvABC resolvase is indispensable for recom-
binational repair in sbcB15 mutants of Escherichia coli. J Bacteriol 184,
4141–4147.
Edited by: J. Green
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 05:11:16
1205