Uracil-containing DNA in Drosophila: stability, stage-specific accumulation, and developmental involvement.

Abstract

Base-excision repair and control of nucleotide pools safe-guard against permanent uracil accumulation in DNA relying on two key enzymes: uracil-DNA glycosylase and dUTPase. Lack of the major uracil-DNA glycosylase UNG gene from the fruit fly genome and dUTPase from fruit fly larvae prompted the hypotheses that i) uracil may accumulate in Drosophila genomic DNA where it may be well tolerated, and ii) this accumulation may affect development. Here we show that i) Drosophila melanogaster tolerates high levels of uracil in DNA; ii) such DNA is correctly interpreted in cell culture and embryo; and iii) under physiological spatio-temporal control, DNA from fruit fly larvae, pupae, and imago contain greatly elevated levels of uracil (200-2,000 uracil/million bases, quantified using a novel real-time PCR-based assay). Uracil is accumulated in genomic DNA of larval tissues during larval development, whereas DNA from imaginal tissues contains much less uracil. Upon pupation and metamorphosis, uracil content in DNA is significantly decreased. We propose that the observed developmental pattern of uracil-DNA is due to the lack of the key repair enzyme UNG from the Drosophila genome together with down-regulation of dUTPase in larval tissues. In agreement, we show that dUTPase silencing increases the uracil content in DNA of imaginal tissues and induces strong lethality at the early pupal stages, indicating that tolerance of highly uracil-substituted DNA is also stage-specific. Silencing of dUTPase perturbs the physiological pattern of uracil-DNA accumulation in Drosophila and leads to a strongly lethal phenotype in early pupal stages. These findings suggest a novel role of uracil-containing DNA in Drosophila development and metamorphosis and present a novel example for developmental effects of dUTPase silencing in multicellular eukaryotes. Importantly, we also show lack of the UNG gene in all available genomes of other Holometabola insects, indicating a potentially general tolerance and developmental role of uracil-DNA in this evolutionary clade.

Full text

Page 1

Uracil-ContainingDNAinDrosophila:Stability,Stage-Specific
Accumulation, and Developmental Involvement
Villo ˝ Muha1., Andra ´s Horva ´th1., Ange ´la Be ´ke ´si1, Ma ´ria Puka ´ncsik1, Barbara Hodoscsek1,
Ga ´bor Mere ´nyi1, Gergely Ro ´na1, Ju ´lia Batki1, Istva ´n Kiss2, Ferenc Jankovics2, Pe ´ter Vilmos2,
Miklo ´s Erde ´lyi2, Bea ´ta G. Ve ´rtessy1,3*
1Institute of Enzymology, Research Center for Natural Sciences, Hungarian Academy of Science, Budapest, Hungary, 2Institute of Genetics, Biological Research Centre
(BRC), Hungarian Academy of Science, Szeged, Hungary, 3Department of Applied Biotechnology and Food Sciences, University of Technology and Economics, Budapest,
Hungary
Abstract
Base-excision repair and control of nucleotide pools safe-guard against permanent uracil accumulation in DNA relying on
two key enzymes: uracil–DNA glycosylase and dUTPase. Lack of the major uracil–DNA glycosylase UNG gene from the fruit
fly genome and dUTPase from fruit fly larvae prompted the hypotheses that i) uracil may accumulate in Drosophila genomic
DNA where it may be well tolerated, and ii) this accumulation may affect development. Here we show that i) Drosophila
melanogaster tolerates high levels of uracil in DNA; ii) such DNA is correctly interpreted in cell culture and embryo; and iii)
under physiological spatio-temporal control, DNA from fruit fly larvae, pupae, and imago contain greatly elevated levels of
uracil (200–2,000 uracil/million bases, quantified using a novel real-time PCR–based assay). Uracil is accumulated in genomic
DNA of larval tissues during larval development, whereas DNA from imaginal tissues contains much less uracil. Upon
pupation and metamorphosis, uracil content in DNA is significantly decreased. We propose that the observed
developmental pattern of uracil–DNA is due to the lack of the key repair enzyme UNG from the Drosophila genome
together with down-regulation of dUTPase in larval tissues. In agreement, we show that dUTPase silencing increases the
uracil content in DNA of imaginal tissues and induces strong lethality at the early pupal stages, indicating that tolerance of
highly uracil-substituted DNA is also stage-specific. Silencing of dUTPase perturbs the physiological pattern of uracil–DNA
accumulation in Drosophila and leads to a strongly lethal phenotype in early pupal stages. These findings suggest a novel
role of uracil-containing DNA in Drosophila development and metamorphosis and present a novel example for
developmental effects of dUTPase silencing in multicellular eukaryotes. Importantly, we also show lack of the UNG gene in
all available genomes of other Holometabola insects, indicating a potentially general tolerance and developmental role of
uracil–DNA in this evolutionary clade.
Citation: Muha V, Horva ´th A, Be ´ke ´si A, Puka ´ncsik M, Hodoscsek B, et al. (2012) Uracil-Containing DNA in Drosophila: Stability, Stage-Specific Accumulation, and
Developmental Involvement. PLoS Genet 8(6): e1002738. doi:10.1371/journal.pgen.1002738
Editor: R. Scott Hawley, Stowers Institute for Medical Research, United States of America
Received August 22, 2011; Accepted April 13, 2012; Published June 7, 2012
Copyright: ? 2012 Muha et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the Hungarian Scientific Research Fund (OTKA K68229, NK84008, CK78646, K75774, NKTH-OTKA H07-BEL74200), the
National Innovation Office, the New Hungary Development Plan (TA´MOP-4.2.1/B-09/1/KMR-2010-0002), Howard Hughes Medical Institutes #55005628 and
#55000342, and Alexander von Humboldt-Stiftung, Germany. The funders had no role in study design, data collection and analysis, decision to publish or
preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: vertessy@enzim.hu (BGV)
. These authors contributed equally to this work.
Introduction
In wild-type organisms, notably excepting some rare bacterio-
phages with thymineless DNA genomes [1,2], uracil in DNA is
thought to occur only transiently and at very low frequency (,20/
million bases) as a damage product [3,4]. Efficient base-excision
DNA repair together with fine-tuned control of nucleotide pools
safe-guard against uracil in DNA relying on two key enzymes:
uracil–DNA glycosylase (UDG) [5] and dUTPase [6].
UDG deficiency, in combination with thymidylate synthase
inhibition or depleted dUTPase activity, was reported to lead into
notable uracil accumulation in DNA. However, deficiency in
uracil–DNA glycosylase on its own contributes only slightly to the
genomic uracil content [7–10]. Since dUTPase deficiency or
silencing can be rescued by depleted activity of UDG [11–14], it
can be argued that UDG is a major factor that renders uracil–
DNA, formed in absence of dUTPase, intolerable for cells. Uracil–
DNA glycosylases represent a superfamily that involves enzymes
with specialized functions. Among the superfamily members,
UNG is reported to be the most abundant one that also possesses
the highest activity in removing uracils from any context of both
single-stranded and double-stranded DNA [5]. SMUG has similar
attributes to UNG but prefers uracil-containing ssDNA as
substrate [15]. MBD4 and TDG recognize mismatched uracil or
thymine bases that base-pair with guanine [16]. The latter enzyme
is known to function within CpG islands, where thymine is formed
after spontaneous methyl-cytosine deamination [17].
Through hydrolyzing dUTP to dUMP and pyrophosphate,
dUTPase serves a dual role in cell physiology: it produces a
precursor for thymidylate synthesis, and removes dUTP from the
deoxynucleotide pool. Eukaryotic and bacterial DNA polymerases
incorporate deoxyuridine into DNA with a rate that depends on
PLoS Genetics | www.plosgenetics.org1 June 2012 | Volume 8 | Issue 6 | e1002738

Page 2

the cellular dUTP concentration [6,10]. Therefore dUTPase has a
significant indirect impact in prevention of uracil incorporation
into DNA. Lack of dUTPase is supposed to induce thymine to
uracil substitution that does not alter the genetic code; however
UDG still removes this kind of base modification supposedly
resulting in genome instability. This process may serve as an
explanation for overall lethality of dUTPase deficiency [6].
Here, we propose that a condition similar to simultaneous
deficiency in both dUTPase and UDG is present in Drosophila
larvae. The Drosophila genome does not encode the major uracil–
DNA glycosylase UNG [18], therefore uracil–DNA tolerance may
be expected. Drosophila larvae contain both proliferating primordial
tissues of imago and differentiated tissues that degrade during
metamorphosis. Our previous study reported Drosophila dUTPase
[19,20] expression only in the first one of the two kinds of tissues
[21,22]. We wished to investigate if the stage- and tissue-specific
expression of dUTPase is translated into differences in the uracil
content of genomic DNA and whether this putative pattern has
developmental significance.
We now report high resistance of Drosophila cell lines to
fluorodeoxyuridine (FdUR) that induced high uracil–DNA levels.
We show that in addition to tolerance of uracil-substituted DNA,
Drosophila cells correctly interpret this unusual DNA both in vitro and
in vivo. Moreover, inwild type Drosophilawe observe stage- and tissue-
specific changes in uracil-content of DNA that are inversely
correlated to dUTPase expression and cellular fate. We propose
that the observed pattern is due to the lack of the ung gene from the
Drosophila genome paralleled with developmental down-regulation of
dUTPase in larval tissues. To check whether the absence of
dUTPase may be causative for uracil accumulation in DNA, we
show that silencing of dUTPase in larvae expands the uracil–DNA
pattern to imaginal tissues, as well. Interestingly, dUTPase silencing
results in abnormal DNA strand breaks, cell death and develop-
mental arrest in early pupal stage. This may indicate that tolerance
of highly uracil-containing DNA is also stage-specific, and other
tolerance factors, in addition to the lack of UNG, may appear in this
specific developmental phase. To our knowledge, our study is the
first one that reports uracil–DNA appearance in absence of
dUTPase in a wild type complex eukaryotic organism and
demonstrates the developmentalpattern of itstoleranceand stability.
Results/Discussion
Since the Drosophila genome lacks UNG, we wished to test
whether Drosophila cells show similar characteristics to UNG
deficient organisms. The drug 59-fluorodeoxyuridine (FdUR),
frequently used as an inhibitor of thymidylate biosynthesis [23,24],
leads to perturbation of nucleotide levels and cell death. Loss of
uracil–DNA glycosylase activity was found to lead to fluoropyr-
imidine resistance in E. coli [25], yeast [26] and C. elegans [27].
Deficient uracil–DNA glycosylase activity was also reported to be
required for increased genomic uracil content after FdUR
exposure in E. coli and mammalian cells [7–10]. Therefore,
response to FdUR treatment may be an indicator for testing
cellular uracil–DNA glycosylase activity. We observed that the
Drosophila S2 cell line shows only small decrease in viability in the
presence of 1 mM of FdUR, while HeLa cells, possessing the ung
gene, show strong lethality at this drug concentration (Figure 1A).
As a dose-dependent response to increasing concentrations of
FdUR, uracil accumulation in genomic DNA of Drosophila S2 cells
became highly elevated (up to approx. 450 uracil/million bases)
(Figure 1B). Both the observed relatively high resistance for FdUR
and the cellular response of genomic uracil incorporation may be
explained by the fact that Drosophila cells lack significant uracil–
DNA glycosylase activity.
We then asked if uracil-containing chemically unusual DNA
may be tolerated and interpreted in Drosophila cells. Uracil–DNA
specific cell response was provoked by transfecting cells with
exogenous plasmid uracil–DNA. The expression signal of the
fluorescent reporters (eYFP or dsRedMonomer) encoded by
plasmids produced in wild type (normal plasmid) or dut-1, ung-1
(uracil-plasmid, approx. 5500 uracil/million bases [8,10]) E. coli
were followed. The gene span for the eYFP or DSRed-monomer
reporters included the promoter (CMV immediate early or
metallothionein), the reporter protein gene (DSRed-monomer or
eYFP), and the SV40 polyadenylation signal. The total length of
these gene spans comprises 1515 or 1299 nucleotides. If
Figure 1. Tolerance and stability of uracil-containing DNA in D.
melanogaster. (A) Dose-response curve upon FdUR treatment followed
by Alamar Blue assay. (B) FdUR leads to uracil accumulation in DNA of
Drosophila S2 cells. Data indicate that increased level of uracil is well-
tolerated in Drosophila, but not in human cells. Data are presented as
mean 6 s.e.m. (C) Drosophila S2 (top panels) and human HeLa cells
(bottom panels) were transfected with normal plasmid (left panels) or
uracil-containing plasmid (right panels). Expression of YFP in Drosophila
S2 cells or dsRedMonomer in HeLa cells indicates stability of the DNA.
(D) Microinjection of uracil-plasmid into Drosophila embryo. Non-
injected embryos served as control sample.
doi:10.1371/journal.pgen.1002738.g001
Author Summary
The usual paradigm confines ‘‘normal’’ DNA of living cells
to a well-defined restricted chemical space populated with
only four bases (adenine, thymine, guanine, and cytosine)
and some of their methylated derivatives (e.g. 59-methyl-
cytosine). Uracil is not considered to be a ‘‘normal’’ DNA
base, except in several bacteriophages. On the contrary,
uracil is generally considered to be an error in DNA. We
show that Drosophila cells interpret uracil-substituted DNA
as normal DNA, due to lack of two repair enzymes.
Importantly, this unusual trait is under developmental
control and applies only for animals before pupation.
Metamorphosis is drastically perturbed by silencing of
dUTPase, responsible for keeping uracil out of DNA. Our
results argue that in Drosophila, and perhaps in other
Holometabola insects as well, uracil–DNA plays a dedicat-
ed physiological role.
Role of dUTPase in Drosophila Development
PLoS Genetics | www.plosgenetics.org2 June 2012 | Volume 8 | Issue 6 | e1002738

Page 3

synthesized in the double mutant dut-1, ung-1 E.coli strain, the
average number of uracil substitutions on a single strand within
these gene spans corresponds to 8.3 and 7.1, respectively. Human
(HeLa) cells, possessing UNG, showed appreciable fluorescent
signal only when transformed with normal plasmid, indicating that
uracil-containing plasmid was not interpreted probably due to its
degradation. On the contrary, Drosophila S2 cells could express
reporter genes encoded either on normal or uracil containing
plasmid at comparable level (Figure 1C and Figure S1).
Having established that Drosophila cells in in vitro culture may
tolerate and do correctly interpret uracil-containing DNA, we
wished to assess the fruit fly physiological response to uracil–DNA
at the organism level. Therefore similarly to the cell culture
reporter-assay, uracil–DNA plasmid was introduced to Drosophila
embryo. Upon microinjection of pP{Gal4VP16} expression vector
(where the 1000 nt segment includes the P-element promoter, the
Gal4-VP16 fusion gene and the hsp70 terminator that may have
an average number of 5.5 uracil substitutions if produced in
double mutant E. coli) into P{mw+UASeGFP} transgenic animals,
the strong eGFP signal indicated the stability of exogenous DNA
(Figure 1D). In pre-hatching embryos eGFP signal was detected
from both normal and uracil-containing DNA with commeasur-
able intensities. In both cases the expression pattern of the reporter
construct is similar and eGFP is most pronounced in the gut.
These results can be explained only by assuming that the genetic
information stored in uracil-containing DNA serves as a cognate
code for transcription in Drosophila cells. Such ability of the fruitfly
cells is most probably due to lack of UNG under which condition
uracil–DNA does not get rapidly degraded.
The other key factor responsible for keeping uracil out of DNA
is the enzyme dUTPase, the importance of which is even more
substantiated in D. melanogaster that lacks UNG. During develop-
ment of D. melanogaster, high dUTPase protein levels can be
observed only in embryonic stages (Figure 2A). As shown by
Western blotting and immunohistochemistry, starting from late
embryonic phase, dUTPase expression level is decreased drasti-
cally and remains low during postembryonic stages. Under normal
physiological circumstances of larval development, dUTPase is
predominantly expressed in imaginal discs, the central nervous
system and in the testis (Figure 2) but not in most larval tissues, like
salivary gland and gut. In the ventriculus and in the salivary gland,
some cells are associated with dUTPase expression – these seem to
correspond to the imaginal cells (Figure 2B). In the imago,
dUTPase is present in the ovary of females. Cellular localization of
dUTPase is usually nuclear [21,28–31], however cytoplasmic
occurrence is also evident in the nurse cells of mature follicles
within ovaries, as described previously [21]. The present results
are in agreement with our more limited earlier observations at the
protein level [21]. We also quantified the dUTPase mRNA level
by real-time PCR and found that protein and mRNA levels show
similar tendencies during development (Figure 3A and 3B).
We were interested to learn whether coincidence of uracil–DNA
tolerance and dUTPase down-regulation results in uracil accu-
mulation in the genome of affected developmental stages and
larval tissues. Determination of genomic uracil level was carried
out by applying a recent quantitative real-time PCR-based method
[10]. Data presented in Figure S2 and Figure 3C argue that DNA
purified from embryo is relatively uracil-free, whereas in larval
stages, uracil becomes much accumulated. The presence of uracil–
DNA in larvae was further confirmed by using an independent
method, the UNG-ARP assay (Figure S3) [8]. Uracil content also
varies according to tissue type within the larvae: the salivary gland,
a representative tissue of pupal degradation, accumulates high
levels of uracil; while imaginal discs, which do not undergo
abundant metamorphosis-coupled cell death, contain much less
uracil (Figure 3D). Uracil level in larval tissues is comparable to
that previously reported in the double mutant dut1-1, ung-1 E. coli
strain [8,10], amounting to several thousands of uracils per million
bases. The above discussed pattern of uracil accumulation in
different stages and tissues shows a clear negative correlation to the
expression pattern of dUTPase (compare the respective develop-
mental stages in Figure 2 and Figure 3).
In order to investigate if perturbation of the wild-type pattern of
uracil levels in DNA may interfere with normal development, we
Figure 2. Stage- and tissue-specific distribution of dUTPase protein levels in D. melanogaster. Western blotting (A) and
immunohistochemistry (B) was performed on selected developmental stages and tissues. Embryo 0–6 h (E1), embryo 0–24 h (E2), 1stlarvae (1L),
2ndlarvae (2L), early 3rdlarvae (3L1), wandering 3rdlarvae (3L2), pupae before head eversion (P1), pupae after head eversion (P2) and pupae 50–60 h
after puparium formation (P3). For Western blotting, actin was used as loading control. Note that dUTPase protein levels are down-regulated during
larval stages and expression is confined to specific tissues.
doi:10.1371/journal.pgen.1002738.g002
Role of dUTPase in Drosophila Development
PLoS Genetics | www.plosgenetics.org3 June 2012 | Volume 8 | Issue 6 | e1002738

Page 4

aimed to silence dUTPase in transgenic D. melanogaster strains
(Table S1). Efficient silencing could be successfully achieved in a
setup resulting in well distinguishable F1 phenotypes: non-silenced
animals were characterized by GFP expression and curly wings,
whereas silenced animals had no markers [32] (Figure S4). Overall
silencing resulted in efficiently decreased dUTPase protein levels
in larvae and pupae (Figure 4A). RNAi silencing may not operate
appropriately in early embryo due to maternal effects, but this
stage is out of the scope of our present experiment with transgenic
strains. We observed that dUTPase silencing did not perturb
normal life and development of larvae. The silenced versus non-
silenced larvae of F1 generation were selected based on GFP
expression, and the time interval between the egg laying and
puparium formation did not show any significant alteration
(Figure 4B). Importantly, effective silencing of dUTPase in
imaginal discs and larval brain did not cause any observable
morphological changes in tissue morphology (Figure 5A). At pupal
stage, however, dUTPase silencing induced 100% lethality, i.e. no
silenced animals could develop into imago (observation is based on
counting 2350 curly winged control imagos resulting from the first
generation of crossings) (Figure 4C).
One possible proof for RNAi specificity is a rescue by the
corresponding complementary DNA [33,34]. Therefore, to test
the specificity of dUTPase RNAi, the RNAi construct was co-
expressed with a dUTPase transgene that led to the expression of
dUTPase protein, as detected in the rescued animals (cf. method
described in Text S1 and shown on Figure S4C and S4D). We
observed a full rescue of the lethality caused by the RNAi
confirming that the RNAi phenotype was due to a reduction of
dUTPase level (Table S2). Rescued animals underwent develop-
ment and metamorphosis just like the wild type animals. In the
larval, prepupal and imago stages, rescued animals were carefully
investigated for morphology and no distinct traits were observed.
In the detailed phenotype analysis of dUTPase-silenced animals,
morphologic observations indicated serious adverse effects and
developmental arrest at or before the pupal stage P5 [35]
(Figure 5B, Figures S5 and S6). Upon removing the puparium
of the silenced animal, defects were identified in the inner texture
of everted discs, and head sack, as well as in the development of
adult eye. These defects are permanent, and are not simply due to
slow down of normal development. Moreover, 3–4 days after
puparium formation, tissues showing morphological traits associ-
ated with the larval stage were identified within dissected samples
(Figure S7 versus Figure S8). Noteworthy, the typical organ
structures of the wild type adult (legs, wings, Malpighian tubules,
Yellow Body, eyes etc.) never appeared in the silenced animals.
Darkened tissues that may result from cell death, necrosis or
histolysis were also observed e.g. in the prothoracic region and
organs (Figure 5B and Figure S8) [36]. This result argued that
although dUTPase is dispensable in larval tissues [21], the
presence of the enzyme at the developmental stage of early
puparium formation is essential for normal development during
metamorphosis.
Tissue specific silencing in the dorsal wing surface of imaginal
discs was also developed as described in Figure S9 (cf. [37]). The
great majority of silenced animals developed curly wings, and a
significant portion of them even showed blisters on the wings.
Curling and blistering of the wings was suggested to result from
Figure 4. Silencing of dUTPase in Drosophila larvae and pupae.
Western blots in (A) show that the protein level of dUTPase is under
detection limit in silenced animals. Actin served as loading control. (B)
Curves show the relative number of silenced and non-silenced animals
that have undergone puparium formation at the given time point after
egg deposition. Inflection points of the curves represent the mean time
of puparium formation characteristic for the given population. dUTPase
silencing did not perturb the time interval required for puparium
formation. (C) Graph shows the number of counted dead animals
relative to number of hatched curly winged control flies. Among these
dead animals, three groups with distinct morphological traits charac-
teristic for wandering larvae (w3L), prepupae (preP), and pupal stage P5
(P5) were identified and counted. (D) Genomic uracil content of
dUTPase silenced and control tissues from 3rdlarvae.
doi:10.1371/journal.pgen.1002738.g004
Figure 3. D. melanogaster genomic DNA uracil content inversely
correlates with dUTPase expression. (A) Changes of dUTPase
mRNA level throughout fruitfly development: embryo (E), 1stlarvae (L1),
2ndlarvae (L2), late 3rdlarvae (L3) and pupae (P). Note that dUTPase is
down-regulated in larvae. (B) Comparison of dUTPase RNA level in the
larval tissues salivary gland and imaginal tissue. Data are presented as
mean of triplicates 6 s.e.m. mRNA level was measured by RT-qPCR and
dUTPase mRNA level was normalized to Rp49 mRNA level. (C) Uracil
content of D. melanogaster genome in different developmental stages:
embryo (E), 1stlarvae (L1), 2ndlarvae (L2), late 3rdlarvae (L3) and pupae
(P). Embryonic sample was used as reference since it was shown to
contain undetectable levels of uracil in DNA (cf. Figure S2). (D)
Comparison of genomic uracil content in wild type imaginal disc and
salivary gland of 3rdlarvae. Data are presented as mean 6 s.e.m.
doi:10.1371/journal.pgen.1002738.g003
Role of dUTPase in Drosophila Development
PLoS Genetics | www.plosgenetics.org4 June 2012 | Volume 8 | Issue 6 | e1002738

Page 5

abnormal cell death in dorsal wing surface [38,39]. These results
suggest that dUTPase silencing within a well-defined tissue
segment may be specifically and exclusively associated with
phenotypic effects in the very same tissue segment.
To analyze whether dUTPase silencing changed the distribution
of uracil–DNA within different larval tissues, we assayed genomic
uracil content of imaginal discs from both non-silenced and silenced
larvae of F1 generation. As Figure 4D shows, imaginal discs from
dUTPase silenced larvae accumulated a high amount of uracil in
their genome (1097+/255 uracil/million bases, to be compared
with 131+/269 uracil/million bases in the non-silenced animals)
that approximated the genomic uracil content of salivary gland.
Further increase in uracil–DNA content of salivary gland DNA was
not observed in dUTPase silenced larvae, indicating that silencing
was effective in tissues that normally express dUTPase but had no
effect on uracil–DNA in tissues that normally do not express
dUTPase. This result confirms that uracil appearance in DNA
depends on dUTPase expression, thus dUTPase activity is causative
in maintaining DNA with low levels of uracil.
To analyze if the effects of dUTPase silencing may lead to DNA
damage or cell death, we applied TUNEL and phospho-Histone
H2Av assays [40,41]. We observed that imaginal discs isolated
from 3rdstage wandering larva of dUTPase-silenced animals show
strong enrichment in TUNEL positive cells (Figure 6A and 6B).
TUNEL staining in the tissues indicate primarily cell death in the
phase of DNA fragmentation. To address the question whether
dUTPase depletion violates genome integrity we stained nuclei for
phospho-Histone H2Av. H2Av histone modification by the ATR/
ATM kinases indicates DNA double strand breaks (DSBs) in the
proximity of the foci [41]. We observed numerous phospho-H2Av
foci in dUTPase silenced wing imaginal discs, while no such foci
were visible in the wild type (Figure 6C). These results suggest a
potential correlation between dUTPase activity and DNA
integrity. The increased level of DNA damage observed in our
experiments in the dUTPase-silenced animals may result from
excessive processing of uracil-containing DNA that concludes to
DNA fragmentation.
Uracil–DNA measurements provided direct evidence that larval
tissues of D. melanogaster that undergo developmental degradation
accumulate uracil in genomic DNA. Up to date, presence of highly
uracil-substituted genomic DNA within wild type organisms was
reported only in some viruses; e.g. in bacteriophages [1,2] and
recently in HIV. To our knowledge, the present study may present
a first example for naturally occurring permanent existence of
uracil–DNA in a free-living complex eukaryotic organism (cf.
Figure 3C and 3D). This condition requires developmental down-
regulation of dUTPase and absence of UNG in the case of D.
melanogaster. Involvement of dUTPase in determining genomic
uracil content by regulating dUTP levels was confirmed by
dUTPase silencing that resulted in appearance of uracil–DNA in
imaginal disc tissues (cf. Figure 4D). Silencing of dUTPase resulted
in developmental defects and DNA strand breaks (cf. Figure 5 and
Figure 6). We also reported that uracil–DNA is tolerated and
interpreted at least from embryonic to 3rdlarval stages.
The extraordinary situation of tolerance and stability of uracil–
DNA may not be exclusively present in D. melanogaster as absence
of ung is ubiquitous among Holometabola (Table 1). As uracil–
DNA naturally occurs in larval tissues that are sentenced to death,
we consider that uracil–DNA might be linked to metamorphosis
and tissue degradation. Further investigations should be taken to
describe the mechanism, its impact and its putative role.
Pupal lethality was observed in flies affected by mutations or
silencing of purine biosynthesis enzymes [42,43]. These animals
showed apoptosis in developing imaginal primordium during
metamorphosis that were also observable as darkened tissues. In
these studies, deficient deoxynucleotide biosynthesis may have
resulted in imbalanced dNTP levels and increased ratio of
mismatches in DNA [44]. Overrepresented DNA modifications
like cytosine methylation also result in pupal lethality [45,46]. In
this case increased methylation pattern of DNA inhibits transcrip-
tional activity and cell cycle progression.
We can suggest two hypotheses to explain pupal intolerance of
uracil–DNA. First, similarly to the case of hypermethylated DNA,
uracil–DNA may show a decreased response and interaction with
transcriptional regulators, activators or other morphogenetic
factors required specifically during pupal metamorphosis. Second,
one or more factor(s), functional only in the pupal stage, may
process uracil–DNA resulting in genome instability and defects in
cell cycle progression or cell death [47,48]. Beyond UNG that is
missing from Drosophila, other uracil–DNA glycosylases would be
suspect for this role. In agreement with this suggestion, expression
patterns of other uracil–DNA glycosylases (SMUG and Thd1) and
AP endonucleases (Rrp1, RpS3 and RpLP0) indicate relative
Figure 5. Morphological consequences of dUTPase silencing. In
larvae (A) and pupae (B). (A) Immunohistochemistry of wing and eye
discs, and brain of non-silenced (control) and silenced larvae for
dUTPase (red) and DAPI staining for DNA (blue) demonstrate on one
hand highly effective silencing; and on the other hand no observable
morphological changes within these tissues. (B) Wild type pupae
(control) in stage P6 (cf. Figure S5) and dUTPase silenced pupae at
corresponding time after puparium formation in dorsal and ventral view
are shown, after puparium removal. Wild type traits, Malpighian tubules
(white arrows), Yellow Body (white asterix), developing adult eye (white
arrowheads) are not observable on silenced animals. Instead, darkened
(apoptotic/necrotic or melanized) tissues (red arrowheads) can be
visualized on these pupae. Note the basically different inner texture of
the everted discs (white boxes) and head sack (white circles).
doi:10.1371/journal.pgen.1002738.g005
Role of dUTPase in Drosophila Development
PLoS Genetics | www.plosgenetics.org5 June 2012 | Volume 8 | Issue 6 | e1002738

Page 6

upregulation during metamorphosis after their lowest expression
in larval stages according to microarray data (Table S3) [49].
Developmental downregulation of these base-excision DNA repair
pathway (BER) enzymes may contribute to uracil–DNA appear-
ance and tolerance in larval stages, and their upregulation may
initiate uracil–DNA processing, destabilization or degradation.
The pupal uracil–DNA intolerance pathways hypothesized above
are intended to be analyzed in further experiments.
Taken together, we suggest three different factors as being
responsible for the stage-specific elevated level of uracil in larval
DNA: 1) lack of ung gene, 2) absence of dUTPase and 3) decreased
levels of enzymes involved in uracil removal. We also conclude
that dUTPase is essential for the full completion of the Drosophila
life cycle, although its absence may be tolerated in the larval
stages. Although fruit flies, and perhaps other Holometabola
insects, are unique in possessing a developmental period when
uracil–DNA can be tolerated, preservation and transmission of
genetic information for several generations still requires dUTPase.
Materials and Methods
Uracil–DNA stability assays
For the uracil–DNA stability assays, pRm-eYFP, pDsRedMono-
mer-N1, pP{Gal4VP16} (kind gift from La ´szlo ´ Sipos) were amplified
in E.coli K12 XL1Blue strain and CJ236 dut-1, ung-1 strain (NEB).
Plasmids were purified with Plasmid Midi Kit (QIAGEN). Uracil
contentoftheplasmidswascheckedwithUDGandAPendonuclease
treatment followed by standard agarose gel electrophoresis [48].
Figure 6. dUTPase silencing results in cell death and DNA strand breaks in larval imaginal discs. (A) Imaginal discs were isolated from
wild type and dUTPase silenced wandering 3rdlarvae and stained for TUNEL assay (shown as red dots). Discs from silenced animals showed highly
increased TUNEL staining. (B) TUNEL positive cell counts in imaginal discs from wild type and dUTPase silenced wandering 3rdlarvae. Error bars
represent the standard error of mean. (C) Imaginal discs from wild type and dUTPase silenced 3rdwandering larvae stained against phospho-H2Av
foci (white dots, some of these are appointed by white arrowheads). dUTPase depleted discs showed several nuclei with phospho-H2Av foci
indicating DNA damage. Scale bar represents 50 mm.
doi:10.1371/journal.pgen.1002738.g006
Table 1. Occurrence of genes encoding dUTPase and UNG in different insects.
Insecta HolometabolaDipteradUTPase UNG
333
Drosophila melanogaster (fruit fly)
+
+
+
+
+
+
+
+
+
+
2
333
Culexpipiens(mosquito)
2
333
Aedesaegypti(mosquito, yellow fever)
2*
333
Anopheles gambiae(mosquito, malaria)
2
33
Bombyxmori(silkmoth)
2
33
Triboliumcastaneum(red flour beetle)
2
33
Apismellifera(honey bee)
2
33
Nasoniavitripennis(parasitoid wasp)
2
3
Acyrthosiphonpisum(pea aphid)
2
3
Pediculushumanuscorporis(body louse)
+
The gene for dUTPase is ubiquitous, but the gene of the major uracil–DNA glycosylase, ung is not encoded in the genome of Holometabola species.
*In the genome of Aedes aegypti strain Liverpool, an unexpected ung sequence was found, showing very high (87%–94%) similarity to the ung gene of
Comamonadaceae, a family of Proteobacteria, arguing for its bacterial origin.
doi:10.1371/journal.pgen.1002738.t001
Role of dUTPase in Drosophila Development
PLoS Genetics | www.plosgenetics.org6 June 2012 | Volume 8 | Issue 6 | e1002738

Page 7

Uracil-containing plasmid stability in cell culture.
fection of pRm-eYFP-N-C* into Drosophila S2 cells was carried out
in the presence of Cellfectine (Invitrogen) following the manufac-
turer’s instructions and expression was induced from the metallo-
thionein promoter at 25uC by addition of 700 mM CuSO4and
overnight incubation. pDsRedMonomer-N1 was transfected into
HeLa cells by using Lipofectamine (Invitrogen). Samples were
visualized with a Leica DMLS fluorescence microscope 48 hours
after transfection.
Uracil-containingplasmid
embryos.
pP{Gal4VP16} plasmids were injected into 0–
30 min pP{mw+UASeGFP} transgenic Drosophila embryos (kind
gift of La ´szlo ´ Sipos). For each experiment, app. 40 dechorionated
embryos were microinjected. They were aligned on a glass
coverslip, dried for 30 min, than covered with 10S Voltalef oil
before injection. Plasmid concentration was adjusted to 1 mg/ml,
by dilution in standard injection buffer. GFP signal was detected in
pre-hatching embryos, 22 h after injection. Embryos without
injection served as a negative control.
Trans-
stability in Drosophila
Cell viability assay
Drosophila Schneider 2 (S2) cells and human HeLa cells were
cultured in 96 well plates at 56104cells/well or 26103cells/well,
respectively. The culture media used were Serum Free Insect
Medium (Sigma, S3777) supplemented with 1% penicillin–
streptomycin solution for S2 cells; and DMEM-F12 (Sigma,
D8437) supplemented with 10% FBS and 1% penicillin–strepto-
mycin for HeLa cells. FdUR (Sigma) was added at a final
concentration range of 0.1–1000 mM. After 96 hours in culture,
cell viability was quantified by Alamar Blue assay (BioSource). The
experiment was repeated in triplicates.
Quantitative measurement of the uracil content in DNA
samples
In order to quantify uracil content of DNA, a real-time
quantitative PCR-based assay was used [10]. Genomic DNA was
isolated and digested with NheI. DNA fragments of 4–5 kb were
purified from gel. Real-time PCR was performed on Mx3000P
qPCR System (Agilent Technologies) using EvaGreen dye
(Biotium) and PfuTurbo Hotstart DNA polymerase and PfuTurbo
Cx Hotstart DNA polymerase (Stratagene). A segment with 963
base length defined by the primers (puBSd-Fw 59-TCGGGAT-
GACTTTTGGGTTCTG-39 and puBSd985R 59-CGCGGTT-
TAACACAGCGTCGG-39) is amplified during the PCR reaction.
Two-fold dilution series were prepared from DNA samples. Three
or more parallel measurements were performed.
dUTPase silencing by RNA interference and rescue of RNAi
UAS-IR stocks were obtained from Vienna Drosophila RNAi
Centre (VDRC) [32]. Strain #21883 and #21884 was used for
dUTPase silencing. Rescue was performed by co-expression of the
RNAi construct with a dUTPase transgene. For details see Text
S1, Tables S1 and S2, Figures S4 and S9.
Immunohistochemistry and Western blot analysis
Western blot analysis of larval organs and stage specific
dUTPase expression was performed as described previously [21].
For immunohistochemistry, applied primary antibodies were anti-
dUTPase (1:10000) or mouse anti-phospho-Histone H2A.X
(Ser139) (Millipore) (1:250). The latter was shown to recognize
Drosophila phospho-Histone H2Av [50]. Tissues were fixed in
50% n-Heptane and 50% PEM-formaldehyde (100 mM PIPES,
1 mM MgCl2, 1 mM EGTA, 2.5% Tween-20, 4% PFA,
pH=6.9) for 30 minutes, with vigorous shacking at room
temperature (RT). Samples were washed with inactivating buffer
(50 mM TRIS, 150 mM NaCl, 0.5% Tween-20, pH=7.4).
Blocking was performed in the following: 5% goat serum, 1.5%
BSA, 0.1% Tween-20, 1% Triton-X 100, 0,001% NaN3, in PBS,
pH=7.4 for 4 hours at RT. Tissues were incubated in primary
antibody diluted in blocking buffer (1:10000), at 4uC for 16 h.
Samples were further washed with blocking buffer for 8 h at RT.
Secondary antibody was applied to visualize dUTPase staining
(Alexa 543 conjugated anti rabbit IgG, Molecular Probes) in
blocking buffer (1:1000) for 2 h, RT. DAPI was used for DNA
staining. After further washing steps, samples were mounted in
FluorSave (Calbiochem). Images were either acquired with a Zeiss
LSCM 710 or a Leica DM IL LED Fluo microscope equipped
with a Leica DFC345 FX monochrome camera.
TUNEL assay
TUNEL assay was carried out as described in [40]. After careful
dissection in PBS, imaginal discs were fixed (0.1M PIPES,
pH=6.9, 1 mM EDTA, 1% Triton X-100, 2 mM MgSO4, 1%
formaldehyde) for 30 min at RT, washed for times with PBS
buffer also containing 0.1% Triton X-100 (PBT) and two times
with PBT 5X (PBS, 0.5% Triton X-100) (10 min each), and
transferred into permeabilization solution (0.1M sodium citrate in
PBT) for 30 min at 65uC. Discs were washed twice with PBT 5X,
three times in PBT and incubated in reaction buffer (30 mM Tris-
HCl, pH=7.2, 140 mM sodium cacodylate, 1 mM cobalt
chloride) for 30 min RT. Reaction was carried out in reaction
buffer containing 0.2 unit/microL terminal deoxynucleotidyl
transferase (NEB) and 5 microM Cy5-dUTP (GE Healthcare)
for 1 hour RT. The reaction was stopped with PBT 5X and
washed three times with PBT and finally with PBS. Nuclei were
stained by Hoechst. Discs were imaged by laser scanning confocal
microscopy (Zeiss LSCM 710). The number of TUNEL positive
cells was determined by ImageJ (Rasband, W.S., ImageJ, National
Institutes of Health, Bethesda, Maryland, USA, http://rsb.info.
nih.gov/ij/, 1997–2004.)
Supporting Information
Figure S1
normal (T pl.) or uracil-substituted plasmids (U pl.). (A) Drosophila
S2 cells, (B) HeLa cells. The number of observed fluorescent cells is
also presented within the bars together with the total number of
scored cells (shown in brackets).
(PDF)
Percentage of fluorescent cells upon transfection with
Figure S2
limit. Uracil content of Drosophila embryonic genome compared to
that of DNA plasmid purified from wild-type E. coli. Both of the
samples showed a value under the detection limit.
(PDF)
Genomic uracil content of embryo is under detection
Figure S3
uracil–DNA in Drosophila larvae. For negative and positive
controls, genomic DNA samples from XL1 Blue and CJ236 ung-
1, dut-1 E.coli strains were used respectively. CJ236 ung-1, dut-1
E.coli strain produces DNA with high uracil content (approx. 5500
uracil/million bases [8,10]).
(PDF)
Ung-ARP assay. UNG-ARP assay shows presence of
Figure S4
Drosophila larvae and pupae and for rescue of dUTPase RNAi.
Crossing schemes are shown on panel A and C: Act-Gal4 means
Gal4 gene coupled with actin 5C promoter that result in
ubiquitous and constitutive expression of yeast transcription
Scheme of crossing for silencing of dUTPase in
Role of dUTPase in Drosophila Development
PLoS Genetics | www.plosgenetics.org7 June 2012 | Volume 8 | Issue 6 | e1002738

Page 8

factor, Gal4 in transgenic D. melanogaster driving transcription of
silencing element (IR) following the UAS promoter. F1 generation
has two genotypes: Act-Gal4/UAS-IR animals express dsRNA for
dUTPase silencing, and have no markers; in UAS-IR/CyO, GFP
animals, the silencing element is not activate, curly wing (CyO)
and GFP markers expressed. Silenced and non-silenced animals
are distinguishable at larvae/pupae and imago stages on the basis
of GFP (panel B) and CyO markers, respectively. Crossing scheme
for silencing is shown on panel C: UAS-dUTPase-FLAG stands
for the rescue construct. Two relevant categories of the F1
generation can be unambiguously distinguished based on the
phenotype of the marker mutations of the CyO, SM6b, and TM3
balancer chromosomes. The TM3 phenotype marks the gene
silenced progenies, while the rescued animals show noTM3
phenotype. Panel D shows Western blot for dUTPase in silenced
versus rescued animals. Note the absence of dUTPase protein in
silenced animals (silencing alleles 21883 and 21884), whereas the
presence of dUTPase proteins in the rescued animals (rescuing
alleles DMDUT20 and DMDUT29). Equivalent total protein
loading was verified by developing the blot also against tubulin
using anti-Tubulin (E7, provided by M. Klymkowsky; Develop-
mental Studies Hybridoma Bank, University of Iowa, Iowa city,
IA).
(PDF)
Figure S5
arrow shows the stage P5 (around 12–14 h after puparium
formation) until lethality due to dUTPase silencing appear.
(PDF)
Summary of pupal developmental processes. Red
Figure S6
Drosophila pupae. Wild type (upper panels) and dUTPase silenced
(bottom panels) pupae were compared in stages P4, P5–6, P6–7,
and P9. Every panel shows four views of the same pupa: dorsal
(upper two) and ventral (bottom two) with and without its
puparium. Specific differences appear at or before P5: Malpighian
tubules (arrows) and Yellow Body (asterices) never appears in
dUTPase silenced pupae.
(PDF)
Developmental arrest caused by dUTPase silencing in
Figure S7
puparium formation. Wild type pupa was dissected at stage P11
where adult organs have already developed (A). Dissected
Malpighian tubules (arrows on B) and Yellow Body (asterices on
B) of wild type pupa these organs have never identified within
dUTPase silenced pupae.
(PDF)
Wild type structures of pharate adults 3 days after
Figure S8
puparium formation. Three days after puparium formation,
dissected tissues of silenced pupae still preserve larval traits: testis
is oval (A), foregut and gastric caeca show larval characteristics (B,
D, asterices), Malpighian tubules (B, arrows) are thin characteristic
for larval ones, and brain (C, white arrowhead) also preserves the
basic structure of larval one. Darkened tissues may have resulted
from necrosis, apoptosis or melanisation [36].
(PDF)
Larval traits in dissected silenced pupae 3 days after
Figure S9
dorsal compartment of Drosophila wing imaginal discs. Crossing
scheme is shown on panel (A): virgin females of the MS1096 Gal4
Scheme of crossing for silencing of dUTPase in the
enhancer trap line expressing Gal4 preferentially in the dorsal
compartment of the wing and carrying UAS-Dicer2 in homozy-
gous form on the second chromosome (Bloomington stock
No. 25706) were crossed to males carrying the Gal4 inducible
silencing element (UAS-IR) in homozygous form on the second
chromosome. The silencing element was activated by the MS1096
driver [37] in female progenies only while F1 males served as an
internal negative control where no silencing occurred. Silenced
females exhibited dorsally curled wing phenotype (panel B) often
with blisters. The penetrance of the phenotype was around 85%.
About 35% of the silenced female progeny also showed blistering
wings (panel C). Male progenies had no wing phenotype. Panel D
shows the expression pattern of the MS1096 driver in the dorsal
compartment of the wing disc visualized by crossing MS1096
females to UAS-MoesinCherry [51] males (panel D) (red
fluorescent staining in the wing disc). MoesinCherry overexpress-
ing female progeny had no wing phenotype.
(PDF)
Table S1
RNAi stocks.
(PDF)
Genomic position of UAS-IR constructs in dUTPase
Table S2
phenotype. Table shows the results of the rescue crosses. UAS-
IR/SM6b; UAS-dUTPase-FLAG/TM3 males were crossed to
Act-Gal4/CyO females (Figure S4). Two UAS-IR (21883 and
21884) and two transgenic rescue lines (DMDUT20 and
DMDUT29) were combined. Number of progenies of the relevant
F1 categories is shown. Gene silencing was complete since no
UAS-IR/Act-GAl4; TM3/+ adult progeny was observed. How-
ever, when the dUTPase transgene was present, rescued animals
survived to adulthood.
(PDF)
dUTPase transgene rescues the dUTPase RNAi
Table S3
array data available on FlyBase were used. Table shows mRNA
level for genes involved in different DNA repair pathways,
elements of uracil–DNA repair are highlighted on grey back-
ground. Q indicates mRNA level decrease, q mRNA level
increase, < no stage-specific change. Note that the overall base
excision repair is down-regulated during larval development, but
other DNA repair processes are not.
(PDF)
Uracil–DNA repair is perturbed in Drosophila. Micro-
Text S1
Materials and Methods and Supplementary References.
(DOC)
Supplementary information. Includes Supplementary
Acknowledgments
We thank Prof. Tibor Vellai and A´gnes Rego ˝s for kind help in using the
Zeiss stereomicroscope and La ´szlo ´ Sipos for kind help in embryo
microinjections.
Author Contributions
Conceived and designed the experiments: VM AH AB MP GR JB PV ME
BGV. Performed the experiments: VM AH AB MP BH JB IK GM GR FJ
PV ME. Analyzed the data: VM AH AB JB MP GM GR JB IK FJ PV ME
BGV. Wrote the paper: VM AH AB MP GR PV ME BGV.
References
1.Duncan BK, Warner HR (1977) Metabolism of uracil-containing DNA:
degradation of bacteriophage PBS2 DNA in Bacillus subtilis. J Virol 22:
835–838.
2.Kiljunen S, Hakala K, Pinta E, Huttunen S, Pluta P, et al. (2005) Yersiniophage
phiR1-37 is a tailed bacteriophage having a 270 kb DNA genome with
thymidine replaced by deoxyuridine. Microbiology 151: 4093–4102.
Role of dUTPase in Drosophila Development
PLoS Genetics | www.plosgenetics.org8 June 2012 | Volume 8 | Issue 6 | e1002738

Page 9

3. Lindahl T (1993) Instability and decay of the primary structure of DNA [see
comments]. Nature 362: 709–715.
Krokan HE, Drablos F, Slupphaug G (2002) Uracil in DNA–occurrence,
consequences and repair. Oncogene 21: 8935–8948.
Visnes T, Doseth B, Pettersen HS, Hagen L, Sousa MM, et al. (2008) Review.
Uracil in DNA and its processing by different DNA glycosylases. Philos Trans R
Soc Lond B Biol Sci.
Vertessy BG, Toth J (2009) Keeping uracil out of DNA: physiological role,
structure and catalytic mechanism of dUTPases. Acc Chem Res 42: 97–106.
Andersen S, Heine T, Sneve R, Konig I, Krokan HE, et al. (2005) Incorporation
of dUMP into DNA is a major source of spontaneous DNA damage, while
excision of uracil is not required for cytotoxicity of fluoropyrimidines in mouse
embryonic fibroblasts. Carcinogenesis 26: 547–555.
Lari SU, Chen CY, Vertessy BG, Morre J, Bennett SE (2006) Quantitative
determination of uracil residues in Escherichia coli DNA: Contribution of ung,
dug, and dut genes to uracil avoidance. DNA Repair (Amst) 5: 1407–1420.
Luo Y, Walla M, Wyatt MD (2008) Uracil incorporation into genomic DNA
does not predict toxicity caused by chemotherapeutic inhibition of thymidylate
synthase. DNA Repair (Amst) 7: 162–169.
10. Horvath A, Vertessy BG (2010) A one-step method for quantitative
determination of uracil in DNA by real-time PCR. Nucleic Acids Res 38: e196.
11. el-Hajj HH, Wang L, Weiss B (1992) Multiple mutant of Escherichia coli
synthesizing virtually thymineless DNA during limited growth. J Bacteriol 174:
4450–4456.
12. Gadsden MH, McIntosh EM, Game JC, Wilson PJ, Haynes RH (1993) dUtp
pyrophosphatase is an essential enzyme in Saccharomyces cerevisiae. Embo J
12: 4425–4431.
13. Dengg M, Garcia-Muse T, Gill SG, Ashcroft N, Boulton SJ, et al. (2006)
Abrogation of the CLK-2 checkpoint leads to tolerance to base-excision repair
intermediates. EMBO Rep 7: 1046–1051.
14. Guillet M, Van Der Kemp PA, Boiteux S (2006) dUTPase activity is critical to
maintain genetic stability in Saccharomyces cerevisiae. Nucleic Acids Res 34:
2056–2066.
15. Pettersen HS, Sundheim O, Gilljam KM, Slupphaug G, Krokan HE, et al.
(2007) Uracil–DNA glycosylases SMUG1 and UNG2 coordinate the initial steps
of base excision repair by distinct mechanisms. Nucleic Acids Res 35:
3879–3892.
16. Hardeland U, Bentele M, Jiricny J, Schar P (2003) The versatile thymine DNA-
glycosylase: a comparative characterization of the human, Drosophila and
fission yeast orthologs. Nucleic Acids Res 31: 2261–2271.
17. Sousa MM, Krokan HE, Slupphaug G (2007) DNA-uracil and human
pathology. Mol Aspects Med 28: 276–306.
18. Adams MD, Celniker SE, Holt RA, Evans CA, Gocayne JD, et al. (2000) The
genome sequence of Drosophila melanogaster. Science 287: 2185–2195.
19. Takacs E, Barabas O, Petoukhov MV, Svergun DI, Vertessy BG (2009)
Molecular shape and prominent role of beta-strand swapping in organization of
dUTPase oligomers. FEBS Lett 583: 865–871.
20. Fiser A, Vertessy BG (2000) Altered Subunit Communication in Subfamilies of
Trimeric dUTPases. Biochem Biophys Res Commun 279: 534–542.
21. Bekesi A, Zagyva I, Hunyadi-Gulyas E, Pongracz V, Kovari J, et al. (2004)
Developmental Regulation of dUTPase in Drosophila melanogaster. J Biol
Chem 279: 22362–22370.
22. Kovari J, Barabas O, Takacs E, Bekesi A, Dubrovay Z, et al. (2004) Altered
active site flexibility and a structural metal-binding site in eukaryotic dUTPase:
kinetic characterization, folding, and crystallographic studies of the homo-
trimeric Drosophila enzyme. J Biol Chem 279: 17932–17944.
23. Rich TA, Shepard RC, Mosley ST (2004) Four decades of continuing
innovation with fluorouracil: current and future approaches to fluorouracil
chemoradiation therapy. J Clin Oncol 22: 2214–2232.
24. Merenyi G, Kovari J, Toth J, Takacs E, Zagyva I, et al. (2011) Cellular response
to efficient dUTPase RNAi silencing in stable HeLa cell lines perturbs
expression levels of genes involved in thymidylate metabolism. Nucleosides
Nucleotides Nucleic Acids 30: 369–390.
25. Wang Z, Mosbaugh DW (1988) Uracil–DNA glycosylase inhibitor of
bacteriophage PBS2: cloning and effects of expression of the inhibitor gene in
Escherichia coli. J Bacteriol 170: 1082–1091.
26. Seiple L, Jaruga P, Dizdaroglu M, Stivers JT (2006) Linking uracil base excision
repair and 5-fluorouracil toxicity in yeast. Nucleic Acids Res 34: 140–151.
27. Kumar S, Aninat C, Michaux G, Morel F (2010) Anticancer drug 5-fluorouracil
induces reproductive and developmental defects in Caenorhabditis elegans.
Reprod Toxicol 29: 415–420.
4.
5.
6.
7.
8.
9.
28. Tinkelenberg BA, Fazzone W, Lynch FJ, Ladner RD (2003) Identification of
sequence determinants of human nuclear dUTPase isoform localization. Exp
Cell Res 287: 39–46.
29. Bozoky Z, Rona G, Klement E, Medzihradszky KF, Merenyi G, et al. (2011)
Calpain-catalyzed proteolysis of human dUTPase specifically removes the
nuclear localization signal peptide. PLoS ONE 6: e19546. doi:10.1371/
journal.pone.0019546.
30. Merenyi G, Konya E, Vertessy BG (2010) Drosophila proteins involved in
metabolism of uracil–DNA possess different types of nuclear localization signals.
FEBS J 277: 2142–2156.
31. Muha V, Zagyva I, Venkei Z, Szabad J, Vertessy BG (2009) Nuclear localization
signal-dependent and -independent movements of Drosophila melanogaster
dUTPase isoforms during nuclear cleavage. Biochem Biophys Res Commun
381: 271–275.
32. Dietzl G, Chen D, Schnorrer F, Su KC, Barinova Y, et al. (2007) A genome-
wide transgenic RNAi library for conditional gene inactivation in Drosophila.
Nature 448: 151–156.
33. Hsu YC, Chern JJ, Cai Y, Liu M, Choi KW (2007) Drosophila TCTP is
essential for growth and proliferation through regulation of dRheb GTPase.
Nature 445: 785–788.
34. Lopez TV, Lappin TR, Maxwell P, Shi Z, Lopez-Marure R, et al. (2011)
Autocrine/paracrine erythropoietin signalling promotes JAK/STAT-dependent
proliferation of human cervical cancer cells. Int J Cancer 129: 2566–2576.
35. Bainbridge SP, Bownes M (1981) Staging the metamorphosis of Drosophila
melanogaster. J Embryol Exp Morphol 66: 57–80.
36. Minakhina S, Steward R (2006) Melanotic mutants in Drosophila: pathways and
phenotypes. Genetics 174: 253–263.
37. Capdevila J, Guerrero I (1994) Targeted expression of the signaling molecule
decapentaplegic induces pattern duplications and growth alterations in
Drosophila wings. EMBO J 13: 4459–4468.
38. Morris EJ, Michaud WA, Ji JY, Moon NS, Rocco JW, et al. (2006) Functional
identification of Api5 as a suppressor of E2F-dependent apoptosis in vivo. PLoS
Genet 2: e196. doi:10.1371/journal.pgen.0020196.
39. Lee SB, Cho KS, Kim E, Chung J (2003) blistery encodes Drosophila tensin
protein and interacts with integrin and the JNK signaling pathway during wing
development. Development 130: 4001–4010.
40. Vicente-Crespo M, Pascual M, Fernandez-Costa JM, Garcia-Lopez A,
Monferrer L, et al. (2008) Drosophila muscleblind is involved in troponin T
alternative splicing and apoptosis. PLoS ONE 3: e1613. doi:10.1371/journal.-
pone.0001613.
41. Madigan JP, Chotkowski HL, Glaser RL (2002) DNA double-strand break-
induced phosphorylation of Drosophila histone variant H2Av helps prevent
radiation-induced apoptosis. Nucleic Acids Res 30: 3698–3705.
42. Ji Y, Clark DV (2006) The purine synthesis gene Prat2 is required for Drosophila
metamorphosis, as revealed by inverted-repeat-mediated RNA interference.
Genetics 172: 1621–1631.
43. Holland C, Lipsett DB, Clark DV (2011) A Link Between Impaired Purine
Nucleotide Synthesis and Apoptosis in Drosophila melanogaster. Genetics 188:
359–367.
44. Hoffmann JS, Cazaux C (1998) DNA synthesis, mismatch repair and cancer.
Int J Oncol 12: 377–382.
45. Lyko F, Ramsahoye BH, Kashevsky H, Tudor M, Mastrangelo MA, et al. (1999)
Mammalian (cytosine-5) methyltransferases cause genomic DNA methylation
and lethality in Drosophila. Nat Genet 23: 363–366.
46. Weissmann F, Muyrers-Chen I, Musch T, Stach D, Wiessler M, et al. (2003)
DNA hypermethylation in Drosophila melanogaster causes irregular chromo-
some condensation and dysregulation of epigenetic histone modifications. Mol
Cell Biol 23: 2577–2586.
47. Bekesi A, Pukancsik M, Haasz P, Felfoldi L, Leveles I, et al. (2011) Association of
RNA with the uracil–DNA-degrading factor has major conformational effects
and is potentially involved in protein folding. FEBS J 278: 295–315.
48. Bekesi A, Pukancsik M, Muha V, Zagyva I, Leveles I, et al. (2007) A novel
fruitfly protein under developmental control degrades uracil–DNA. Biochem
Biophys Res Commun 355: 643–648.
49. Gelbart W, Emmert D (2010) Flybase High Throughput Expression Pattern
Data Beta Version. Flybase.
50. Wells BS, Johnston LA (2012) Maintenance of imaginal disc plasticity and
regenerative potential in Drosophila by p53. Dev Biol 361: 263–276.
51. Millard TH, Martin P (2008) Dynamic analysis of filopodial interactions during
the zippering phase of Drosophila dorsal closure. Development 135: 621–626.
Role of dUTPase in Drosophila Development
PLoS Genetics | www.plosgenetics.org9 June 2012 | Volume 8 | Issue 6 | e1002738