ABSTRACT
The
rfa1-M2
and
rfa1-M4
Saccharomyces cerevisiae
mutants, which are altered in the 70 kDa subunit of replication protein A (RPA)
and sensitive to UV and methyl methane sulfonate (MMS), have been analyzed for
possible checkpoint defects. The G1/S and intra-S DNA damage checkpoints are defective in the
rfa1-M2
mutant, since
rfa1-M2
cells fail to properly delay cell cycle progression in response to UV
irradiation in G1 and MMS treatment during S phase. Conversely, the G2/M DNA
damage checkpoint and the S/M checkpoint are proficient in
rfa1-M2
cells and all the checkpoints tested are functional in the
rfa1-M4
mutant. Preventing S phase entry by
[alpha]
-factor treatment after UV irradiation in G1 does not change
rfa1-M4
cell lethality, while it allows partial recovery of
rfa1-M2
cell viability. Therefore, the hypersensitivity to UV and MMS treatments
observed in the
rfa1-M4
mutant might only be due to impairment of RPA function in DNA repair, while the
rfa1-M2
mutation seems to affect both the DNA repair and checkpoint functions of Rpa70.
Replication protein A (RPA, also known as RFA or SSB) is a heterotrimeric single-stranded DNA binding protein composed of three subunits of ~70, 34 and 14 kDa, whose structure and function is conserved in all
eukaryotic organisms from yeast to man (
1
-
4
). RPA was initially identified as a protein factor essential for
in vitro
SV40 DNA replication (
1
-
3
) but, more recently, it has been implicated in several other DNA transactions,
such as DNA repair, recombination and transcription (
5
-
13
).
The DNA binding activity was found to be associated with the Rpa70 subunit (
3
,
4
), while the functions of Rpa34 and Rpa14 are still poorly understood.
Interactions between individual subunits have suggested a role for Rpa14 in RPA
complex formation (
14
,
15
). Antibodies specifically interacting with Rpa34 or Rpa14 inhibit DNA
replication and DNA repair
in vitro
, but do not affect DNA binding (
16
-
18
), indicating that these polypeptides are required for other RPA functions.
Rpa34 is phosphorylated at the G1/S boundary of the cell cycle (
19
-
20
) and it becomes hyperphosphorylated as a consequence of DNA damage caused by
ionizing or UV radiation (
21
-
22
).
In vitro
studies indicate that cyclin-dependent kinases (CDKs) and DNA-dependent protein kinase (DNA-PK) are involved in Rpa34 phosphorylation events (
20
,
23
-
26
), but the functional significance of Rpa34 phosphorylation has not yet been
assessed
in vivo
.
DNA damage caused by genotoxic agents delays cell cycle progression of
eukaryotic cells through a network of highly conserved surveillance mechanisms,
called checkpoints, whose function is likely to delay cell cycle progression to
allow time for DNA repair (for recent reviews see
27
-
30
). The DNA damage checkpoints act at three stages: the G1 -> S transition (G1/S), the rate of S phase progression (intra-S), the G2 -> M phase transition (G2/M). The search for mutations abolishing
cell cycle arrest caused by genotoxic agents has allowed the identification of
a number of checkpoint genes in
Saccharomyces cerevisiae
and in
Schizosaccharomyces pombe
(reviewed in
29
-
30
). In budding yeast the
RAD9
,
RAD53/MEC2/SAD1/SPK1
,
RAD17
,
RAD24
,
MEC1/ESR1
and
MEC3
genes are all required for the DNA damage checkpoints listed above (
30
-
38
). Among them, the
RAD9
,
RAD17
,
RAD24
and
MEC3
gene products have also been implicated directly in DNA repair processes (
30
,
36
). Moreover, increasing evidence indicates that replication proteins are
involved in the surveillance mechanism linking the completion of S phase to
entry into mitosis (S-M checkpoint) (
39
-
42
).
Since RPA is required for both DNA replication and repair, we asked whether this
multifunctional and essential protein complex also played a role in some of the
checkpoints acting at different stages of the cell cycle. In this manuscript we
provide evidence that the
rfa1-M2
mutant, altered in the p70 subunit of budding yeast RPA, fails to properly
delay cell cycle progression in response to DNA damage in G1 or during S phase,
suggesting that RPA may be involved in the interconnections between DNA
replication, DNA repair and cell cycle control.
Yeast strains K699-M2 and K699-M4 have been obtained by replacing the wild-type
RFA1
chromosomal copy with the mutant alleles (
9
) in strain K699 (MAT
a
ade2-1 trp1-1 can1-100 leu2-3,112 his3-11,15 ura3
) by the two-step procedure (
43
). Briefly, replacement was obtained by transforming K699 cells with the
Sac
I-linearized pML27 and pML28 YIp5-derivative plasmids carrying, respectively, the
rfa1-M2
and
rfa1-M4
mutant alleles (
9
). Correct replacements have been tested by Southern blot analysis. Standard
genetic techniques and media were as described (
44
).
To measure cell cycle delay at the G1/S boundary in response to UV treatment,
log phase cultures were blocked in G1 with 2 [mu]g/ml [alpha]-factor as previously described (
45
), spread on YPD plates and immediately UV irradiated with 40 J/m
2
. Cells were then washed from the plates, rinsed to remove pheromone and
resuspended in fresh YPD at 25oC. At timed intervals, samples were removed for FACS analysis and 4 * 10
2
-1 * 10
4
cells were plated in triplicate onto YPD plates to measure cell survival after
3 days incubation at 25oC.
The intra-S DNA damage checkpoint was analyzed essentially as described by Paulovich
and Hartwell (
35
). Briefly, [alpha]-factor pre-synchronized cells were released from the [alpha]-factor block in YPD medium containing 0.02%
methyl methane sulfonate (MMS). At timed intervals, samples were collected for fluorescence-activated cell sorting (FACS) analysis and measurement of cell survival,
as described above.
To analyze cell cycle delay at the G2/M boundary in response to UV treatment,
log phase cultures were first blocked in G2/M by treatment with nocodazole (5 [mu]g/ml) and dimethylsulfoxide (1%) for 110 min and then plated on YPD and
immediately UV irradiated with 50 J/m
2
. Cells were then washed from the plates, rinsed to remove nocodazole and
resuspended in fresh YPD at 25oC. At timed intervals, cells were collected and the percentage of uni- and bi-nucleate cells was scored microscopically after staining with 4',6-diamidino-2-phenylindole (DAPI).
The details of cell treatment before FACS analysis with a Becton Dickinson
FACScan have already been described (
45
).
We have previously shown that the
rfa1-M2
and
rfa1-M4
mutants are highly sensitive to UV radiation and MMS treatment (
9
). The sensitivity of these
rfa1
mutants to UV and MMS treatments is recessive (unpublished observations) and,
in principle, could be due to defects in DNA repair mechanisms, to altered DNA
damage checkpoints, or to both. As a first step to address this point, we
tested the behavior of the
rfa1-M2
and
rfa1-M4
mutants in response to UV irradiation in G1.
[alpha]-Factor-arrested wild-type and mutant cells were UV irradiated and budding profile and DNA content were monitored
after release from the [alpha]-factor block. As shown in Figure
1
A, bud emergence after [alpha]-factor release was similarly delayed in UV-irradiated wild-type and
rfa1-M4
cells compared with unirradiated controls, while this delay was consistently
reduced in the
rfa1-M2
UV-treated culture. In fact, bud emergence in
rfa1-M2
irradiated cells was earlier by at least 70 min compared with wild-type and
rfa1-M4
cells under the same conditions (Fig.
1
A).
It has recently been shown that slowing of the rate of S phase progression when DNA is chronically damaged with the alkylating agent MMS is genetically controlled (
35
). Since we found that
rfa1-M2
mutant cells are defective in properly delaying cell cycle progression after
DNA damage in G1, we tested their capacity to slow down the rate of DNA
replication when 0.02% MMS was added immediately after [alpha]-factor release and maintained during the subsequent S phase. As
shown in Figure
3
A, wild-type cells treated with MMS replicated their DNA more slowly than
untreated cells. Conversely,
rfa1-M2
cells were defective in slowing down the rate of DNA synthesis in response to
MMS treatment. In fact, although MMS treatment caused a delay in S phase entry
of both wild-type and
rfa1-M2
cultures compared with the untreated controls, MMS-treated
rfa1-M2
cells completed DNA replication at least 90 min earlier than the wild-type under the same conditions (Fig.
3
A). Accordingly, cell survival of
rfa1-M2
cells, compared with the wild-type, was drastically reduced by MMS treatment during S phase (Fig.
3
B).
The
rfa1-M2
and
rfa1-M4
mutants are weakly sensitive to hydroxyurea (HU) treatment (45% and 51% cell survival respectively, compared with 100% survival of the wild-type isogenic strain, after 4 h in 0.2 M HU). Nevertheless, treatment with
0.2 M HU of logarithmically growing or [alpha]-factor pre-synchronized
rfa1-M2
and
rfa1-M4
cells caused their cell cycle arrest as large budded cells with a single
undivided nucleus, short spindles and an S phase DNA content (data not shown).
These data suggest that
rfa1-M2
and
rfa1-M4
cells can properly block entry into mitosis when DNA is not completely
replicated.
Similarly, the G2/M DNA damage checkpoint was not defective in these mutants. As
shown in Figure
4
,
rfa1-M2
and
rfa1-M4
cells did not prematurely enter into mitosis when cells were UV irradiated in
G2 after nocodazole arrest, although cell survival (11 and 0.13% in
rfa1-M2
and
rfa1-M4
cultures respectively) was dramatically reduced compared with wild-type cells (83%) under the same conditions. As a control, we used the
mec3
[Delta] strain, which is altered in this checkpoint (
38
), and, accordingly, failed to delay nuclear division in response to UV
irradiation in G2 (Fig.
4
).
The RPA single-stranded DNA binding protein is a multifunctional protein required for
several DNA transactions (
5
-
13
). Indeed, yeast
rfa1
mutants are defective in DNA replication and recombination and show
hypersensitivity to UV and MMS treatments (
9
,
11
,
12
). This last phenotype is likely to be related to the fact that RPA might
function at the earliest stage of nucleotide excision repair (NER), as
indicated by its physical interaction with the damage recognition protein XPA
and the endonuclease XPG (
8
). In addition, its single-stranded DNA binding activity may be required for gap filling repair
synthesis (
6
,
7
).
The data presented in this manuscript indicate that the
RFA1
gene product is also required for properly delaying cell cycle progression
after UV irradiation in G1 and continuous DNA damage during S phase. In fact,
rfa1-M2
mutant cells are defective in these two checkpoint mechanisms. However,
sensitivity of
rfa1-M2
cells to UV irradiation in G1 can only partially be ascribed to a checkpoint
defect, since preventing S phase entry by [alpha]-factor treatment after UV irradiation allows only a partial
recovery of cell viability. These data reveal a checkpoint function for Rpa70
and, at the same time, further support a role for RPA in DNA repair. A function
for Rpa70 in DNA repair
in vivo
is also strengthened by the observation that loss of cell viability of the
rfa1-M4
mutant, after UV irradiation in G1 and MMS treatment during S phase, does not appear to be related to checkpoint defects. Therefore, while both DNA replication and DNA repair are defective in the two
analyzed mutants, the effect of
rfa1
mutations on checkpoints appears to be allele specific. Production and
characterization of more
rfa1
alleles will allow one to establish whether the checkpoint function of Rpa70 is
dependent upon its repair/replication function and whether it is required only
for the subset of DNA damage checkpoints impaired in the
rfa1-M2
mutant, or also for other surveillance mechanisms. Furthermore, since RPA
appears to be involved in regulating UV-inducible genes (
13
), the function of RPA in DNA damage checkpoints might be related to its
possible role in modulating the expression of checkpoint or DNA repair genes.
It is still unknown whether the role of RPA in checkpoints is limited to the
Rpa70 single-stranded DNA binding subunit or is a function of the whole RPA complex. We
have previously found that
rfa2
mutants, altered in the p34 yeast RPA subunit, are defective in proper S phase
progression (
46
). However, the UV sensitivity of the
rfa2
mutants so far analyzed is indistinguishable from that of wild-type cells and they are proficient in all the cell cycle checkpoints
tested (
46
; unpublished observations). Therefore, although hyperphosphorylation of Rpa34
in response to DNA damaging agents points to a role of this subunit in some
repair or checkpoint pathway, no functional data yet support this hypothesis.
A major challenge will be to understand the biochemical basis of Rpa70 function
in the G1 and intra-S DNA damage checkpoints. Single-stranded DNA is an intermediate of many repair pathways and is
generated as a consequence of incomplete DNA replication. Moreover, it has
recently been suggested that accumulation of regions of single strandedness at
telomeres in
cdc13
mutants might constitute the signal for the
RAD9
-dependent G2/M DNA damage checkpoint (
47
). Considering that Rpa70 is a single-stranded DNA binding protein essential for DNA replication and required
for DNA repair, its checkpoint function might be directly related to its single-stranded DNA binding activity. In this view, it will be interesting to
investigate whether Rpa70 can act as a sensor of single-stranded regions or as a target of some of the checkpoint pathways which
are activated by genotoxic agents.
We wish to thank Roberta Fraschini and Raffaella Zaccarini for help in many
experiments and all the members of our laboratory for useful discussions and
criticisms. This work was partially supported by the Progetto Strategico Ciclo
Cellulare e Apoptosi, by a grant from Associazione Italiana per la Ricerca sul
Cancro and by contracts CHRX-CT93-0248 and ERB CHRX-CT94-0685 from the European Union. MPL and HN were supported,
respectively, by fellowships from Fondazione Adriano Buzzati-Traverso and the Human Capital Mobility EU Program.
REFERENCES
Return