| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2
* Groupe ‘Instabilité génétique et cancer’,
Groupe ‘Toxico-résistance’, Institut de Pharmacologie et de Biologie Structurale, CNRS UMR 5089, 31077 Toulouse cedex 4, France; and
Department of Biochemistry, Medical school of Kyushu University, Fukuoka 812-8582, Japan
2Correspondence: Groupe ‘Instabilité génétique et cancer’, Groupe ‘Toxico-résistance’, Institut de Pharmacologie et de Biologie Structurale, CNRS UPR 9062, 205 route de Narbonne, 31077 Toulouse cedex 4, France. E-mail: cazaux@ipbs.fr; salles{at}ipbs.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and/or 
participate to the repair DNA
synthesis step. Using UV and cisplatin-modified DNA templates, we
measured in vitro that extracts from cells
overexpressing the error-prone DNA polymerase ß exhibited a five- to
sixfold increase of the ultimate DNA synthesis activity compared with
control extracts and demonstrated the specific involvement of Pol ß
in this step. By using a 28 nt gapped, double-stranded DNA substrate
mimicking the product of the incision step, we showed that Pol ß is
able to catalyze strand displacement downstream of the gap. We discuss
these data within the scope of a hypothesis previously presented
proposing that excess error-prone Pol ß in cancer cells could perturb
the well-defined specific functions of DNA polymerases during
error-free DNA transactions.—Canitrot, Y., Hoffmann, J.-S., Calsou,
P., Hayakawa, H., Salles, B., Cazaux, C. Nucleotide excision repair DNA
synthesis by excess DNA polymerase ß: a potential source of genetic
instability in cancer cells.
Key Words: cell variants • DNA repair • mutagenesis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. During
these transactions, cells seem to have developed a well-defined set of
DNA polymerases, each one being more or less specific to a given
process.
According to the current consensus, the role of DNA polymerase ß (Pol
ß) is narrowly delimited to base excision repair (BER). Pol ß is
expressed at a constant low level throughout the cell cycle
(3)
and is inducible by some genotoxic treatments
(4)
. Features that distinguish Pol ß from other cellular
polymerases are the lack of associated proofreading activity, its high
infidelity in replicating DNA in vitro, and its poor ability
to discriminate nucleotides at the level of binding (5)
.
In accordance with its low accuracy, Pol ß exhibits the lowest
discrimination against mutagenic analogs of dGTP modified by endogenous
processes (6)
. We recently demonstrated that an excess of
Pol ß in cells resulted in an increased mutagenesis (7)
.
Moreover, we have shown by using a murine cell line mimicking the
chronic myelogenous leukemia (CML) that Pol ß activity is increased
in this model (8)
. Excess Pol ß could thus favor the
genetic instability observed in this pathology, particularly during the
blast crisis. These data suggest that enhanced level of Pol ß may
have a role in cancer predisposition and/or tumor progression
(9)
. At the transcriptional level, Pol ß was actually
overexpressed in many cancer cells (10
, 11)
. High levels
of Pol ß have also been detected at the protein level. For example,
Pol ß was found more than 10-fold over-represented in prostate,
breast, or colon cancer tissues compared to adjacent normal tissues
(12)
. We quote here as additive example that the
intracellular level of Pol ß is more than 10-fold increased in
several ovarian cancer cell lines compared with normal ovarian tissue
(Fig. 1
). Based on these data, we hypothesize that an excess of the error-prone
Pol ß can perturb the well-defined specific functions of error-free
DNA polymerases in some cancer cells. We have already proposed that Pol
ß-mediated gap fillings during DNA transactions such as repair,
replication, or recombination pathways can introduce mutations
(9)
. Such a mutagenic interference could thus participate
to the genetic variability of primary tumors whose consequence is known
to be the emergence of invasive cell variants resistant to the
therapeutic treatment.
|
Among other DNA repair cellular processes, nucleotide excision repair
(NER) is a pathway involved when DNA is exposed to physical or chemical
agents, leading to bulky or helix-altering lesions (13)
.
This repair pathway operates by recognition of lesion, incision of the
damaged strand on either side of the targeted lesion, and excision of
an 
30-nucleotide fragment carrying the lesion. Finally, the filling
of the resulting gap is mediated by DNA polymerases in a proliferating
cell nuclear antigen (PCNA) -dependent manner and is followed by
ligation. In mammalian cell extracts, the absolute dependence of NER on
PCNA (14)
suggests the participation of DNA polymerase
(Pol ) or DNA polymerase (Pol ) in this last step since these
enzymes are known to be stimulated by PCNA (15)
.
Biochemical studies showed that the entire assay could be reconstituted
in the presence of either Pol or Pol (16)
.
To date, no evidence has shown an involvement of Pol ß in NER of
mammalian cells. Here we show that an enhanced level of Pol ß can
perturb the well-orchestrated steps in NER synthesis. We present
evidence that overexpressed error-prone Pol ß can compete with Pol
and/or Pol in the DNA synthesis step and that this involvement
enhances the probability for misincorporation. The putative effect in
cancer cells of the DNA polymerase ß imbalance on mutation
accumulation is discussed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. PUTpol ß harbors a fusion between the
pol ß cDNA and the bleSh gene conferring
resistance to zeocin. Control cells were transfected with a similar
pUT526
plasmid (Cayla, France) harboring only the bleSh
gene. Cells were grown at 37°C in MEM medium (BioMedia) supplemented
with glutamine, fetal calf serum (10%), and antibiotics
(penicillin/streptomycin) in a humidified atmosphere 95%
CO2/5% O2. The
pol ß cDNA and the bleSh gene being
transductionally fusioned, the stability of the transfectants in terms
of continuity of Pol ß overexpression was ensured by the continuous
presence in the culture medium of 250 µg/ml zeocin (Cayla). Ovarian
carcinoma cell lines were growth in RPMI 1640 medium (BioMedia)
supplemented with glutamine, fetal calf serum (10%), and antibiotics
(penicillin/streptomycin).
Preparation of radiolabeled 8-oxo-dGTP
Radiolabeled 8-oxodGTP was prepared as already described
(18)
. Briefly, 6 mM dGTP was incubated at 37°C for
2 h in the dark in a reaction mixture containing 100 mM sodium
potassium (pH 6.8), 30 mM ascorbic acid, and 100 mM
H2O2. Fifty microliters of
the reaction mixture was loaded onto a high-performance liquid
chromatography Spherisorb SAX column equilibrated with 150 mM potassium
phosphate (pH 5.5) and chromatographic separation was carried out at
25°C with the same buffer at a flow rate 1 ml/min. 8-Oxo-dGTP was
monitored with a UV detector (254 nm) and eluted with a retention 1
min later than for dGTP. Fractions containing 8-oxo-dGTP were combined,
applied to a DEAE-MemSep 1000 cartridge equilibrated with 50 mM
triethylammonium hydrogen carbonate (pH 7.0), and eluted with a linear
gradient (50–500 mM) of triethyl ammonium hydrogen carbonate (pH 7.0)
at a flow rate 1 ml/min. The fractions containing 8-oxo-dGTP were
lyophilized, dissolved in 20 mM sodium phosphate (pH 6.8), and stored
at -20°C. Purity and quantity of the preparation were examined in
electrophoretical measurements and UV spectrum analyses.
In vitro DNA repair reactions
Whole cell extract preparation
CHO extracts preparation was performed according to the
previously described protocol (19)
. Briefly, CHO cells
were washed with ice-cold phosphate-buffered saline, harvested by
centrifugation, and the cell pellet was suspended in hypotonic buffer
(10 mM Tris-HCl, pH 7.5, 10 mM KCl, 10 mM MgCl2,
1 mM DTT) containing protease inhibitors. The cells were disrupted with
20 strokes of the tight-fitting pestle in a dounce homogenizer during
incubation at 2°C. Nuclei were harvested by centrifugation and the
nuclear proteins were extracted in hypotonic buffer containing 350 mM
NaCl. Cytosolic and nuclear extracts proteins were precipitated by
addition of ammonium sulfate. The precipitates were resuspended in
dialysis buffer (50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 100 mM mono-K
glutamic acid, and 10% glycerol) and dialyzed for 2 h at 4°C.
The extracts were frozen in liquid nitrogen and stored at -70°C.
Preparation of plasmids and treatment with damaging agents
The 2959 bp pBS (pBluescript KS+; Stratagene, San Diego, Calif.)
and the related 3738 bp pHM14 plasmid (gift from R. D. Wood, ICRF,
U.K.) were prepared by alkaline lysis method from Escherichia
coli JM109. Both plasmids were carefully purified by one cesium
chloride and two neutral sucrose gradient centrifugation as described
(20)
. pBluescript KS+ plasmid was treated with either
cisplatin according to Hansson and Wood (21)
or with UV
light (300 J/m2) at 254 nm. pHM plasmids were treated with
10 mM MMS as described (22)
.
In vitro repair synthesis assay
The repair synthesis assay was performed as described
(23)
. The reaction mixture (50 µl) contained 300 ng each
of damaged pBluescript KS+ and untreated pHM14 closed circular
plasmids, 2 µCi 
[32P] dCTP (800 Ci/mmol),
200 µg cell extract protein, and 70 mM potassium glutamate in
reaction buffer containing 45 mM HEPES-KOH (pH 7.8), 7.4 mM
MgCl2, 0.9 mM DTT, 0.4 mM EDTA, 2 mM ATP, 20 µM
each dGTP, dATP and dTTP, 4 µM dCTP, 40 mM phosphocreatine, 2.5 µg
phosphocreatine kinase, 3.4% glycerol, and 18 µg bovine serum
albumin (BSA) as described. When specified, 10 µM radioactive
[32P] 8-oxo-dGTP, was added to a similar
reaction mixture containing 20 µM each dATP, dCTP and dTTP. When
noticed, rat Pol ß, purified as described (24)
, was
added. One unit of Pol ß corresponds to 1 nmol of dNTP incorporated
into acid-insoluble materials at 37°C in 60 min by using as a
substrate an activated calf thymus DNA preincubated with DNAase I.
Incubation was 3 h at 30°C. Before the electrophoresis on a 1%
agarose gel containing 0.5 µg/ml ethidium bromide, plasmid DNA was
purified from reaction mixtures and linearized with EcoRV,
except for the measurement of the complete gap filling activity.
In vitro repair incision assay
The incision assay was performed as already reported (25
, 26)
. Briefly, repair reaction took place as described above
except that deoxyribonucleotides were omitted and 4.5 µM aphidicolin
was added. After 2 h at 30°C, the reaction was stopped by the
addition of 0.5% sodium dodecyl sulfate (SDS) and 0.1 M EDTA. Plasmids
were treated by gentle manual agitation with 200 µg/ml proteinase K
(42°C, 1 h), purified by phenol-chloroform-isoamyl alcohol
extraction, ethanol precipitated, washed in 70% ethanol, dried, and
redissolved in TE buffer. DNA was incubated (10 min at room
temperature) in reaction mixture containing 90 mM HEPES-KOH (pH 6.6),
10 mM MgCl2, 0.2 µCi of
[32P] dCTP, 2 mM dithiothreitol, 20 µM of
each of dGTP, dATP, dTTP and 1 unit of E. coli DNA
polymerase I large fragment. The reaction was stopped by addition of
EDTA 50 mM. DNA was extracted, precipitated, washed, linearized by
EcoRV, and electrophoresed on a 1% agarose gel containing
0.5 µg/ml ethidium bromide.
Quantification of repair
Gels were dried and processed on a PhosphorImager (Storm System,
Molecular Dynamics, Sunnyvale, Calif.) and radioactivity was quantified
(Image quant 1.1 software). Photographs of the ethidium bromide-stained
gels were scanned and processed to normalize for plasmid DNA recovery.
Strand displacement assay
To serve as a DNA template, an 85-mer oligonucleotide was
hybridized to a 5'-32P-labeled 20-mer primer and
a looped 43-mer (see Fig. 6A
). This 28 nucleotide-gapped,
double-stranded DNA template (5 ng) was replicated for 1 h
in vitro by various amounts of purified Pol ß in reactions
(15 µl) containing 45 mM HEPES-KOH (pH 7.8), 7 mM
MgCl2, 1 mM DTT, 0.4 mM EDTA, 3.4% glycerol, 65
mM potassium glutamate, 18 µg of BSA, 200 µM each dATP, dCTP, dGTP,
and dTTP. At the end of the reaction, 5 µl of stopping buffer (90%
formamide/0.1% xylene cyanol/0.1% bromophenol blue/0.1 mM EDTA) was
added. Samples were denatured for 10 min at 70°C and loaded onto a
15% polyacrylamide/7 M urea/30% formamide gel.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Control extracts from CHO UV4 and
UV135 mutant cells (respectively deficient for the incisions 5' and
3'-side of the lesion) exhibited no NER activity under our assay
conditions (data not shown). Thus, the repair activity of Sh and
Polß::Sh cell extracts can confidently be attributed to
true NER. In addition, mixing both extracts in a same reaction restored
a signal close to that of wild-type cell extracts (data not shown),
showing that our preparations were proficient. As shown in Fig. 2
, protein extracts obtained from pol
ß::Sh cells (11)
constitutively
overexpressing 7–10 times Pol ß (7)
exhibited a
five- to sixfold higher repair synthesis activity on UVC-treated
plasmid than extracts from control cells (Sh). To ensure
that this increased activity was specific of the Pol ß-mediated
action, we used the chain terminator ddCTP, which has been demonstrated
to be a good substrate of Pol ß but a poor substrate of the
replicative DNA polymerases and (11
, 27)
. The
presence of ddCTP in the pool of nucleotides during the NER step of
synthesis decreased the signal of incorporation from Pol
ß-overexpressing extracts on UVC-treated plasmids (Fig. 2)
. On the
other hand, ddCTP had no effect on the NER process mediated by
Sh cell extracts, showing predominance of the action of DNA
polymerases and in control cells. This result suggests a
specific contribution of Pol ß in the enhanced signal obtained with
Pol ß-overexpressing cell extracts. A complementary demonstration was
obtained by incubating control Sh cell extracts with an
increased amount of purified Pol ß. The addition of four units of
purified protein, which correspond to the specific activity measured in
Pol ß-overexpressing cell extracts (data not shown), increased the
incorporation of radiolabeled nucleotides on UV-treated plasmids (Fig. 2)
.
|
BER activity of Sh and pol
ß::Sh cell extracts
Because it is known that UVC induced the formation of pyrimidine
dimers recognized by the NER pathway but also of minor lesions such as
thymine glycols recognized by the BER, the increased repair synthesis
might be due to NER, BER, or both. To evaluate the residual
participation of BER in the yield of radiolabeled incorporation, we
performed two sets of control experiments. First, a methyl-methane
sulfonate (MMS) damaged plasmid DNA was incubated in a repair reaction
since MMS provokes the formation of base methylation repaired by the
BER but not the NER pathway. As shown in Fig. 3A
, no significant variation in the repair synthesis signal
was observed with extracts from Pol
ß::Sh cells compared to the extracts
from Sh cell extracts. This result is in accordance with
cellular survival data showing that Pol
ß::Sh cells are as sensitive to MMS
as Sh cells (7)
and shows that the action of
Pol ß in BER is not the limiting step. In other terms, since it has
been shown that null pol ß mutants are sensitive to MMS
(28)
, the intracellular concentration of endogenous Pol
ß seems to be saturating in normal cells. Second, a cisplatin-damaged
plasmid DNA was incubated in the repair reaction since adducts produced
by this drug are mostly recognized in vitro by the NER
pathway (29)
. As expected, a five- to sixfold increase in
the repair synthesis activity was found with extracts from
pol ß::Sh cells by comparison to the
control Sh cells (Fig. 3B
). This repair signal
was dependent on addition of ddCTP or four units of purified Pol ß as
reported with the UVC-damaged plasmid DNA (Fig. 2)
. However, the signal
restitution by purified Pol ß is to a lesser extent, maybe because
recombinant Pol ß purified from E. coli is less efficient
with regard to cisplatin than that extracted from eukaryotic cells.
|
NER incision activity of Sh and pol
ß::Sh cell extracts
The variation of the repair synthesis activity in the assay might
be due to either an increased activity in incision of the damaged DNA,
an increase of the length of the repair patch, or both. To alleviate
the ambiguity, an incision assay (25)
was performed with
extracts from Sh and pol
ß::Sh cells. No difference of incision activity
in a NER reaction with cisplatin or UVC-damaged plasmid DNA was
observed (Fig. 4
). Another control was realized by measuring the NER activity using
NER-deficient extracts from CHO UV4 and UV135 mutant cells. No
significant signal was detected with such extracts, complemented or not
with purified Pol ß (data not shown), showing as expected that this
DNA polymerase is not able to complement the UV4/UV135 defects and does
not play a role in incision. Taken together, our data demonstrate that
when overexpressed into the cell, Pol ß participates in the NER
reaction through the DNA polymerization step.
|
Complete gap filling activity by Sh and
pol ß::Sh cell extracts
In vitro, the NER reaction proceeds up to the ligation
step as already shown (23)
. Therefore, it was of interest
to examine whether Pol ß could promote a complete gap filling when
involved in the repair patch synthesis. We conducted similar
experiments as described in Figs. 2
and 3B
, except that
plasmids were not linearized after incubation. By using such
substrates, the production of an only partial gap DNA synthesis would
result in an impediment for ligation process and therefore in an
accumulation of the radioactivity in open circular forms. In contrast,
a majority of radioactive materials in closed circular forms would
reflect a complete gap filling DNA synthesis and subsequent ligation.
As observed in Fig. 5
, higher incorporation of radioactive nucleotide in the closed circular
form of the repaired plasmids damaged with cisplatin or UVC light was
observed for both Sh and pol
ß::Sh cell extracts, strongly suggesting that
the DNA polymerization step went to completion in the absence or
presence of excess of Pol ß. Furthermore, addition of four units of
purified Pol ß to normal cell extracts after the 3 h repair
reaction for an additional hour did not result in an increased
incorporation of radioactivity in open or closed circular forms,
indicating that the enhanced repair synthesis activity in
pol ß::Sh cell extracts was not due
to completion of repair reactions blocked at intermediate steps (data
not shown). Taken together, these data show that Pol ß when
substituting for Pol / is able to promote completion of the gap
filling with increased size of the repair patch.
|
DNA strand displacement by Sh and pol
ß::Sh cell extracts.
To investigate the molecular basis of the enhanced repair
synthesis activity mediated by an excess of Pol ß, we measured the
ability of pol ß::Sh cell extracts to
displace the DNA strand opposite the template after completion of the
gap filling. We thus synthesized a 28 nt gapped duplex DNA mimicking
the substrate of the DNA polymerases after the NER excision step. To
discriminate between a ligation process and a continuous DNA synthesis
mediated by a strand displacement, we constructed a 85/20/43 substrate
(see Fig. 6A
) in which the 43-mer strand 5' to the gap, annealed to the
85/20 primed template strand, possesses a loop of 6 extra bases placed
in the hybridized part of the duplex. We reasoned that gap filling plus
ligation would yield a 91-mer product whereas strand displacement
accompanying DNA synthesis without ligation process would result in a
85-mer product. The outcome of these experiments is shown in Fig. 6B
, where one can distinctly see a Pol ß-mediated
appearance of the 85-mer by addition of purified Pol ß in cell
extracts (Fig. 6B
). In conclusion, strand displacement DNA
synthesis catalyzed by excess Pol ß may account for the enhanced
repair synthesis in recombinant cells.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, a specific context
that has been shown to occur in various tumors such as glioma and
lymphoma (10
, 11)
, breast, colon, and prostate
(12)
as well as in ovarian tumor cell lines (Fig. 1)
.
Because the genome of tumor cells has been shown to carry increased
level of DNA damage, an up-regulation of DNA repair proteins may be a
first response. The accumulation of alkylation intermediates can, for
example, lead to the activation (30)
in some cancer
tissues of the pol ß promoter through an increased
DNA-alkylating agent-dependent phosphorylation of the cAMP response
element binding protein 1. It could also be envisaged an up-regulation
of the Pol ß expression with the aim of complementing a deficiency of
replicative DNA polymerases such as Pol , which is altered in
several cancer cells (31
, 32)
. Moreover, we have shown
that Pol ß-mediated mutation accumulation is dependent on the cell
proliferation state, being about twice as important during stationary
than in log-phase (data not shown). Thus, it can be assumed
(9)
that Pol ß could complement replicative DNA
polymerases, which would be down-regulated during the numerous
nonproliferating states of the long carcinogenesis process
(33)
where the tumor is prevented from growing because of
physiological barriers, deficiencies in blood supply, inadequate pH,
etc. Aneuploidy processes found in nearly all major human tumors
(2)
could also account for the overexpression. For
instance, three copies of
polß-containing chromosome 8 can
be found in acute myeloid leukemia (34)
. A more subtle
regulation of the Pol ß expression is also likely to take place
during the carcinogenesis process. For example, we have shown that the
presence of the bcr-abl tyrosine kinase, the hallmark of
chronic myelogenous leukemia is concomitant to an increase of both the
level and activity of Pol ß (8)
.
Incidence of Pol ß excess on NER synthesis
DNA polymerase ß is a 39 kDa protein with both
nucleotidyl-transferase and 5'-deoxyribose phosphodiesterase activities
involved in the BER pathway as well as in meiosis (28
, 35)
. It is expressed at a constant low level throughout the cell
cycle (3)
and is inducible by some genotoxic treatments
(4)
. It has been described as the least accurate mammalian
DNA polymerase, resulting from many features that characterize the
enzyme: 1) the lack of associated proofreading activity
(36)
, 2) its poor ability to discriminate
nucleotides (37)
, 3) its capacity to
efficiently catalyze error-prone translesion synthesis across
intrastrand cross-links (19
, 38)
, and 4) its
distributive mode of DNA synthesis (37)
.
The high accuracy of the DNA polymerases involved in the patch
synthesis of NER pathway (Pol and ) results in an error-free
restitution of the nucleotide sequence (13)
. Using
UV-treated DNA plasmids and cell extracts, we present evidence that the
inaccurate Pol ß can play a key role in NER in the specific context
of its overexpression (Fig. 1)
. We have verified that this interference
is due neither to the BER-associated action of Pol ß (Figs. 2
, 3)
nor
to a putative involvement of Pol ß in the incision step (Fig. 4)
. Our
data show that Pol ß in excess acts during the final NER gap filling
step (Fig. 5)
. The consequence of this involvement is the displacement
of the DNA strand downstream of the gap (Fig. 6)
. Considering the
particular Pol ß features, we also tested the extent of
misincorporation of the oxidative base 8-oxo-G, known to potentially
lead to the 8oxoG:A mispairing after DNA replication (39)
,
by including 32P-radiolabeled 8-oxo-dGTP into the
cell extract-mediated repair reaction of UV treated plasmids. Cell
extracts from Pol ß overexpressing cells showed a higher ability to
incorporate radiolabeled 32P 8-oxo-dGTP than cell
extracts from control cells (data not shown). Taken together, these
results lead us to propose that in excess, Pol ß could compete with
or substitute for error-free DNA polymerases and during the
accurate NER DNA synthesis, switching it to an error-prone process
(Fig. 7
).
|
A model for Pol ß-mediated NER DNA synthesis
In the base excision repair reaction, Pol ß is involved not only
in the single-nucleotide patch repair but also in long patch repair
although its role in this pathway remains unclear. The role of Pol ß
in the long patch repair was shown by using pol ß null cell extracts
that do not repair a reduced abasic site normally repaired through the
long patch pathway (40)
. The reaction can be reconstituted
with either DNA polymerase ß or , suggesting that both DNA
polymerases can substitute for each other in long patch BER
(41)
. In addition, it has been shown recently by using
cell extracts and a uracil-containing plasmid DNA substrate that the
excision of a 3 nt oligonucleotide produced after strand displacement
and flap incision is highly dependent on Pol ß (42)
.
Indeed, the oligonucleotide release was strongly reduced in the
pol ß null cell extract (43)
. The
reconstitution of long gap filling in pol
ß null cell extracts was also achieved in a
dose-dependent manner by adding purified Pol ß (42)
.
Finally, a very small amount of purified Pol ß was sufficient to
restore the defect (42)
, showing that Pol ß plays an
essential role in long patch BER.
Taken together, these experiments indicate that DNA repair synthesis
performed by Pol ß is required for excision in long patch BER. If
this 2- to 8-nucleotide long patch BER route is minor in mammalian
cells, a context of Pol ß overexpression could thus reinforce this
route. We hypothesize here that a comparable situation in presence of
excess Pol ß may occur in the NER 27–29 nt DNA synthesis. Our
experiments indicate that NER synthesis, in which Pol ß is not
involved in normal conditions (16
, 44)
, can be
accomplished by overexpressed Pol ß as in long patch BER. It could
substitute for the PCNA-containing replication machinery by promoting
the initiation of the DNA synthesis (hypothesis A in the model
presented in the Fig. 7
) or by switching the elongation complex and
complete the gap (hypothesis B in Fig. 7
).
If we consider that the intervention of Pol ß in NER could be the
mirror image of what happens in the long patch pathway BER, it is
likely that Pol ß acts alone in the NER DNA elongation step instead
of as part of a protein complex. It could thus promote the DNA
synthesis, maybe by forming homo oligomers along the DNA substrate, the
protein being able to undergo indefinite self-association
(45)
. Indeed, if the BER long patch pathway is performed
by several proteins such as Pol ß, Pol , or Pol , PCNA , RPA,
FEN1 structure-specific endonuclease, and DNA ligase I (41
, 46
47
48)
, it has not been proved that these partners act through
the formation of a physical complex. No scaffold protein such as XRCC1,
involved in the short-patch pathway, was actually detected in the long
patch route (49)
. In addition, Pol ß-protein analysis
showed physical interactions and functional coordination only between
Pol ß and BER proteins such as human apurinic Ape endonuclease
(50)
, DNA ligase I (48)
, and the scaffold
protein XRCC1 (49)
, which functions only in the
short-patch complex containing also DNA ligase III and PARP proteins
(51)
. Last, DNA polymerase ß is present neither in
multiprotein replication complexes/replication foci (52)
nor in RPA/PCNA-containing replication complexes recently underscored
close to the replication fork in the long patch BER synthesis
(47)
.
According to its biochemical features, we have shown here that
involvement of Pol ß in the NER gap filling reaction resulted in
strand displacement, resulting in the observed enhanced labeled
materials incorporation in our NER assay. PCNA , engaged in the
formation of dynamic complexes with a number of alternative proteins,
which facilitates the excision during long patch BER (53)
,
could promote the release of the generated flap by stimulating the
FEN-1 endonuclease (54)
, the subsequent ligation step
being mediated by the NER DNA ligase I (Fig. 7)
. This hypothetical
model matches a recent observation in yeast showing that UV-induced
lesions can be repaired by a process involving the flap endonuclease
FEN-1 (55)
. However, we cannot exclude a PCNA requirement
at a step preceding DNA resynthesis, as it has been shown in DNA
mismatch repair (56)
.
Incidence on DNA mutagenesis of the error-prone Pol ß
interference
Early studies of the fidelity of DNA synthesis by Pol ß using
single-stranded templates demonstrated that the average Pol ß error
rates for single-base substitution (7x10-4) and
single-base deletion (3–9x10-4) were
substantially higher than those measured for the replicative DNA
polymerases, consistent with its lack of 3' 
5' proofreading
exonuclease activity (36)
. This low-fidelity mode of DNA
synthesis was recently examined during the filling of gapped DNA,
generating clustered multiple substitutions (57)
.
Considering that such types of errors were observed in experiments
using UV-irradiated shuttle vectors (58)
, these data
suggest that these errors could have resulted partially from gap
filling synthesis in vivo by Pol ß through the NER
pathway.
Obstacles such as nutritional requirement, inadequate blood supply, and
barriers generated by normal matrices limit the expansion of cancer
cells during the long-time of cancer process. During these successive
periods of nonproliferation, where DNA replication is affected, Pol
ß, whose level is constant throughout the cell cycle, may play a key
but mutagenic role by becoming a predominant DNA polymerase. In
bacteria, DNA recombination has been proposed to be associated to
mutational events occurring during the plateau phase (59)
.
Since it is a part of recombination nodules (60)
, Pol ß
could be involved, for example, in the gap filling of recombination
intermediates.
In the light of our data showing that an excess DNA polymerase ß
results in a mutator phenotype in mammalian cells (7)
, we
proposed that in Pol ß-overexpressing cells such as the
above-described cancer cells, Pol ß could enhance genetic instability
(9)
. We hypothesized that an excess of this error-prone
enzyme could perturb error-free DNA transactions by interfering with
the action of the accurate and processive DNA polymerases involved in
these processes. Here we showed that nucleotide excision repair could
be affected. By using as a model the nucleoside 8-oxo-dG, we also
demonstrated the increased capacity of the enzyme for misincorporation
of mutagenic base analogs. The involvement in cancer cells of Pol ß
in such a pathway could therefore result in a mutagenic response toward
some therapeutic treatments based on NER-repaired genotoxic drugs, such
as cisplatin, melphalan, or mechlorethamine (7)
. Variant
cells could thus emerge from the treated tumor mass, some of them being
likely to gain a growth advantage and consequently to be selected for
their invasive and chemoresistant potentials.
We cannot rule out that a ‘non-natural’ intervention of the error-prone Pol ß (through the context of its overexpression) in error-free DNA transactions pathways may be generalized to processes other than NER involving a long patch DNA synthesis, such as DNA homologous recombination, mismatch repair or DNA replication. Further investigations regarding these hypotheses will allow a better understanding of one of the multiple ways that confer genetic variability during tumor progression.
Pol ß belongs to a special group of enzymes, recently the focus
of attention of many groups, also called ‘mutases’ (61)
,
which are part of stress-inducible processes. This enzyme group also
includes DNA polymerases Pol 
, Pol 
, Pol IV, and Pol V
(62)
. A better knowledge of such enzymes could represent a
novel approach to find pharmaceutical means, such as mutase inhibitors,
for protecting people exposed to DNA damaging agents from mutagenic
alterations during the course of cancer therapies.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication December 21, 1999.
Revision received March 13, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß are able to incorporate anti-HIV deoxynucleotides into DNA. J. Biol. Chem. 267,21459-21464This article has been cited by other articles:
|
|
 |
|
  F. Boudsocq, P. Benaim, Y. Canitrot, M. Knibiehler, F. Ausseil, J. P. Capp, A. Bieth, C. Long, B. David, I. Shevelev, et al. Modulation of Cellular Response to Cisplatin by a Novel Inhibitor of DNA Polymerase {beta} Mol. Pharmacol., May 1, 2005; 67(5): 1485 - 1492. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Bavoux, A. M. Leopoldino, V. Bergoglio, J. O-Wang, T. Ogi, A. Bieth, J.-G. Judde, S. D. J. Pena, M.-F. Poupon, T. Helleday, et al. Up-Regulation of the Error-Prone DNA Polymerase {kappa} Promotes Pleiotropic Genetic Alterations and Tumorigenesis Cancer Res., January 1, 2005; 65(1): 325 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Canitrot, J.-P. Capp, N. Puget, A. Bieth, B. Lopez, J.-S. Hoffmann, and C. Cazaux DNA polymerase {beta} overexpression stimulates the Rad51-dependent homologous recombination in mammalian cells Nucleic Acids Res., September 27, 2004; 32(17): 5104 - 5112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Fotiadou, O. Henegariu, and J. B. Sweasy DNA Polymerase {beta} Interacts with TRF2 and Induces Telomere Dysfunction in a Murine Mammary Cell Line Cancer Res., June 1, 2004; 64(11): 3830 - 3837. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Dalal, J. L. Kosa, and J. B. Sweasy The D246V Mutant of DNA Polymerase {beta} Misincorporates Nucleotides: EVIDENCE FOR A ROLE FOR THE FLEXIBLE LOOP IN DNA POSITIONING WITHIN THE ACTIVE SITE J. Biol. Chem., January 2, 2004; 279(1): 577 - 584. [Abstract] [Full Text] [PDF]
|
|
|
|
