(Received for publication, January 29, 1997, and in revised form, May 19, 1997)
From the Departments of Radiology and Radiation
Oncology, Albert Einstein College of Medicine, Bronx, New York 10461, § Pennington Biomedical Research Center, Louisiana State
University, Baton Rouge, Louisiana 70808, and ¶ Department of
Pediatrics, Section of Pediatric Endocrinology, Wells Center for
Pediatric Research and Department of Biochemistry and Molecular
Biology, Indiana University School of Medicine, Indianapolis, Indiana
46202
The Drosophila ribosomal protein S3
has been previously demonstrated to cleave DNA containing 8-oxoguanine
residues and has also been found to contain an associated
apurinic/apyrimidinic (AP) lyase activity that cleaves phosphodiester
bonds via a ,
-elimination reaction. The activity of this protein
on DNA substrates containing incised AP sites was examined. A
glutathione S-transferase fusion protein of S3 was found to
efficiently remove sugar-phosphate residues from DNA substrates
containing 5
-incised AP sites as well as from DNA substrates
containing 3
-incised sites. Removal of 2-deoxyribose-5-phosphate as
4-hydroxy-2-pentenal-5-phosphate from a substrate containing 5
-incised
AP sites occurred via a
-elimination reaction, as indicated by
reaction of the released sugar-phosphate products with sodium
thioglycolate. The reaction for the removal of
4-hydroxy-2-pentenal-5-phosphate from the substrate containing
3
-incised AP sites was dependent on the presence of the
Mg2+ cation. These findings suggest that the S3 ribosomal
protein may function in several steps of the DNA base excision repair pathway in eukaryotes and may represent an important DNA repair function for the repair of oxidative and ionizing radiation-induced DNA
damage.
Cellular DNA is constantly exposed to both exogenous and
endogenous agents that can produce oxidative base damage (1). To remove
this damage and restore the integrity of the DNA, specific DNA repair
pathways have evolved (1, 2). The major pathway for the removal of
oxidative base damage is the DNA base excision repair pathway, found in
prokaryotes and eukaryotes (3). In this pathway, oxidized DNA bases are
removed by specific DNA glycosylases, leaving apurinic/apyrimidinic
(AP)1 sites in the DNA (1, 4). These AP
sites are then acted on by either a lyase activity that cleaves the DNA
via a -elimination reaction leaving a 3
unsaturated sugar-phosphate
product (4-hydroxy-2-pentenal-5-phosphate) or an endonuclease that
leaves a 5
deoxyribose-phosphate product (1, 4). The sugar-phosphate
products are removed by deoxyribophosphodiesterase (dRpase) activities
that have been identified in bacteria (5-7) and have also been shown
to exist in eukaryotic cells (8, 9). Removal of the
sugar-phosphate products allows the subsequent restoration of
nucleotides by DNA polymerases and the re-formation of intact
phosphodiester bonds by DNA ligases.
Several of the enzymes involved in DNA base excision repair have been
found to be multifunctional proteins. For example, the Escherichia coli Fpg protein, the product of the
mutM gene, is a DNA glycosylase that removes 8-oxoguanine
and formamidopyrimidines from DNA and also has an associated AP lyase
activity as well as an activity that excises 2-deoxyribose-5-phosphate
at 5-incised AP sites via a
-elimination reaction (7, 10).
Recently, the Drosphila ribosomal protein S3 has been shown
to have an enzymatic activity that cleaves DNA containing 8-oxoguanine
residues and an associated AP lyase activity that cleaves
phosphodiester bonds via a
,
-elimination reaction (11). This
protein was able to rescue the H2O2 sensitivity
of an E. coli mutM strain and completely abolish the mutator
phenotype of mutM caused by 8-oxoguanine-mediated G
T
transversions (11).
The similarity in DNA repair activities of the E. coli Fpg
protein and the Drosophila S3 protein suggested the
possibility that the S3 protein may contain an integral dRpase
activity. We investigated the dRpase activities of a glutathione
S-transferase fusion protein of S3 (GST-S3) on DNA
substrates containing either 5- or 3
-incised AP sites. The GST-S3
fusion protein, and not GST alone, was able to remove sugar-phosphate
products from both types of substrates. An unsaturated sugar-phosphate
product is released from the substrate containing 5
-incised AP sites
by
-elimination. The removal of
trans-4-hydroxy-2-pentenal-5-phosphate from the substrate
containing 3
-incised AP sites is via a hydrolytic reaction requiring
Mg2+. These S3-associated dRpase activities suggest that
this ribosomal protein may be active in several steps of the DNA base
excision repair pathway.
Bacteriophage M13mp18 single-stranded
DNA and a M13 24-mer sequencing primer (47) were purchased from U. S. Biochemical Corp. The large fragment (Klenow) of DNA polymerase I was
purchased from Boehringer Mannheim. E. coli uracil-DNA
glycosylase was purchased from Epicentre Technologies, and endonuclease
IV was prepared as described previously (12). Endonuclease III was a
generous gift of Dr. Richard Cunningham (State University of New York, Albany, NY). Sodium thioglycolate and spermidine were obtained from
Sigma.
Drosophila ribosomal gene S3 was overexpressed
as a GST fusion construct in a bacterial strain
(mutM::mini-tet) (13) that is defective
for one of the major EDTA-resistant activities in E. coli
for the removal of DNA 5-terminal deoxyribose phosphates (7).
The conditions for overexpression and purification of GST and GST-S3
were identical to those reported previously (11).
A M13
DNA substrate containing 33P-labeled AP sites was prepared
essentially as described previously (7, 14).
[-33P]dUTP was prepared by deamination of
[
-33P]dCTP (DuPont; 2000 Ci/mmol) as described
previously (15) and incorporated into DNA in a reaction (100 µl)
containing 1 µg of single-stranded M13mp18 DNA, 30 ng of 24-mer
primer, 2.5 nmol of dATP, dCTP, and dGTP, 0.25 nmol of dTTP, 0.15 nmol
of dUTP, 15 µCi of [
-33P]dUTP, 50 mM
Tris-HCl, pH 7.8, 10 mM MgCl2, 1 mM
dithiothreitol, and 3 units of large fragment (Klenow) of DNA
polymerase I. After incubation at 37 °C for 2 h, the reaction
was supplemented with an additional 2.5 nmol of dATP, dTTP, dGTP, and
dCTP and 3 units of large fragment (Klenow) of DNA polymerase I, and
incubation was continued for 1 h. The reaction was stopped by the
addition of 4 µl of 0.5 M Na2EDTA, followed
by incubation at 65 °C for 5 min. The DNA was purified from
unincorporated nucleotides on a 1-ml Sephadex G-50 column, precipitated
with ethanol, and then lyophilized. The
[33P]dUMP-containing M13 DNA was treated with uracil-DNA
glycosylase and subsequently treated with either endonuclease IV to
create a substrate containing 5
-incised AP sites or with endonuclease III to create a substrate containing 3
-incised AP sites as described previously (5).
DNA dRpase activity was assayed in a reaction
measuring either the release of 2-deoxyribose-5-phosphate from a
M13mp18 DNA substrate containing 5-incised AP sites or
4-hydroxy-2-pentenal-5-phosphate from a M13mp18 DNA substrate
containing 3
-incised AP sites. A typical reaction (50 µl) contained
220 fmol of M13mp18 DNA substrate containing incised AP sites, 100 ng
of GST-S3 enzyme, 30 mM Hepes-KOH, pH 7.4, 50 mM KCl, 1 µg/ml bovine serum albumin, 1 mM
dithiothreitol, 0.05% Triton X-100, and 0.1 mM
Na2EDTA. Some reactions were supplemented with 5 mM MgCl2. Release of sugar-phosphate products
was determined either by precipitation with trichloroacetic acid in the
presence of Norit charcoal or by anion exchange HPLC, as described
previously (5, 14).
M13mp18 double-stranded
DNA substrate containing 5-incised AP sites (220 fmol; 50,000 cpm) was
incubated with either 100 ng of GST-S3 enzyme or 2 mM
spermidine in the absence of enzyme in a 50-µl reaction containing
100 mM sodium thioglycolate, 30 mM Hepes-KOH,
pH 7.4, 50 mM KCl, 1 µg/ml bovine serum albumin, 1 mM dithiothreitol, 0.05% Triton X-100, and 0.1 mM Na2EDTA for 30 min at 37 °C. The reaction
mixture was then injected onto a Brownlee MPLC AX cartridge column (4.6 mm × 3 cm), which was eluted with 6 ml of 25 mM
KH2PO4, pH 4.5, followed by a 1-ml linear
gradient to 250 mM KH2PO4, pH 4.5, followed by 15 ml of 250 mM KH2PO4, pH 4.5, at a flow rate of 1 ml/min. Fractions (0.5 ml) were collected, and the radioactivity contained in each fraction was determined by
liquid scintillation counting.
To examine dRpase activities associated with the Drosophila ribosomal protein S3, the protein was purified as a GST fusion product. The purity of this enzyme was >95%. To eliminate any possible contamination with the E. coli Fpg protein, the GST-S3 was isolated from an E. coli mutM strain (11, 13).
To determine whether GST-S3 was capable of removing a 5-terminal
deoxyribose-phosphate product, the enzyme was incubated with a M13mp18
DNA substrate containing 5
-incised AP sites produced by treatment of a
[33P]dUMP-containing substrate with uracil-DNA
glycosylase and E. coli endonuclease IV. As shown in Fig.
1A, a dRpase activity acting on a 5
-incised
AP site will release either 2-deoxyribose-5-phosphate by a hydrolytic
mechanism or trans-4-hydroxy-2-pentenal-5-phosphate by a
-elimination reaction. As seen in the time course in Fig. 2, efficient release of sugar-phosphate was catalyzed by
GST-S3. This reaction was not dependent on the addition of 5 mM MgCl2. Reaction of 0.2 pmol of substrate
with 100 ng of purified GST did not result in the release of
sugar-phosphate (data not shown). The apparent Km
for the release of sugar-phosphate was determined by Lineweaver-Burk
analysis, as shown in Fig. 3, and found to have a value
of 0.07 µM; the value of Vmax is
0.015 pmol/min. The turnover number (kcat) was
calculated to be 0.13/min using an enzyme preparation that was 36 h old. It has been previously demonstrated that GST-S3 loses activity
rapidly upon storage (15-fold decrease within 72 h after
purification) (11).
Mechanism of Release of the Sugar-Phosphate Product from a DNA Substrate Containing 5
The release of
sugar-phosphate from a DNA substrate containing 5-incised AP sites by
the action of GST-S3 was not dependent on the addition of
MgCl2 and suggested that this reaction may proceed via a
-elimination reaction. The E. coli Fpg protein removes
5
-terminal deoxyribose phosphate groups by such a mechanism (7). To
determine whether the release of sugar-phosphate occurred via
-elimination, GST-S3 was incubated with the M13 DNA substrate containing 5
-incised AP sites in the presence of sodium thioglycolate. It has been shown previously that this reagent will react with an
unsaturated sugar-phosphate product, resulting in the formation of
products with altered mobilities on anion exchange chromatography (7,
8). When the M13 DNA substrate containing 5
-incised AP sites was
incubated with sodium thioglycolate and the
-elimination catalyst
spermidine (2 mM), the unsaturated sugar-phosphate product released by spermidine treatment formed products that reacted with
sodium thioglycolate, as seen in Fig. 4A.
When the DNA substrate was incubated with sodium thioglycolate and
GST-S3, the released products co-migrated with the products produced by
spermidine treatment, as seen in Fig. 4B. These results
strongly suggest that the mechanism of release of the 5
-terminal
deoxyribose-phosphate by GST-S3 is via a
-elimination reaction.
GST-S3 Removes Sugar-Phosphate Products from a DNA Substrate Containing 3
To determine whether GST-S3 was
capable of removing a 3-terminal sugar-phosphate product
(trans-4-hydroxy-2-pentenal-5-phosphate) as shown in Fig.
1B, the enzyme was incubated with a M13mp18 DNA substrate
containing 3
-incised AP sites produced by treatment of a
[33P]dUMP-containing substrate with uracil-DNA
glycosylase and E. coli endonuclease III, a known
-elimination catalyst (AP lyase). As seen in the time course in Fig.
5, the product
trans-4-hydroxy-2-pentenal-5-phosphate was released from the
substrate by GST-S3. Product identity was verified by anion exchange
HPLC as shown in Fig. 6. This reaction was absolutely
dependent on the addition of 5 mM MgCl2.
Reaction of 0.2 pmol of substrate with 100 ng of purified GST alone did not result in the release of sugar-phosphate product (data not shown).
The apparent Km for the release of
trans-4-hydroxy-2-pentenal-5-phosphate was determined by
Lineweaver-Burk analysis, as shown in Fig. 7, and found
to have a value of 0.075 µM; the value of
Vmax is 0.012 pmol/min. The turnover number
(kcat) was calculated to be 0.08/min using an
enzyme preparation that was 36 h old.
The results presented here demonstrate that Drosophila
ribosomal protein S3 is able to remove preincised AP sites that exist either 3 or 5
to the initial incision event. Similar to the Fpg
protein (7) but not to other dRpase-like activities in E. coli (5, 6), S3 is shown to remove incised 5
AP sites by a
-elimination reaction. Interestingly, Fpg is the only other purified
activity that also shares with S3 the ability of S3 to act on
8-oxoguanine residues in DNA that is subsequently followed in
vitro by the formation of a
,
-elimination product. Unlike Fpg, however, is the ability of S3 to also remove 3
-incised AP sites
through a Mg2+- dependent hydrolytic mechanism. Whereas Fpg
lacks this activity,2 the removal of 3
AP
sites nevertheless has been observed for another dRpase that exists in
E. coli (5) that has a Km value (0.06 µM) for the release of this product very close to that
determined for S3 (0.075 µM). The in vitro
ability of S3 to liberate an AP site existing at a 3
terminus is also
shared by 5
-acting AP endonucleases present in both prokaryotes and eukaryotes (17, 18). That S3 is able to carry this reaction out
in vivo as well as in vitro comes from studies
examining the rescue of the methyl methane sulfonate sensitivity of a
bacterial strain lacking the major 5
-acting AP endonucleases (RPC501), where Drosophila S3 was able to significantly protect RPC501
from cell killing (11). As an explanation of this result, it was noted
that once S3 catalyzed a
-elimination reaction at an AP site formed
by methyl methane sulfonate damage to DNA, the ribosomal protein
dissociated from the DNA and on a second encounter cleaved 5
to the
abasic site by
-elimination, thus leaving a 3
phosphoryl group that
was hypothesized to be less deleterious to the cell than a 3
-modified
deoxyribose produced by
-elimination. However, based on the results
presented here, it is conceivable that in the presence of endogenous
Mg2+, on the second encounter S3 completely released the
existing 4-hydroxy-2-pentenal-5-phosphate, thus leaving a
one-nucleotide gap and an efficient 3
terminus for DNA polymerase I to
fill.
In eukaryotic organisms, recent results indicate that rat DNA
polymerase can catalyze the release of 5
-incised AP sites (9). It
is unknown whether the same is true for the Drosophila
polymerase, but if so, it would suggest that S3 would only act in a
back-up role for the removal of 5
deoxyribose-phosphate, or it is
possible that S3 has a more important role in the proliferating cell
nuclear antigen-dependent abasic site repair using DNA
polymerase
(19).
We have been examining another Drosophila ribosomal protein, PO, that also contains DNase activity, but unlike S3, possesses only modest AP endonuclease activity. However, similar to S3 is the ability of PO to significantly reverse the sensitivity of E. coli RPC501 to MMS (20). Thus, both ribosomal proteins are capable of acting as DNA repair proteins in vivo although it remains to be seen whether, in fact, this is the case in Drosophila. It should be noted that both PO (20) and S3 (21) are associated with the nuclear matrix in Drosophila cells, therefore suggesting roles that are distinct from protein translation.
In conclusion, it seems that Drosophila S3 has a broader
substrate specificity toward the removal of an AP site than any
eukaryotic protein thus far characterized. Additionally,
Drosophila S3 is capable of acting on DNA containing
formamidopyrimidine (FapyGua) and 8-oxoguanine residues. Taken
together, the combination of these activities could result in a
one-nucleotide gap by processing oxidatively damaged bases by a
concerted N-glycosylase/AP lyase activity, followed by S3
dissociating from the substrate, and, in the presence of
Mg2+, subsequently removing the remaining
4-hydroxy-2-pentenal-5-phosphate. The combination of these activities
would suggest a more efficient means for preparing the DNA for gap
filling, as opposed to recruiting a 5-acting AP endonuclease to remove
the blocked 3
termini in preparation for polymerase
.