BIOLOGY OF REPRODUCTION 57, 1367-1374 (1997)
Role of Deoxyribonucleic Acid Polymerase E in Spermatogenesis in Mice
Dia Kamel, 3 Zachary B. Mackey, 4 Tiina Sjoblom, 5 Christi A. Walter, 6 John R. McCarrey, 7 Lahja Uitto, 3
Heidi Palosaari, 3 Jaana Lahdetie, s Alan E.Tomkinson, 4 and Juhani E. Syvioja2 ,3
Biocenter Oulu and Department of Biochemistry, 3 University of Oulu, FIN-90570 Oulu, Finland
Department of Molecular Medicine,4 Institute of Biotechnology, The University of Texas Health Science Center at
San Antonio, San Antonio, Texas 78245
Department of Medical Genetics and Center for Reproductive and Developmental Medicine,5 University of Turku,
FIN-20520 Turku, Finland
Department of Cellular and Structural Biology, 6 The University of Texas Health Science Center at San Antonio,
San Antonio, Texas 78284
Department of Genetics,7 Southwest Foundation for Biomedical Research, San Antonio, Texas 78228
ABSTRACT
Previous studies on DNA polymerase e indicate that this enzyme is involved in replication of chromosomal DNA. In this
study, we examined the expression of DNA polymerases a, 8,
and e during mouse testis development and germ cell differentiation. The steady-state levels of mRNAs encoding DNA polymerase e and the recombination enzyme Rad51 remained constant during testis development, whereas the mRNA levels of
DNA polymerases a and 8 declined from birth until sexual maturity. Immunohistochemical staining methods, using a stagespecific model of the seminiferous epithelium, revealed dramatic differences between DNA polymerase a and distribution.
As expected, DNA polymerase a and proliferating cell nuclear
antigen showed relatively strong immunostaining in mitotically
proliferating spermatogonia and even stronger staining in preleptotene cells undergoing meiotic DNA replication. The distribution of Rad51 was similar, but there was a dramatic peak in
late pachytene cells. In contrast, DNA polymerase was detectable in mitotically proliferating spermatogonia but not in the
early stages of meiotic prophase. However, DNA polymerase e
reappeared in late pachytene cells and remained through the
two meiotic divisions, and was present in haploid spermatids up
to the stage at which the flagellum starts developing. Overall,
the results suggest that DNA polymerase e functions in mitotic
replication, in the completion of recombination in late pachytene cells, and in repair of DNA damage in round spermatids.
In contrast, DNA polymerases a and appear to be involved in
meiotic DNA synthesis, which occurs early in meiotic prophase,
in addition to functioning in DNA replication in proliferating
spermatogonia.
INTRODUCTION
Five distinct DNA polymerases, designated o, , y, 8,
and , have been identified in mammalian cell extracts [1].
DNA polymerases or, 13,6, and e are located in the nucleus,
whereas DNA polymerase y is predominantly a mitochondrial enzyme. Three of the nuclear DNA polymerases---,
8, and E-belong to a group of a-like DNA polymerases
Accepted August 7, 1997.
Received May 13, 1997.
'This work was supported by grants from the National Research Councils for the Natural Sciences U.E.S.) and Health Sciences (J.L.), Academy
of Finland, and grants HD23126 .R.M), E505798 (C.A.W.), CA61335
(C.A.W.), and GM47251 (A.E.T.) from the United States Department of
Health and Human Services. Z.B.M. was supported by the Training Program inthe Molecular Basis of Breast Cancer CDAMD17-94-J-4147 from
the 2Department of the Army.
Correspondence. FAX: 358-8-553 1141; e-mail:
[email protected]
that share many structural and catalytic properties [1]. The
homologues of these enzymes in yeast are all essential for
cell viability and nuclear DNA replication [2-5]. The other
nuclear enzyme, DNA polymerase , functions in base excision repair [6, 7]. In contrast, DNA polymerase y is required for the replication of mitochondrial DNA [8].
DNA polymerases 3 and are sufficient for replication
of the circular genome of the mammalian virus SV40 in
vitro [9-11] and apparently also in vivo [12]. In this model
system, DNA polymerase a-primase is responsible for the
synthesis of short RNA-DNA precursor chains [13] at the
replication origin that prime synthesis of both the leading
and lagging strands. These RNA-DNA precursors are extended by DNA polymerase
[13-15]. The switch from
DNA primer synthesis to DNA synthesis by DNA polymerase depends on replication factor C and proliferating
cell nuclear antigen (PCNA), which replace the DNA polymerase a-primase complex with DNA polymerase [10].
Consequently, DNA polymerase
is responsible for extending RNA-DNA precursors into Okazaki fragments during the discontinuous synthesis of the lagging strand and
the processive elongation of the leading strand. The components that are essential for the complete in vitro replication of a plasmid containing SV40 origin of replication
have been identified, and the minimal system has been reconstituted from purified proteins [16].
The structure and catalytic properties of human DNA
polymerase
have been thoroughly studied [17-20]. Although this enzyme does not seem to be involved in the
replication of SV40 DNA, a recent study showing crosslinking of DNA polymerases ao, , and to nascent cellular
DNA in replicating chromosomes suggests that DNA polymerase plays a role in the replication of cellular DNA or
in a process closely associated with replication [12]. This
observation is consistent with the proliferation-dependent
expression of human DNA polymerase [21]. DNA polymerase has also been implicated in repair of UV damage
in human fibroblasts [22]. Furthermore, the carboxyl-terminal domain of Saccharomyces cerevisiae DNA polymerase appears to act as a sensor of DNA replication progression, coordinating the transcriptional and cell cycle responses to blockage of DNA replication [23, 24].
Spermatogenesis consists of three main phases: mitotic
proliferation, meiosis, and spermiogenesis. Cells at different stages within these phases can be identified by histochemical methods. Furthermore, essentially pure populations of many different spermatogenic cell types can be
isolated. Thus, spermatogenesis provides an in vivo model
1367
1368
KAMEL ET AL.
FIG. 1. Tissue distribution of mouse DNA polymerase EmRNA. RT-PCR
was performed on total RNA samples isolated from tissues as described
in Materials and Methods. PCR products were separated by PAGE and
detected by autoradiography. The 333-bp PCR products amplified by
primers specific to mouse DNA polymerase E cDNA are indicated. 3Actin was included as a control, and the 514-bp PCR product was amplified using specific primers.
for assessing the involvement of proteins in DNA replication, in meiotic recombination, and in DNA repair processes that occur during spermatogenesis. In order to assess the
role of DNA polymerase E, we studied the occurrence of
its mRNA and immunoreactive protein during spermatogenesis and compared this pattern with those of DNA polymerase , PCNA, and RadS1. The results of these studies
suggest that DNA polymerase
is not only involved in
DNA replication but also contributes to the completion of
meiotic recombination and postmeiotic DNA repair in haploid cells.
MATERIALS AND METHODS
Antibodies
Mouse monoclonal antibodies G1A and H3B against the
catalytic subunit of human DNA polymerase have been
described earlier [25]. Mouse monoclonal antibody
ICT1026195 against the catalytic subunit of human DNA
polymerase a and a polyclonal rabbit antibody against
Rad51 protein were kindly donated by Dr. Heinz-Peter Nasheuer (Institut fur Molekulare Biotechnologie e.V., Jena,
Germany) and by Dr. Efim Golub (Yale University, New
Haven, CT), respectively. A PCNA monoclonal antibody,
PC10, was purchased from Boehringer Mannheim (Mannheim, Germany).
Immunohistochemical Staining of Paraffin-Embedded
Sections
Testes from Balb C mice were fixed in 10% neutral buffered formalin for 6-24 h and embedded in paraffin. Sections (5 jim thick) were cut and mounted on poly-L-lysine
coated slides. Deparaffinized sections were first incubated
in 0.3% H2 0 2 in methanol for 30 min to block endogenous
peroxidase. After rehydration and three washes in Tris-buffered saline (TBS; 50 mM Tris-HC1, pH 7.4, 150 mM
NaCl) of 5 min each, the sections were treated in a micro-
FIG. 2. Steady-state levels of DNA polymerases (Pol) uc,8, and E and
Rad51 mRNAs during mouse testis development. The levels of mRNAs
for the proteins in the testes of 5-, 15-, 25-, and 83-day-old mice were
determined by RT-PCR with poly(A)+ RNA samples isolated from the testes
of mice of these ages as described in Materials and Methods. PCR products were separated by PAGE and detected by autoradiography. The 333bp PCR product was amplified using primers specific to DNA polymerase
E,the 546-bp PCR product using primers specific to DNA polymerase a,
the 1056-bp PCR product using primers specific to DNA polymerase 8,
and the 711-bp PCR product using primers specific to Rad51. -Actin
was included as a control, and the 514-bp PCR product was amplified
using specific primers.
wave oven twice for 5 min in 10 mM citrate buffer (pH 6).
After incubation at room temperature for at least 20 min,
slides were washed three times in TBS. Nonspecific binding
was blocked by incubating the slides in 1.5% normal serum
for 30 min. The primary antibody diluted in 0.1% BSA was
applied to the slides, which were then mounted with a coverslip and incubated overnight at 4C or for 1 h at room
temperature in a humidified chamber.
After incubation, the slides were kept at room temperature for 30 min, washed three times in TBS, and then incubated for 30 min at room temperature with a biotinylated
secondary antibody (Sigma, St. Louis, MO). After washing
with TBS, the distribution of antibodies in the sections was
detected by using the Vectastain Kit (Vector Laboratories,
Burlingame, CA) as recommended by the manufacturer.
The slides were then washed with TBS before being counterstained with periodic acid-Schiff (PAS) to visualize the
development of the acrosome in the spermatids and to permit identification of cells in different stages of the epithelial
cycle [26]. Finally, the slides were cleared, dehydrated, and
mounted.
The monoclonal antibody H3B was used for immunohistochemical staining of DNA polymerase . A similar but
less intensive staining pattern of seminiferous tubules was
obtained when the antibody G1A was used instead. Since
the antibodies H3B and G1A were raised against distinct
regions of DNA polymerase E [25], the similar staining patterns indicate the specificity of binding. When the antibody
H3B was first mixed with the excess of the recombinant
polypeptide that was used to raise it, no staining occurred.
1369
ROLE OF DNA POLYMERASE E IN SPERMATOZOA
Pol ax
c
0
1.0
0.8
I.
FIG. 3. Western analysis of DNA polymerase
catalytic polypeptide
from extracts of the testes of 5-, 15-, 25-, and 60-day-old mice. The extracts were prepared and Western analysis performed with a monoclonal
antibody G1A 125] as described in Materials and Methods.
0
S
0o.
U
0.4
ccS
0.2
Scoring of Slides
Immunoreactivity in different types of spermatogenic
cells was evaluated semiquantitatively in the following
way: - no immunoreactivity or only background staining,
+ weak immunoreactivity, ++ strong immunoreactivity,
+ + + very strong immunoreactivity. The stages of the seminiferous epithelium (I-XII) and the cell types in the sections counterstained with PAS were identified according to
the criteria of Oakberg [26] and Russell et al. [27].
Western Analysis
Protein samples were separated by SDS-PAGE and
transferred onto Immobilon P filters (Millipore, Bedford,
MA). The filters were washed with TBS, blocked by incubation with 5% nonfat milk in TBS for 1 h, and then
incubated overnight with a primary antibody diluted in TBS
supplemented with 0.05% Tween-20. After being washed
with TBS, the blots were incubated for 2 h with goat antimouse IgG conjugated with alkaline phosphatase (Bio-Rad,
Richmond, CA; 1:3000 dilution in TBS/Tween solution),
washed with TBS, and then incubated with the colorimetric
reagents 5-bromo-4-chloro-3-indolyl phosphate and nitro
blue tetrazolium in 100 mM Tris-HCl (pH 9.5).
o0.0
m
0
U,
el
_
5;
L
_i
<
0
Z
Q
0
E
0.
c
wuJ
'
0
C
0
o
a
S
c
1.
S,
e
0
0
LU
."
0
cr
1.0
Rad51
C
o0
Wj
0.8-
X
0.6
-
0.4
Z
0.2
0.0
.
'
._
..
0
0)
N
0.
W
I
)
a
LU
c1
p:O~~~~~~~9
FIG. 4. Determination of DNA polymerases and E and Rad51 mRNA
levels by RT-PCR with total RNA samples isolated from different testis cell
types, as described in Materials and Methods. PCR products were separated by PAGE and detected by autoradiography. The 333-bp PCR product
was amplified using primers specific to DNA polymerase , the 546-bp
PCR product using primers specific to DNA polymerase , and the 711bp PCR product by primers specific to Rad51. -Actin was included as a
control, and the 514-bp PCR product was amplified by specific primers.
C
I ]
I
L~~UJ
WC
E~~c
FIG. 5. Graphic representation of the data shown in Figure 4. After
quantification of the PCR products by phosphorimage analysis, the values
obtained for each of the cDNA products were standardized relative to the
[3-actin cDNA product.
1370
KAMEL ET AL.
FIG. 6. Photographs of immunohistochemically stained and PAS-counterstained sections of mouse testis and different stages in cross sections of the
seminiferous tubules of A, B)DNA polymerase e, C, D) DNA polymerase a, E)Rad51, and F)PCNA. The stages were identified according to the criteria
of Oakberg [26] and Russell et al. [27]. Brown staining resulting from the Vectastain ABC kit indicates immunoreactivity, while the pink stain of the
PAS reaction visualizes acrosome development and is used for staging. Roman numerals inside the tubules indicate corresponding seminiferousepithelial
stages in the mouse. Sg, spermatogonia; p, preleptotene spermatocytes; Is, leptotene spermatocytes; ps, pachytene spermatocytes; st, round spermatids.
x200.
ROLE OF DNA POLYMERASE e IN SPERMATOZOA
1371
Enrichment of Specific Spermatogenic Cell Types
Enriched populations of germ cells were prepared by
StaPut gradient separation as described previously [28].
Germ cell preparations were > 85% homogeneous as determined by phase contrast microscopy.
Reverse Transcription-Polymerase Chain Reaction
(RT-PCR)
Poly(A+ ) was isolated from testis tissue and germ cell
populations as described [29]. For isolation of total RNA,
mouse tissues were lysed in guanidium isothiocyanate, and
the RNA was purified by cesium chloride centrifugation
[30]. RT-PCR was performed as described [28]. Primers for
mouse DNA polymerases (a, , , and Rad5 1 were
AGCCAGTAGAAAGGGTGGAACA
(forward) and
GGCACTTCTGCCGAGTATCTAA (reverse), GGCAGTTTCCACTGCAGACAT (forward) and AAAGGGAAGCGGTCCACCTTTA (reverse), CCACCTTCTCCAGGAGTCTGAA (forward) and GGCCTCCTCAGAGAATTCACCA (reverse), and GGGACAGTCATGGCTATGCAAA
(forward) and GCCCTGAGTAGTCTGTTCTGTA (reverse), respectively. Expected products were 546, 1056,
333, and 711 base pairs (bp) long, respectively. Mouse [3actin primers were from Stratagene (La Jolla, CA) and produce a product of 514 bp from cDNA but a larger product
from genomic DNA.
RESULTS
Expression of DNA Polymerase
mRNA in Mouse Tissues
Expression of DNA polymerase e catalytic subunit was
examined in various mouse tissues by RT-PCR. High
steady-state levels of DNA polymerase mRNA were detected in both the spleen and testis (Fig. 1), tissues that
contain proliferating cells. Interestingly, expression of DNA
polymerase was significantly higher in the testes of older
animals (> 2 yr old) than in the testes of sexually mature
younger animals (> 1 yr old). DNA polymerase mRNA
was detectable in all the other tissues after longer exposure
(data not shown), indicating that this gene is ubiquitously
expressed. A similar expression pattern was observed for
another DNA replication enzyme, DNA polymerase ot catalytic subunit (data not shown). The association of high
levels of DNA polymerase expression with proliferating
tissues is in agreement with our previous studies quantifying the levels of DNA polymerase E in different mouse
tissues by immunoblotting [21].
Expression of mRNAs Encoding the Catalytic Subunits of
DNA Polymerases a, 8, and e, and Rad51 in Developing
Mouse Testis
To further investigate the role of DNA polymerase in
the mammalian testis, we initially monitored DNA polymerase
expression as a function of testis development.
For comparison, we also examined the expression of the
replicative DNA polymerases, a and 8, and a recombination
enzyme, Rad51 (Fig. 2). As expected, DNA polymerases a
and 8 were highly expressed in the testes of young animals,
which contain a high proportion of proliferating cells. The
decline in the steady-state levels of these mRNAs with increasing age correlates with the declining contribution of
proliferating cells to the testis [31]. In contrast, the steadystate levels of DNA polymerase e mRNA remained essentially constant during testis development. A similar pattern
FIG. 7. Photographs of mouse testis sections stained with DNA polymerase E antibody and counterstained with PAS: higher magnification of
seminiferous tubules. A) Stage I showing strongly stained early spermatids
(small arrows) and a type A4 spermatogonium (short arrow). A pachytene
spermatocyte shows only weak background staining (long arrow). B)Stage
X showing strongly stained nuclei of late pachytene spermatocytes (long
arrows) and a type A2 spermatogonium (short arrow). Leptotene spermatocytes show no staining (small arrows). 1200.
of expression was observed for the RAD51 gene, which
encodes a recombination enzyme. To confirm that the
amount of polypeptide present correlated with mRNA expression levels, we quantified the amount of DNA polymerase in extracts from testes at the same stage of development by immunoblotting (Fig. 3) and found that the
level of DNA polymerase polypeptide correlated with the
level of DNA polymerase mRNA. The difference in the
expression pattern of DNA polymerase compared with
the expression patterns of DNA polymerases a and suggests that, although these enzymes may function together
during DNA replication, they have distinct roles in the later
stages of spermatogenesis.
Expression of mRNAs for DNA Polymerases a and
Rad51 in Enriched Testicular Cell Populations
and
In an attempt to understand the cellular basis for the
different patterns of gene expression observed during testis
1372
KAMEL ET AL.
FIG. 8. Schematic illustration of the intensities of immunohistochemical staining of A) DNA polymerase e, B) DNA polymerase oa, C) Rad51 protein,
and D) PCNA. Organization of the germ cells in defined cell associations (stages) of the cycle of the seminiferous epithelium in the mouse is shown
by vertical columns and indicated by Roman numerals I-XII according to the classification of Oakberg 126]. Immunoreactive cells are indicated as
follows: + weak staining, ++ strong staining, and +++ very strong staining. PI, preleptotene; L, leptotene; Z, zygotene; P, pachytene; Di, diplotene
spermatocytes; M, meiotic divisions. A,-A4 , In, and B are different types of spermatogonia.
development, we measured the steady-state levels of DNA
polymerase o, DNA polymerase e, and Rad51 mRNAs as
a function of germ cell differentiation (Figs. 4 and 5). Consistent with their roles in DNA replication, steady-state levels of DNA polymerase at and e mRNAs were elevated in
proliferating spermatogonia compared with the levels in
Sertoli cells, a nondividing somatic cell type within the
testis. As the cells entered meiotic prophase, there was a
significant increase in DNA polymerase oa expression at the
leptotene-zygotene stage, and then the level of DNA polymerase a expression remained relatively constant until the
round spermatid stage. The level of DNA polymerase e
expression was similar in spermatogonia and during the
early stages of meiotic prophase. In contrast to DNA polymerase at, an increase in DNA polymerase E expression did
not occur until the late pachytene stage of meiotic prophase.
Expression of the RAD51 gene, whose product presumably
functions in meiotic recombination, increased as the germ
cells progressed along the differentiation pathway, reaching
a peak at the late pachytene stage of meiotic prophase and
then declining.
Immunostaining of DNA Polymerases a and e, PCNA,
and Rad51 in Spermatogenic Cells of the Mouse
Seminiferous Epithelium
To determine whether steady-state levels of DNA polymerases a and e, and Rad51 mRNA levels correlate with
protein levels in differentiating germ cells, we examined
the distribution of these enzymes and PCNA within seminiferous tubules by immunohistochemistry (Figs. 6 and 7).
ROLE OF DNA POLYMERASE E IN SPERMATOZOA
A compilation of the data obtained from seminiferous tubules at different stages is shown in Figure 8.
DNA polymerase E showed clear immunostaining in the
type A 1-A 4 intermediate and type B spermatogonia, whereas no staining was detectable in preleptotene cells, the first
cell type of meiotic prophase, even though the number of
these cells close to the basal lamina was high. Cells in the
early stages of meiotic prophase (leptotene, zygotene, and
pachytene) also had no DNA polymerase e staining, but
weak staining was detectable in early pachytene spermatocytes at stage II of the seminiferous epithelial cycle. The
staining intensity increased as cells progressed through the
pachytene phase and was observed throughout the meiotic
divisions (stage XII) until the haploid phase. This staining
pattern correlates well with DNA polymerase e levels measured in enriched spermatogenic cell populations (see Figs.
4 and 5). DNA polymerase staining was absent at stage
IX, concomitant with the remodelling of the sperm nucleus.
The Sertoli cells in the epithelium did not show any clear
immunostaining, but the testosterone-producing Leydig
cells in the interstitium between the tubules showed weak
staining.
The distribution of DNA polymerase ta was quite different from that of DNA polymerase . As expected, DNA
polymerase ot immunoreactivity was observed in the mitotically dividing populations of the spermatogonial cell
types (A 1-A 4 , In, and B spermatogonia). However, unlike
DNA polymerase E, DNA polymerase a was present in the
preleptotene cells and became most intensive at stage VIII,
during which maximal DNA synthesis, often referred to as
meiotic DNA synthesis, takes place [32]. DNA polymerase
a staining remained at this high level during meiotic prophase until disappearing abruptly in mid pachytene spermatocytes at stage VIII. The staining pattern of PCNA was
very similar to that of DNA polymerase a. Haploid cells
lacked detectable DNA polymerase c and PCNA reactivity.
RadS1 immunoreactivity was detectable in mitotically
proliferating spermatogonia and increased in intensity in
preleptotene spermatocytes at stage VII. As cells progressed
through meiotic prophase, the staining intensity continued
to increase and remained high almost throughout the pachytene phase, disappearing only in the late pachytene spermatocytes at stages IX-X. No staining was observed at later
stages. For DNA polymerase a and RadS1, the mRNA levels correlated with protein expression during the early
stages of germ cell differentiation; in the latter stages the
mRNAs were detectable, but the protein products were not.
DISCUSSION
In this study, we examined the expression patterns of
replicative DNA polymerases, a DNA polymerase accessory protein PCNA, and a recombination enzyme Rad51
during testis development and germ cell differentiation. The
relative abundance of DNA polymerase a in proliferating
somatic and germ cells was expected, since it is the only
DNA polymerase in eukaryotic cells that has associated
subunit with primase activity. Interestingly, expression of
DNA polymerase aoincreases rapidly to a maximum level
early in meiotic prophase and then remains at this level for
the duration of meiotic prophase. The increase in DNA
polymerase a polypeptide levels is coincident with a process known as meiotic DNA synthesis. The role of meiotic
DNA synthesis in meiotic recombination is not well understood. The expression pattern for PCNA, which functions as a sliding clamp tethering either DNA polymerase
B or DNA polymerase
1373
to the DNA template, was similar
to that of DNA polymerase a and was consistent with that
found in a previous study [33]. Since the expression pattern
of DNA polymerase , but not DNA polymerase E, was
also the same as the DNA polymerase a expression pattern,
we suggest that DNA polymerase a, PCNA, and DNA
polymerase 8 function together in meiotic DNA synthesis,
as they do in SV40 DNA replication [16, 34, 35].
Since the expression pattern of DNA polymerase was
clearly different from that of the other DNA replication
proteins during both testis development and germ cell differentiation, it appears that DNA polymerase plays a distinct role in meiotic recombination. Elevated levels of DNA
polymerase mRNA and protein occur at the end of the
pachytene phase and in round spermatids, suggesting that
this enzyme may be involved in meiotic recombination
events and/or DNA repair mechanisms, in addition to its
role in DNA replication in proliferating spermatogonia.
Higher levels of DNA polymerase expression were detected in the testes of older animals. It is possible that the
relative proportion of round spermatids increases in older
animals or, alternatively, that higher levels of DNA damage
may occur in the round spermatids of older animals and
this increase in damage requires an increase in DNA repair
involving DNA polymerase E. In contrast, similar or reduced levels of expression of enzymes implicated in meiotic recombination like DNA ligase III [28] and Rad51, and
DNA polymerase a were detected in older animals (data
not shown). This may reflect a decrease in spermatogenesis
with age.
Although DNA polymerase e and recombination enzyme
Rad51 exhibit a similar expression pattern during testis development, the distribution of these proteins during germ
cell differentiation is clearly different. Expression of RadS 1
increased during meiotic prophase, reaching a maximum in
late pachytene spermatocytes. This distribution implicates
Rad51 in meiotic recombination, in which it presumably
functions in DNA strand exchange reactions [36]. Accurate
localization of mouse testis Rad51 in meiotic chromosomes
of spermatocytes revealed its presence throughout the
pachytene phase, and at lower intensities in the synaptonemal complex in diakinesis and the meiotic metaphase
[37, 38]. This suggests that Rad51 may be involved at several different steps in meiosis ranging from homology
searches in early prophase to metaphase I.
The appearance of DNA polymerase e in late pachytene
cells suggests that it may be involved in DNA synthesis
reactions required for the completion of meiotic recombination. Alternatively, or in addition, DNA polymerase
may contribute to the high DNA repair activity measured
by unscheduled DNA synthesis in round spermatids [39,
401. Although these cells also contain high levels of DNA
polymerase 3, which is required for base excision repair
[7, 41], it is more likely that DNA polymerase E is responsible for the unscheduled DNA synthesis repair because
DNA polymerase 3 synthesizes only small repair patches,
usually one nucleotide in length. Furthermore, DNA polymerase e has been implicated in the gap-filling step of nucleotide excision repair [22, 42]. It was also found to be a
component of a recombination complex that is able to repair deletions and double-strand breaks [43].
In conclusion, our results indicate that DNA polymerase
a, DNA polymerase 8, and DNA polymerase e function in
DNA replication occurring in spermatogenic cells. Differences in expression patterns during germ cell differentiation
suggest that DNA polymerases a and 8 function in meiotic
1374
KAMEL ET AL.
DNA synthesis, early in meiotic prophase, whereas DNA
polymerase E appears to be involved in the latter stages of
meiotic recombination, and in DNA repair in haploid round
spermatids.
ACKNOWLEDGMENTS
We thank William Ramos and Dominick Trolian for technical assistance, Helmut Pospiech for comments on the manuscript, and Heinz-Peter
Nasheuer and Efim Golub for the gifts of the antibodies against DNA
polymerase a and Rad5I, respectively.
REFERENCES
1. Wang TS-E Cellular DNA Polymerases. In: DePamphilis ML (ed.),
DNA Replication in Eukaryotic Cells. Cold Spring Harbor, NY: Cold
Spring Harbor Laboratory Press; 1996: 461-493.
2. Morrison A, Araki H, Clark A, Hamatake RK, Sugino A. A third
essential DNA polymerase in Saccharomyces cerevisiae. Cell 1990;
62:1143-1151.
3. Araki H, Ropp PA, Johnson AL, Johnston LH, Morrison A, Sugino
A. DNA polymerase II, the probable homolog of mammalian DNA
polymerase e, replicates chromosomal DNA in the yeast Saccharomyces cerevisiae. EMBO J 1992; 11:733-740.
4. Budd ME, Campbell JL. DNA polymerases 6 and E are required for
chromosomal replication in Saccharomyces cerevisiae. Mol Cell Biol
1993; 13:496-505.
5. Francesconi S, Park H, Wang TS-E Fission yeast with DNA polymerase 8 temperature-sensitive alleles exhibits cell division cycle phenotype. Nucleic Acids Res 1993; 21:3821-3828.
6. Fornace AJ Jr, Zmudzka B, Hollander C, Wilson SH. Induction of 3polymerase mRNA by DNA-damaging agents in chinese hamster ovary cells. Mol Cell Biol 1989; 9:851-853.
7. Sobol RW, Horton JK, Kiihn R, Gua H, Singhal RK, Pasad R, Rajewsky K, Wilson SH. Requirement of mammalian DNA polymerase3 in base-excision repair. Nature 1996; 379:183-186.
8. Foury E Cloning and sequencing of the nuclear gene MIPI encoding
the catalytic subunit of the yeast mitochondrial DNA polymerase. J
Biol Chem 1989; 264:20552-20560.
9. Nethanel T, Kaufmann G. Two DNA polymerases may be required
for synthesis of the lagging DNA strand of simian virus 40. J Virol
1990; 64:5912-5918.
10. Tsurimoto T, Melendy T, Stillman B. Sequential initiation of lagging
and leading strand synthesis by two DNA polymerase complexes at
the SV40 DNA replication origin. Nature 1990; 346:534-539.
11. Waga S, Stillman B. Anatomy of a DNA replication fork revealed by
reconstitution of SV 40 DNA replication in vitro. Nature 1994; 369:
207-212.
12. Zlotkin T, Kaufmann G, Jiang Y, Lee MYW, Uitto L, Syvaoja JE,
Dornreiter I, Fanning E, Nethanel T. DNA polymerase epsilon may
be dispensable for SV40- but not cellular DNA replication, EMBO J
1996; 15:2298-2305.
13. Nethanel T, Zlotkin T, Kaufmann G. Assembly of simian virus 40
Okazaki pieces from DNA primers is reversibly arrested by ATP depletion. J Virol 1992; 66:6634-6640.
14. Murakami Y, Wobbe CR, Weissbach L, Dean FB, Hurwitz J. Role of
DNA polymerase a in simian virus 40 DNA replication in vitro. Proc
Natl Acad Sci USA 1986; 83:2869-2873.
15. Prelich G, Stillman B. Coordinated leading and lagging strand synthesis during SV40 DNA replication in vitro requires PCNA. Cell
1988; 53:117-126.
16. Waga S, Bauer G, Stillman B. Reconstitution of complete SV40 DNA
replication with purified replication factors. J Biol Chem 1994; 269:
10923-10934.
17. Syvaoja JE, Linn S. Characterization of a large form of DNA polymerase 8 from HeLa cells that is insensitive to proliferating cell nuclear antigen. J Biol Chem 1989; 264:2489-2497.
18. Syvioja J, Suomensaari S, Nishida C, Goldsmith JS, Chui GSJ, Jain
S, Linn S. DNA polymerases ac, 8 and E: three distinct enzymes from
HeLa Cells. Proc Natl Acad Sci USA 1990; 87:6664-6668.
19. Kesti T, Syvaoja JE. Identification and tryptic cleavage of the catalytic
core of HeLa and calf thymus DNA polymerase e. J Biol Chem 1991;
266:6336-6341.
20. Kesti T, Frantti H, Syvaoja JE. Molecular cloning of the cDNA for
the catalytic subunit of human DNA polymerase epsilon. J Biol Chem
1993; 268:10238-10245.
21. Tuusa J, Uitto L, Syvaoja JE. Human DNA polymerase E is expressed
during cell proliferation in a manner characteristic of replicative DNA
polymerases. Nucleic Acids Res 1995; 23:2178-2183.
22. Nishida C, Reinhard P, Linn S. DNA repair synthesis in human fibroblasts requires DNA polymerase 6. J Biol Chem 1988; 263:501-510.
23. Navas TA, Zhou Z, Elledge SJ. DNA polymerase links the DNA
replication machinery to the S phase checkpoint. Cell 1995; 80:2939.
24. Navas TA, Sanchez Y, Elledge SJ. Rad9 and DNA polymerase e form
parallel sensory branches for transducing the DNA damage checkpoint
signal in Saccharomyces cerevisiae. Genes & Dev 1996; 10:26322643.
25. Uitto L, Halleen J, Hentunen T, Hoyhtya M, Syvaoja JE. Structural
relationship between DNA polymerase Eand e* and their occurrence
in eukaryotic cells. Nucleic Acids Res 1995; 23:244-247.
26. Oakberg EE A description of spermiogenesis in mouse and its use in
analysis of the cycle of the seminiferous epithelium and germ cell
renewal. Am J Anat 1956; 99:391-413.
27. Russell LD, Ettlin RA, Hikim APS, Cleggand ED. Histological and
Histopathological Evaluation of Testis. Clearwater, FL: Cache River
Press; 1990.
28. Mackey ZB, Ramos W, Levin DS, Walter CA, McCarrey JR, Tomkinson AE. An alternative splicing event, which occurs in mouse
pachytene spermatocytes, generates a form of DNA ligase III with
distinct biochemical properties that may function in meiotic recombination. Mol Cell Biol 1997; 17:989-998.
29. Walter CA, Lu J, Bhakta M, Zhou Z-Q, Thompson LH, McCarrey JR.
Testis and somatic XRCC I DNA repair gene expression. Somatic Cell
Mol Genet 1994; 20:451-461.
30. Glisin V, Crkvenjakov R, Byus C. Ribonucleic acid isolated by cesium
chloride centrifugation. Biochemistry 1974; 13:2633-2637.
31. Bellve AR, Cavicchia JC, Millette CF, O'Brien DA, Bhatnagar YM,
Dym M. Spermatogenic cells of the prepubertal mouse: isolation and
morphological characterization. J Cell Biol 1977; 74:68-85.
32. Monesi V. Autoradiographic study of DNA synthesis and the cell cycle in spermatogonia and spermatocytes of mouse testis using tritiated
thymidine. J Cell Biol 1962; 14:1-18.
33. Wrobel KH, Bickel D, Kujat R. Immunohistochemical study of seminiferous epithelium in adult bovine testis using monoclonal antibodies
against Ki-67 protein and proliferating cell nuclear antigen (PCNA).
Cell Tissue Res 1996; 283:191-201.
34. Weinberg DH, Kelly TJ. Requirement for two DNA polymerases in
the replication of simian virus DNA in vitro. Proc Natl Acad Sci USA
1989; 86:9742-9746.
35. Lee S-H, Pan Z-Q, Kwong AD, Burgers PMJ, Hurwitz J. Synthesis
of DNA by DNA polymerase E in vitro. J Biol Chem 1991; 266:
22707-22717.
36. Baumann P, Benson FE, West SC. Human RadS5 protein promotes
ATP-dependent homologous pairing and strand transfer reactions in
vitro. Cell 1996; 87:757-766.
37. Ashley T, Plug AW, Xu J, Solari AJ, Reddy G, Golub EI, Ward DC.
Dynamic changes in Rad51 distribution on chromatin during meiosis
in male and female vertebrates. Chromosoma 1995; 104:19-28.
38. Haaf T, Golub EI, Reddy G, Radding CM, Ward DC. Nuclear foci of
mammalian RadS 1 recombination protein in somatic cells after DNA
damage and its localization in synaptonemal complexes. Proc Natl
Acad Sci USA 1995; 92:2298-2302.
39. Liihdetie J, Kaukopuro S, Parvinen M. Genotoxic effects of ethyl
methanesulfonate and X-rays in different stages of rat spermatogenesis
studied by inhibition of DNA synthesis and induction of DNA repair
in vitro. Hereditas 1983; 99:269-278.
40. Inoue M, Kurihara T, Yamashita M, Tatsumi K. Effects of treatment
with methylmethanesulfonate during meiotic and postmeiotic stages
and maturation of spermatozoa in mice. Mutat Res 1993; 294:179186.
41. Alcivar AA, Hake LE, Hecht NB. DNA polymerase
and
poly(ADP)ribose polymerase mRNA are differentially expressed during the development of male germinal cells. Biol Reprod 1992; 46:
201-207.
42. Aboussekhra A, Biggerstaff M, Shivji MKK, Vilpo JA, Moncollin V,
Podust VN, Protic M, Hubscher U, Egly J-M, Wood RD. Mammalian
DNA nucleotide excision repair reconstituted with purified protein
components. Cell 1995; 80:859-868.
43. Jessberger R, Podust V, Hiibscher U, Berg P. A mammalian protein
complex that repairs double-strand breaks and deletions by recombination. J Biol Chem 1993; 268:15070-15079.