| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Centre de Recherche en Cancérologie de l’Université Laval,* Hôpital Hôtel-Dieu de Québec, Centre Hospitalier Universitaire de Quebéc (CHUQ), Québec; and Unité de Recherche en Génétique Humaine et Moléculaire, Hôpital Saint-François d’Assise, Centre Hospitalier Universitaire de Quebéc (CHUQ), Québec, Canada
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
hel/hel mice) were crossed to PARP-1-null mice. Both Wrnhel/hel and PARP-1-null/Wrnhel/hel cohorts developed more neoplasms than wild-type animals. The tumor spectrum was the same between PARP-1-null/Wrnhel/hel mice and Wrn mutants. However, PARP-1-null/Wrnhel/hel mice developed neoplasms at a younger age. Mouse embryonic fibroblasts derived from such PARP-1-null/Wrnhel/hel mice stop dividing abruptly unlike Wrnhel/hel or PARP-1-null cells. PARP-1-null/Wrnhel/hel fibroblasts were distinguished by an increased frequency of chromatid breaks, complex chromosomal rearrangements, and fragmentation. Finally, experiments have indicated that the PARP-1 enzyme co-immunoprecipitates with the WRN protein in human 293 embryonic kidney cells. These results suggest that Wrn and PARP-1 enzymes may be part of a complex involved in the processing of DNA breaks.
Human WS cells and murine cells with a mutation in the WRN gene homologue are known to be sensitive to drugs that damage DNA at replication forks such as topoisomerase inhibitors.11
These results point to a specific activity of this protein during DNA replication. In this respect, it is interesting to note that in WS fibroblasts the S phase and the whole cell cycle are prolonged.12
It has also been shown that the replication defect detected in WS lymphoblastoid cell lines is associated with an impaired S phase transit.13
At the molecular level the rate of initiation of DNA replication is retarded in WS cells compared to control cells.14,15
Finally, we have observed that the mouse Wrn protein co-purifies with the multiprotein DNA replication complex.11,16
In addition, the WRN protein physically interacts with major components of the DNA replication fork such as proliferation cellular nuclear antigen, replication protein A, topoisomerase I, the p50 small subunit of DNA polymerase 
,17
and flap endonuclease 1 (Fen1).18
The enzyme poly(ADP-ribose) polymerase-1 (PARP-1) is another component of the DNA replication complex.19
PARP-1 physically associates with the DNA polymerase 
-primase complex and is believed to bind to breaks on DNA damage during replication fork progression.20
It rapidly binds to single- or double-strand breaks through its N-terminal DNA-binding domain and uses NAD (ß-nicotinamide adenine dinucleotide) to synthesize poly(ADP-ribose) on a variety of proteins including p53, topoisomerases, histones, and PARP-1 itself.21,22
In addition, PARP-1 targets and regulates via poly(ADP-ribose) the functions of proteins involved in DNA damage checkpoints or DNA repair such as p53, DNA ligase III, DNA-PK, KU70, and XRCC1.21,23,24
Interestingly, p53, DNA-PK and the KU70/80 complex interact with WRN protein as well.17
Chemical or genetic abrogation of PARP-1 activity in cells leads to an increase in the frequency of sister chromatid exchanges and genomic instability25
especially after genotoxic stresses.22
A number of PARP-1 knockout mice have been created by several groups.22 Although, mice lacking a functional PARP-1 develop normally and are not cancer prone, they are hypersensitive to DNA damage.26,27 For example, PARP-1 knockout mice are highly susceptible to nitrosamine carcinogenicity compared to wild-type animals.28 In addition, inhibition of PARP-1 will accelerate tumorigenesis in p53-deficient mice demonstrating a role for PARP-1 in tumor development.29,30 Fibroblasts established from PARP-1 mutant mice have indicated that cells acquire a slower growth rate in culture compared to wild-type fibroblasts.31,32 In addition to a loss of proliferative capacity, PARP-1-null fibroblasts display increased telomere shortening compared to wild-type cells.33 An additional report has indicated that an increase in telomere shortening is only observed in late passage cells and not in primary embryonic tissue cultures.34 Although PARP-1 is cleaved very rapidly during apoptosis, PARP-1-null cells are not more prone to spontaneous apoptosis in vitro.32 It is believed that it is the increased genomic instability (DNA amplification and/or loss of chromatin) that contributes to the delay in cell proliferation in PARP-1 mutant cells.32,34 Finally, microarray analysis on PARP-1-null fibroblasts have indicated that loss of PARP-1 results in deregulation of genes that encode proteins implicated in cancer initiation or progression and in normal or premature aging.35
Because both WRN and PARP-1 proteins are found in the DNA replication complex, it is reasonable to ask whether they act in concert in some aspect of DNA metabolism. In this study, we show that human WRN and PARP-1 enzymes can be co-immunoprecipitated in vivo. The roles of a potential murine Wrn/PARP-1 complex in genomic stability were further investigated in mice. Mice with a deletion of part of the helicase domain of the Wrn gene11 were thus crossed to PARP-1-null mice31 to study the phenotype of double-homozygous mice at the physiological, cellular, and chromosomal levels.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Mice lacking part of the helicase domain of the Wrn gene and PARP-1-null mice were generated by homologous recombination as previously described.11,31
Mice of all possible genotypes were generated by mating homozygous Wrnhel/hel individuals with PARP-1-null mice and intercrossing the F1 and F2 generations to obtain all four desired genotypes. Homozygous animals of each genotype were then crossed to obtain the numbers of mice indicated in the figure legends. The genetic background of all of the animals were both 129/Sv and Black Swiss (129/Sv/Black Swiss genetic background). Mice were maintained under pathogen-free conditions and fed ad libitum with Teklad Global (Madison, WI) 18% protein rodent diet (5% fat). Animals (from littermates) were checked three times a week for any external mass, infection, bleeding, gasping, and overall decrease or change in activity or behavior. Animals that became immobile or moribund were then sacrificed for histological examination of their organs as described previously.36
Mice with skin lesions were sacrificed as soon as there was a sign of infection in the lesion.
Generation and maintenance of the embryonic fibroblasts has been described previously.37 Briefly, cells were maintained in low-glucose Dulbecco’s modified Eagle’s medium supplemented with 10% heat-inactivated calf serum at 37°C in an atmosphere of 5% CO2. Cell proliferation was determined by plating 5 x 104 cells in six-well plates. The cultures were maintained for up to 11 days with changing media every other day. Cells were counted with trypan blue on a hemocytometer. Data were plotted on a graph and cell growth was calculated from the slope of each curve (portion of the curve representing the logarithmic phase of the culture). Human 293 embryonic kidney cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum.
Protein Analysis
Protein extraction, immunoprecipitations, and Western blotting analyses were performed as described.16 Polyclonal antibodies against the C-terminus and the N-terminus portion of human WRN were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and Novus Biologicals (Littleton, CO), respectively. The monoclonal antibody against the N-terminus portion of human PARP-1 was purchased from Transduction Laboratories (Lexington, KY). The polyclonal antibody against the C-terminus portion of human PARP-1 was purchased from Santa Cruz Biotechnology. Finally, all horseradish peroxidase-conjugated secondary antibodies were purchased from Amersham. Proteins on the Western blots were visualized using an ECL kit (Amersham). All antibodies were used as indicated by the manufacturers.
Chromosome Analysis
For each genotype, primary mouse embryonic fibroblasts (MEFs at the indicated population doublings) derived from three embryos of at least two litters were cultured for chromosome analysis. Cells were arrested by adding colcemid to a final concentration of 40 ng/ml for 1 to 3 hours. Cells were then harvested, treated with a hypotonic solution (75 mmol/L KCl) at pH 8.0 for 30 minutes at 37°C, and fixed three times in cold Carnoy’s fixative (3 vol of methanol/1 vol of acetic acid) for 15 minutes.38 Cells were resuspended in fresh fixative and dropped onto glass slides in a 40% humidity chamber at 28°C (Thermotron Industries, Holland, MI). The slides were placed overnight on a warm plate at 60°C before performing chromosome banding. G-band patterns were generated using the GTW (G-bands using trypsin and Wright staining) banding technique. The banding was produced by treating the slides in 0.2x Enzar-T trypsin (Intergen) for 1 to 2 minutes, followed by one wash in pure ethanol, and finally by a 3-minute staining in 8% Wright solution. Chromosomes of 200 metaphases were counted for all four genotypes. To study chromosomal rearrangements, 100 metaphases were analyzed and 15 metaphases karyotyped for the wild-type MEFs.39 For the other genotypes, 150 metaphases were analyzed and 30 karyotyped. The results coming from the different embryos were pooled.
Primed in Situ Labeling
Primed in situ labeling was used to determine telomere length on chromosomes harvested from MEFs derived from two embryos per genotype at 10 to 12 population doublings.40-42
The slides were denatured in 10 mmol/L NaOH/1 mol/L NaCl for 30 seconds at room temperature and then dehydrated in ethanol (70%, 80%, and 100%) at 4°C. Primed in situ labeling reaction solution that contained 4 µl of each of 2.5 µmol/L dATP, dCTP, and 0.25 µmol/L dTTP, 1 µl of 1 mmol/L digoxigenin-11-dUTP (Roche Molecular Biochemicals), 4 µl of 7.5 µmol/L primer [(CCCTAA)7], 2.5 U Taq polymerase (Roche Molecular Biochemicals), 5 µl of 10x PCR buffer and distilled water for a total volume of 50 µl was added on the denatured slides, which were then covered with a coverslip. A single-step primer annealing and strand elongation was performed at 62.5°C for 10 minutes on a thermocycler (PTC-100 16MS slide bloc; MJ Research). At the end of the reaction, the slides were transferred into phosphate-buffered saline (PBS) at 62.5°C for 1 minute and in washing buffer (4x standard saline citrate, 0.2% Tween 20) at room temperature for 5 minutes. To visualize the digoxigenin-labeled DNA in situ, 50 µl of 1% anti-digoxigenin-rhodamine (Roche Molecular Biochemicals) was applied and the slides were covered with a coverslip. After incubation in a moist chamber at 37°C for 30 minutes, the slides were washed in washing buffer (4x standard saline citrate, 0.2% Tween 20) at room temperature for 5 minutes and then in PBS at room temperature for 5 minutes. Chromosome spreads were counterstained using 125 ng/ml of 4',6-diamino-2-phenylindole mixed with 1 mg/ml of 
-phenylenediamine (Sigma).
Image Analysis and Telomere Measurement Analysis
All slides were examined under a Leica DMRB fluorescence microscope equipped with a JAI M300 charge-coupled device camera with a Synchro4 interface board coupled to the Metasystems in situ imaging system (ISIS) software version 4.1.12 and quantified by digital image analysis using the telomere measurement software. Integration times were typically 0.04 second for the 4',6-diamino-2-phenylindole counter stain and 0.6 second for the rhodamine telomere signal. Optimization of the labeling reaction resulted in a mean telomere detection efficiency of 
98%. A good correlation between the values derived from sister chromatid telomere pairs was observed (data not shown). This observation suggests that the measured telomere fluorescence intensity values are directly related to telomere length.43
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Cellular fractionation experiments have indicated that both the murine Wrn and PARP-1 proteins co-purify with the multiprotein DNA replication complex.16,19
To determine whether PARP-1 directly interacts with the WRN enzyme, immunoprecipitation experiments were performed with several different antibodies against WRN or PARP-1 on human 293 embryonic kidney cells. Human cells were used in these experiments because our antibodies against the murine Wrn enzyme did not immunoprecipitate the mouse Wrn protein. A polyclonal antibody against the N-terminus region of the human WRN protein co-immunoprecipitated the PARP-1 enzyme as revealed by Western blotting (Figure 1)
. In contrast, antibodies against the C-terminus portion of human WRN did not immunoprecipitate the PARP-1 enzyme. In similar sets of experiments, the PARP-1 enzyme was immunoprecipitated with antibodies against either the N-terminus or the C-terminus portions of the enzyme. As shown in Figure 1
, both types of antibody immunoprecipitated PARP-1. However, stripping and reprobing of the blots with an antibody against the WRN protein revealed that WRN is detected in the immunoprecipitate only when the antibody against the C-terminus portion of PARP-1 is used (Figure 1)
. These experiments suggest a WRN/PARP-1 complex present in human 293 embryonic kidney cells.
|
To assess the joint role of PARP-1 and Wrn in specific mouse pathology including neoplasm progression, PARP-1-null and Wrnhel/hel mice were crossed to generate PARP-1-null/Wrnhel/hel animals. Littermates of all genotypes were then carefully followed throughout 2 years and scored for the occurrence of neoplasms or any signs of pathology. The average litter size for wild-type animals was 8.5 pups per litter (18 litters) and 6.9 pups per litter (17 litters) for the Wrnhel/hel mice. In contrast, the average litter size for the PARP-1-null mice was 4.8 pups per litter (26 litters) as described previously.22
Finally, the average litter size for PARP-1-null/Wrnhel/hel animals was 4.1 pups per litter (28 litters) which is not significantly different from PARP-1-null mice. This indicates that PARP-1-null/Wrnhel/hel survived embryogenesis. There was no significant difference in average body weight between mice of all four genotypes. As indicated in Table 1
, Wrnhel/hel mutant mice were remarkable with respect to the variety of illnesses and neoplasms they developed36
compared to wild-type animals. As described before,31,32
PARP-1-null mice developed few pathological phenotypes. These mice developed alopecia of the skin, epidermal hyperplasia, or infections. PARP-1-null animals with large skin lesions had to be sacrificed as such lesions became infected and were considered life threatening by the veterinarian. Skin lesions were detected in 17% of the PARP-1-null mice. Such a phenotype was not detected in wild-type or Wrnhel/hel mutant mice. Approximately 10% of the PARP-1-null/Wrnhel/hel mice developed skin lesions similar to PARP-1-null mice. Finally, Table 1
indicates that PARP-1-null/Wrnhel/hel mice could potentially develop any of the pathologies observed with wild-type, Wrnhel/hel, or PARP-1-null mice.
|
. At first glance, Figure 2A
indicates that PARP-1-null/Wrnhel/hel mice developed some types of pathology sooner than Wrnhel/hel mutant or wild-type animals. Log-rank tests on Figure 2A
indicate that the difference between disease-free PARP-1-null/Wrnhel/hel and Wrnhel/hel mutant mice with time is significant (P < 0.0008). The difference between Wrnhel/hel mutant mice and wild-type mice is also significant (P < 0.0001).
|
. Between 6 and 12 months of age, 21% of PARP-1-null/Wrnhel/hel and 17% of PARP-1-null mice were sick. All wild-type animals and 93% of Wrnhel/hel mutant mice were healthy. The difference between PARP-1-null/Wrnhel/hel or PARP-1-null mice and wild-type animals is significant (t test; P < 0.05). Between 12 and 18 months of age, 77% of PARP-1-null/Wrnhel/hel were sick. In contrast, 27% to 35% of wild-type, Wrnhel/hel mutant, and PARP-1-null mice were sick. The differences between PARP-1-null/Wrnhel/hel mice and animals form all of the other genotypes were significant (t-test; P < 0.05). Finally, all PARP-1-null/Wrnhel/hel mice and 88% of Wrnhel/hel mutant animals were sick by 24 months. In contrast, less than 42% of wild-type or PARP-1-null animals were sick. These results indicate that PARP-1-null/Wrnhel/hel mice develop several pathological symptoms sooner than animals of the other genotypes analyzed in this study. However, it takes at least 12 months to see a significant difference between PARP-1-null/Wrnhel/hel and Wrnhel/hel mice. Mutant Mice Develop Multiple Types of Neoplasms
As indicated above, PARP-1-null/Wrnhel/hel mice appeared sick at an earlier age than wild-type or Wrnhel/hel animals. This was so even though all animals were housed under the same conditions. Some of these animals showed a complex phenotype including infections or cardiac fibrosis in addition to neoplasms. In several instances, it was impossible to determine whether the observed cardiac fibrosis or infections were incidental to metabolic system or immune system disorders. As both Wrn and PARP-1 enzymes are involved in genomic stability, an important feature deregulated in cancer, we concentrated our next analyses on animals that were diagnosed with neoplasms. As shown in Table 1
, 27% of wild-type animals (9 of 33 mice) developed some type of hyperplasia or neoplasia at the age of 13 to 24 months. The most common proliferative lesions were either myeloid dysplasia or myeloid leukemia (overall, 12%). The mean age of wild-type mice with neoplasms is 22 months (Figure 3)
. As described before,31,32
PARP-1-null mice develop few pathological phenotypes by 24 months of age. Only 2 animals of 18 analyzed developed a neoplasm (Table 1)
. One PARP-1-null mouse had an hepatocellular carcinoma at 18 months of age and one PARP-1 mutant had developed a myeloid hyperplasia at 11 months of age. No statistical analysis was performed with this cohort (Figure 3)
.
|
hel/hel mutant mice were remarkable with respect to the variety of neoplasms they developed compared to wild-type animals (Table 1)
. Approximately, 45% of Wrnhel/hel mutant mice developed some types of neoplasm compared to 27% for wild-type mice and this difference was statistically significant (Table 1)
. Interestingly, PARP-1-null/Wrnhel/hel mice developed the same type of neoplasms as Wrnhel/hel mutant mice (Table 1)
. Approximately, 47% of PARP-1-null/Wrnhel/hel mice developed a neoplasm. Thus, PARP-1-null/Wrnhel/hel mice are not more cancer prone than Wrnhel/hel animals. However, these mice developed their neoplasms at a younger age than Wrnhel/hel mutant animals. The mean age of PARP-1-null/Wrnhel/hel mice with neoplasms is 14 months (Figure 3)
. Moreover, 18% of PARP-1-null/Wrnhel/hel mice developed more than two types of neoplasm, simultaneously. Only 9% of Wrnhel/hel mice developed more than two types of neoplasms. This difference between both cohorts is significant (chi-square test; P < 0.05 in Table 1
). Finally, the difference between the mean age of Wrnhel/hel and PARP-1-null/Wrnhel/hel animals with neoplasms is significant (Figure 3
; analysis of variance, P < 0.05). All these results suggest that a defect in the PARP-1 gene product accelerates the cancer phenotype seen in the Wrn mutant mice. Analysis of Mutant MEFs
Because reduced growth rate and premature senescence are properties generally associated with the premature aging of human WS fibroblasts,44
we examined this property in fibroblasts derived from PARP-1-null/Wrnhel/hel mice. MEFs from three embryos of each genotype were established as described previously.11
Previous data have indicated that Wrnhel/hel mutant and PARP-1-null cell lines acquired a slower growth rate than wild-type cell lines with the number of passage in culture.11,31
Interestingly, PARP-1-null/Wrnhel/hel MEFs already had a slow growth phenotype at the first passage (approximately five population doublings) in culture compared to cells of all other genotype. After the second passage in culture, it could take up to 8 weeks for the PARP-1-null/Wrnhel/hel MEFs to reach confluence. They stopped growing by the third passage (after 15 to 20 population doublings) in culture. At that point, such cultures could be maintained for several months without any evidence of cell division. Thus, in contrast to the other MEFs that are still growing after the 15 to 20 population doublings in culture, PARP-1-null/Wrnhel/hel MEFs senesce very rapidly. An example of growth curves is given in Figure 4
for MEFs of each genotype after the third passage in culture. Finally, fluorescence-activated cell sorting analyses have indicated that Wrnhel/hel, PARP-1-null, and PARP-1-null/Wrnhel/hel MEFs did not exhibit a significant increase in their apoptotic fraction with the number of passages in culture compared to wild-type cells (data not shown).
|
Chromosome analyses were performed on MEFs of all four genotypes to determine the chromosome composition of the cells at the beginning of the culture. There is an increase in aneuploidy with the number of population doubling in culture for all of the MEFs analyzed. However, as indicated in Table 2
, aneuploidy was already high for the Wrnhel/hel, the PARP-1-null, and the PARP-1-null/Wrnhel/hel MEFs compared to wild-type cells. After 5 to 10 population doublings, more than 70% of wild-type metaphase spreads exhibited a diploid content. In contrast, less than 50% of the metaphase spreads from the Wrnhel/hel, the PARP-1-null, and the PARP-1-null/Wrnhel/hel MEFs were diploid by the 10th population doubling. The differences between wild-type and every mutant MEFs at 5 to 10 population doublings were statistically significant (chi-square test; P < 0.005 or P < 0.001 in Table 2
). These results indicate that MEFs with a mutation either in the PARP-1, the Wrn, or both genes reached a high degree of aneuploidy within fewer cell divisions when compared to wild-type cells.
|
revealed all kinds of chromosomal abnormalities in the mutant cells, such as fragmentation, breaks, and end-to-end fusions including Robertsonian-like configurations, dicentric, and ring chromosomes (Figure 5)
. The sum of all chromosomal abnormalities was always the highest in PARP-1-null/Wrnhel/hel MEFs even early on in cultures (top part of Table 3
). The numbers of end-to-end fusions were similar between the Wrnhel/hel, the PARP-1-null, and the PARP-1-null/Wrnhel/hel MEFs (Table 3)
even after 10 population doublings. However, the number of chromatid breaks was strikingly higher in the PARP-1-null/Wrnhel/hel MEFs. An average of 1.17 fragment per metaphase was detected in these cells after only 10 population doublings. In contrast, an average of 0.02 fragment per metaphase was detected in wild-type cells after 10 population doublings. For the PARP-1-null and the Wrnhel/hel MEFs, 0.22 and 0.09 fragment per metaphase, respectively, was detected after more than 15 population doublings. The difference between PARP-1-null/Wrnhel/hel MEFs and either PARP-1-null or Wrnhel/hel cells is statistically significant (t-test, P < 0.0005). Triradial and quadriradial structures caused by chromatid translocations and double-minutes structure were also more frequent in PARP-1-null/Wrnhel/hel cells (Table 3)
. Importantly, these observations were always similar for all embryos of a specific genotype. Examples of chromosomal rearrangements from PARP-1-null/Wrnhel/hel cells are given in Figure 5
. All these observations indicate that in addition to the rapid increase in aneuploidy, there is also a rapid increase in the extent of chromosomal rearrangements in cells lacking both PARP-1 and Wrn helicase activities. This overall increase in genomic instability is likely to affect the PARP-1-null/Wrnhel/hel cell growth.
|
|
Telomeres are DNA structures composed of TTAGGG repeats required for the maintenance of chromosomal ends. They protect chromosomes from end-to-end fusion, recombination, and degradation. It is known that replicative decline of somatic cells is associated with loss of telomeric repeats.45,46
To determine whether loss of telomere length is responsible for the rapid senescence observed with the PARP-1-null/Wrnhel/hel MEFs, telomeric sequences were analyzed by semiquantitative primed in situ labeling (Figure 5I)
. The analyses were performed on metaphase spreads from PARP-1-null/Wrnhel/hel cells one passage before the senescence phenotype (at the 10th or 12th population doubling). These samples were compared to metaphase spreads from wild-type, Wrnhel/hel, and PARP-1-null cells at exactly the same population doubling. As described previously, PARP-1-null MEFs exhibit shorter telomeres compared to wild-type cells.33
The mean value of telomere fluorescence for the PARP-1-null cell is 53% of that of the wild-type cells (Figure 6)
. Wrnhel/hel MEFs also exhibit shorter telomeres than wild-type cells. Surprisingly, even though PARP-1-null/Wrnhel/hel MEFs senesce more rapidly than PARP-1-null or Wrnhel/hel cells, their mean value of telomere fluorescence is similar to the Wrnhel/hel fibroblasts (Figure 6)
. The mean values of telomere fluorescence for both the Wrnhel/hel and the PARP-1-null/Wrnhel/hel MEFs are 60 to 70% of that of the wild-type cells, respectively. Finally, telomerase activity was examined by using the telomere repeat amplification protocol assay. The levels of telomerase activity were the same in all of the MEFs examined (data not shown). Thus, a simple deletion in the helicase domain of the mouse Wrn protein affects the length of the telomeres. However, the results also indicate that the altered length of the telomeres seen in the PARP-1-null/Wrnhel/hel MEFs is not the only explanation for their abrupt arrest in cell growth.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Neoplasm Formation in Wrnhel/hel and PARP-1-Null/Wrnhel/hel Mice
Several different neoplasms and other pathological findings were observed in Wrnhel/hel mutant mice older than 18 months of age. It was found that 80% of the mice had developed some types of disease including hyperplasias or neoplasms in one of their organs by 24 months of age.36
Interestingly, the PARP-1-null/Wrnhel/hel developed similar types of neoplasms but at a younger age when compared to Wrnhel/hel animals. It is possible that the onset of the pathologies is the same between Wrnhel/hel and PARP-1-null/Wrnhel/hel animals, but the evolution of the pathologies is faster in PARP-1-null/Wrnhel/hel mice. Under such conditions, PARP-1-null/Wrnhel/hel mice appear sick sooner than Wrnhel/hel animals. Thus, it is concluded that a defect in the PARP-1 gene product accelerates the phenotype seen in the Wrn mutant mice.
It has been previously reported that a stable mutant Wrn protein is synthesized from the Wrnhel/hel mice used in this study.11,16
However, this mutant protein does not co-purify with the DNA replication complex16
indicating that the deletion will affect the three-dimensional structure of the whole protein and probably protein-protein interactions in vivo. Nonetheless, we cannot rule out the possibility that this mutant Wrn protein can interact and disrupt the function of other protein complexes involved in genomic stability. This is even more so as a Wrn-null mutation in mouse, which mimics the mutations seen in WS patients, does not reiterate all of the phenotypes described for our Wrnhel/hel mice.48
Importantly, our mutant mice synthesize a stable mutant protein with an intact exonuclease domain that may have abnormal activity affecting genomic stability. Purification and analysis of this mutant Wrn protein is currently under investigation to determine whether it has any functional exonuclease activity. Finally, although Wrnhel/hel mice demonstrate a complex phenotype, they do not recapitulate all of the symptoms described for WS patients. As such, Wrnhel/hel may not be a good model for WS. Nevertheless, our results clearly indicate that a mutation in the helicase domain of the Wrn protein will affect genomic stability.
Interestingly, PARP-1-null MEFs and Wrnhel/hel cells showed similar strong karyotypic instability in vitro and yet PARP-1-null mice did not develop neoplasms like the Wrnhel/hel animals. Preliminary karyotype analyses have indicated that there are some chromosomal rearrangements that can be detected in several PARP-1-null MEFs cultures that are different from those observed in either Wrnhel/hel or PARP-1-null/Wrnhel/hel cells. These preliminary analyses indicate that different sets of genes are mutated in cells of all three genotypes. Such differences in the spectrum of genomic mutation may be responsible for the different outcome observed in vivo for each genotype. Thorough spectral karyotyping analyses and comparative genomic hybridization experiments are underway to test this possibility.
Increased Gnomic Instability and Proliferative Decline in PARP-1-Null/Wrnhel/hel MEFs
It is known that replicative decline of somatic cells is associated with a loss of telomeric repeats and that short telomeres contribute to chromosomal instability in cells.45,46
All mutant MEFs exhibit a decrease in their telomere length compared to wild-type cells. Shorter telomeres might have caused the mutant cells to acquire a slow growth phenotype. However, the proportion of PARP-1-null/Wrnhel/hel chromosomes with very short telomeres is not greater than those from the Wrnhel/hel or PARP-1-null cells, even though PARP-1-null/Wrnhel/hel proliferation declines more rapidly than the other genotypes in culture. PARP-1-null/Wrnhel/hel MEFs are remarkable for the number of metaphase spreads with chromatid breaks, triradial or quadriradial structures, and double-minute chromosomes compared to wild-type, Wrnhel/hel, or PARP-1-null cells. From these results, it is concluded that it is the higher frequency of chromosomal fragments and the higher frequency of chromosomal breaks that are the major contributors of the observed abrupt cell-growth arrest in the PARP-1-null/Wrnhel/hel culture.
Possible Function of Wrn and PARP-1 at DNA Replication Forks
PARP-1 is a nick sensor that also recognizes and binds to double-strand breaks. It facilitates DNA repair by protecting against uncontrolled DNA recombination.48 In addition to its anti-recombinogenic effect, PARP-1 is believed to be part of a DNA break-signaling mechanism recruiting and regulating DNA repair molecules at the sites of DNA damage.48,49 It has been shown that PARP-1-deficient MEFs have a prolonged delay in DNA strand-break resealing leading to chromosomal instability.50 PARP-1 is also believed to participate in the base excision repair pathway.21,22 There is no direct evidence for the participation of WRN in base excision repair. However, WS cells are sensitive to the drug 4-nitroquinoline 1-oxide.51 This agent will cause alkylation of DNA and induce oxidative stress in cells.52 Hence, there is the interesting possibility that the WRN protein might be involved in the efficient removal of certain DNA lesions by a base excision repair pathway. In addition, WRN interacts with Fen1 and proliferation cellular nuclear antigen,16,18 both of which are involved in base excision repair.53 Future experiments should yield insight in the potential involvement of WRN in base excision repair of specific oxidative lesions.
The WRN homologues are considered suppressors of illegitimate recombination. Thus, one possible function of the mouse Wrn protein is to monitor recombinational repair of double-strand breaks at the site of DNA replication.17
Aberrant DNA structures may arise during the process of DNA replication causing collapses of replication forks. Such collapses would generate regions of DNA breaks that may provide substrates for the initiation of recombination. If the Wrn protein is mutated, illegitimate recombination may occur at a high frequency generating abnormal chromosomes as it is observed in our Wrnhel/hel mice.
Finally, several chromatid and chromosomal breaks were detected in PARP-1-null/Wrnhel/hel MEFs metaphase spreads and such DNA breaks persisted in these cells apparently generating chromosomal fragmentation (Figure 5
and Table 3
). In comparison, fewer persistent breaks were detected in Wrnhel/hel and PARP-1-null karyotypes. These results suggest that PARP-1 and Wrn proteins are involved together in the processing of DNA breaks not only during DNA replication, but also after the S phase of the cell cycle.
With respect to DNA damaging agents, it is interesting to note that PARP-1 activity is required for rapid accumulation of p53, activation of p53 sequence-specific DNA binding, and its transcriptional activity after DNA damage.54
The p53 tumor suppressor is required for G1 arrest55
and it is believed that this arrest allows cells time to repair DNA damage before being fixed as mutations. A portion of this p53-dependent cell-cycle control is manifest via p21 induction which is, in turn, responsible for the G1/S cell-cycle checkpoint.37
In addition to cell-cycle arrest, the ability of p53 to induce apoptosis is thought to be an important factor for its tumor suppressor function.56
It has been shown that human WS cells have an attenuation of the p53-dependent apoptotic pathway.57
Moreover, a role for WRN in transcription is also suggested by the observation that overexpression of WRN results in enhanced p53-dependent transcriptional activity.58
Consequently, a suboptimal activation of p53 in cells accumulating DNA breaks, as it is observed in PARP-1-null/Wrnhel/hel cells, would certainly affect cell proliferation. Finally, several PARP-1-null/Wrnhel/hel metaphase spreads have shown rearrangements of the region of chromosome 12 containing the p53 gene. Careful analysis of p53 status in all our MEFs is thus required.
|
Acknowledgements |
|---|
hel/hel and the PARP-1-null mice, respectively; Dr. E. A. Drobetsky for careful editing of the manuscript and helpful comments; A. Julien for technical assistance with the animals; and Dr. L. Turcot-Lemay for the statistical analyses. |
Footnotes |
|---|
Supported in part by grants from the Canadian Institutes of Health Research (to M. L.) and the National Cancer Institute of Canada (with funds from the Canadian Cancer Society to R. D.).
M. L. is a scholar of the Canadian Institutes of Health Research and R. D. is a senior scholar of the Fonds de la Recherche en Santé du Québec.
Accepted for publication February 5, 2003.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Beneke, O. Cohausz, M. Malanga, P. Boukamp, F. Althaus, and A. Burkle Rapid regulation of telomere length is mediated by poly(ADP-ribose) polymerase-1 Nucleic Acids Res., November 1, 2008; 36(19): 6309 - 6317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Tanori, M. Mancuso, E. Pasquali, S. Leonardi, S. Rebessi, V. Di Majo, M.-N. Guilly, F. Giangaspero, V. Covelli, S. Pazzaglia, et al. PARP-1 cooperates with Ptc1 to suppress medulloblastoma and basal cell carcinoma Carcinogenesis, October 1, 2008; 29(10): 1911 - 1919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Beneke and A. Burkle Poly(ADP-ribosyl)ation in mammalian ageing Nucleic Acids Res., December 3, 2007; 35(22): 7456 - 7465. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Brosh Jr and V. A. Bohr Human premature aging, DNA repair and RecQ helicases Nucleic Acids Res., December 3, 2007; 35(22): 7527 - 7544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Deschenes, L. Massip, C. Garand, and M. Lebel In vivo misregulation of genes involved in apoptosis, development and oxidative stress in mice lacking both functional Werner syndrome protein and poly(ADP-ribose) polymerase-1 Hum. Mol. Genet., November 1, 2005; 14(21): 3293 - 3308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Y. Kim, T. Zhang, and W. L. Kraus Poly(ADP-ribosyl)ation by PARP-1: `PAR-laying' NAD+ into a nuclear signal Genes & Dev., September 1, 2005; 19(17): 1951 - 1967. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Beneke and A. Burkle Poly(ADP-Ribosyl)ation, PARP, and Aging Sci. Aging Knowl. Environ., December 8, 2004; 2004(49): re9 - re9. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. von Kobbe, J. A. Harrigan, V. Schreiber, P. Stiegler, J. Piotrowski, L. Dawut, and V. A. Bohr Poly(ADP-ribose) polymerase 1 regulates both the exonuclease and helicase activities of the Werner syndrome protein Nucleic Acids Res., August 3, 2004; 32(13): 4003 - 4014. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Li, S. Navarro, N. Kasahara, and L. Comai Identification and Biochemical Characterization of a Werner's Syndrome Protein Complex with Ku70/80 and Poly(ADP-ribose) Polymerase-1 J. Biol. Chem., April 2, 2004; 279(14): 13659 - 13667. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Susse, C.-J. Scholz, A. Burkle, and L. Wiesmuller Poly(ADP-ribose) polymerase (PARP-1) and p53 independently function in regulating double-strand break repair in primate cells Nucleic Acids Res., February 2, 2004; 32(2): 669 - 680. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. von Kobbe, J. A. Harrigan, A. May, P. L. Opresko, L. Dawut, W.-H. Cheng, and V. A. Bohr Central Role for the Werner Syndrome Protein/Poly(ADP-Ribose) Polymerase 1 Complex in the Poly(ADP-Ribosyl)ation Pathway after DNA Damage Mol. Cell. Biol., December 1, 2003; 23(23): 8601 - 8613. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
