- Email: [email protected]
Involvement of DNA damage and repair systems in neurodegenerative process
Toxicology Letters 139 (2003) 99 /105 www.elsevier.com/locate/toxlet
Short communication
Involvement of DNA damage and repair systems in neurodegenerative process Daniela Uberti, Giulia Ferrari Toninelli, Maurizio Memo * Department of Biomedical Sciences and Biotechnologies, University of Brescia, Via Valsabbina, 19, 25124 Brescia, Italy
Abstract The study summarizes some recent data from our and other groups underlining the contribution to neurodegeneration of two transcription factors known to be involved in DNA damage sensing and repairing: the tumour suppressor gene p53 and the component of the DNA repair system MSH2. Both proteins participate in the cancer prevention machinery for the body as well as in the neurodegenerative process, suggesting that cancer and neurodegenerative disease may share common genetic risk factors for the development and progression of the disease. Here we show that, in neuronal cells, divergent cellular insults, i.e. the exposure to glutamate, b-amyloid (Ab) or H2O2, may converge to a common pathway that initiate with elevation of p53 protein levels. We also found that in SH-SY5Y neuronal cells H2O2 induced the activation of DNA repair system with the nuclear translocation of MSH2, and PCNA. Differently no changes in MSH2 and PCNA cellular distribution were found in undifferentiating SH-SY5Y cells exposed to H2O2. This argues that defects in the repair of, or response to, DNA damage impact significantly on brain function. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: p53 gene product; Excitotoxicity; DNA damage/repair; Apoptosis
1. Introduction Preservation of genomic stability is an essential biological function. Cells engage very efficiently mechanisms involving DNA surveillance/repair proteins that work to maintaining inherited nucleotide sequence of genomic DNA over time. After DNA damage, that can arise both physiologically and pathologically during duplication or after genotoxic stimuli, cells activate intracellular
* Corresponding author. Tel.: /39-030-371-7516; fax: /39030-370-1157 E-mail address: [email protected] (M. Memo).
pathways which are able to recognize the damage, to arrest cell cycle, to recruit DNA repair factors, to repair the damage or induce apoptosis. This definitely relevant process is finalized to prevent the generation and the persistence of impaired cells which may ultimately be detrimental to the organism. Very little is known about the role of DNA damage sensors and repair factors in terminally differentiated, not proliferating cells, like neurons. It is well recognized that mutation of genes related with DNA damage repair are associated with specific cancer-prone syndromes. Interestingly, as recently pointed out by Rolig and McKinnon (2000), many human pathological
0378-4274/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 4 2 7 4 ( 0 2 ) 0 0 4 2 3 - X
100
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
conditions with genetic defects in DNA damage responses are also characterized by neurological deficits. These neurological deficits can manifest themselves during many stages of development, suggesting an important role for DNA repair system during the development and maintenance of the brain. Here we summarize some recent data from our and other groups underlining the contribution to neurodegeneration of at least two transcription factors known to be involved in DNA damage sensing and repairing: the tumour suppressor gene p53 and the component of the DNA repair system MSH2. Both proteins participate in the cancer prevention machinery for the body as well as in the neurodegenerative process. Moreover, they interact with each other to orchestrate DNA repair functions and are transcriptionally self-regulated (Scherer et al., 2000).
2. Contribution of p53 to neuronal degeneration p53, is a transcription factor, whose main function is to control cell cycle progression and apoptotic process. p53 is also recognized to sense DNA damage and participate to DNA repair machinery (see review Almong and Rotter 1997). Loss or inactivation of p53 gene occurs in almost half of all human solid tumours, and is considered a fundamental predisposing event in the pathogenesis of many types of cancer (Purdie et al. 1994). Patients with Li-Fraumeni syndrome, characterized by germ line mutations in p53 gene, present high risk to develop a variety of tumours (Malkin et al., 1990), and mice deficient in p53 display precocious tumour development (Donehower et al., 1992). Interestingly, lack of p53 expression was associated with learning deficiency and behavioural alteration (Amson et al., 2000). p53 protein is upregulated in response to various cellular insults. Different cellular injury, including DNA damage, hypoxia, oxidative stress, ribonucleotide depletion and chemioterapeutic agents, were shown to stabilized p53 protein, which in turn can either cause growth arrest, permitting the induction of DNA repair processes, or alternatively, can direct cells to undergo apoptosis (Ko
and Prives, 1996; Giaccia and Kastan 1998). The involvement of p53 in sensing damaged DNA and in controlling its repair is supported by the observation that p53 directly transactivates the proliferating cell nuclear antigen (PCNA), that is involved in DNA replication and repair, and the GADD45 gene, whose product interacts with PCNA. Furthermore, p53 was shown to bind several transcription factor which are involved in DNA damage repair machinery. Up to now, it is well recognized that, in proliferating cells, versatility in p53 activity, as a response to exogenous or endogenous signals, serves as a control mechanism to preserve the genoma integrity. Remain an open question whether or not p53 exerts the same role in neuronal cells by regulating DNA repair process or inducing apoptosis. In the last decade an emerging body of evidence underscores the particular relevance of the tumour suppressor gene p53 in the central nervous system, because diverse form of neuronal damage have been associated with its induction (Morrison and Kinoshita, 2000). For example, damage resulting from neuronal stimulation by excitatory amino acid has been strongly associated with p53 accumulation. We previously demonstrated that glutamate-induced apoptosis in primary cultures of cerebellar granule cells occurred in a p53 dependent manner (Uberti et al., 1998; Grilli and Memo, 1999; Uberti et al., 2000a,b). In particular, we found that treatment of cerebellar granule cells with a p53 specific antisense oligonucleotide prevented both glutamate-induced p53 expression and apoptosis (Uberti et al., 1998). In vivo studies, using kainic acid systemic administration or quinolinic acid intracerebral injection, further support the involvement of p53 in neurodegeneration by showing an increased p53 mRNA levels in vulnerable brain regions, which displayed signs of apoptosis, including DNA fragmentation (Sakhi et al., 1994; Hughes et al., 1996). Similarly, elevated p53 mRNA levels have been also detected in adrenalectomy-induced degeneration of hippocampal dentate gyrus granule cells (Schreiber et al., 1994). Furthermore, with respect to ischemic cell death, enhanced p53 immunoreactivity has been observed in damaged cortical and striatal
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
101
Fig. 1. p53 expression in neuronal cell phenotypes following different cytotoxic insult. (A) Primary cultures of cerebellar granule cells were exposed to 100 mM glutamate for 15 min, and then returned to the original culture-conditioned media until protein extracts were performed. Primary cultures of cortical neurons were challenged with 25 mM Ab25 35 for different times (8 and 20 h). Oligodendroglialike cells, named OLN93, were treated with 1 mM H2O2 for 5 min and then returned to the growth medium. After designed time periods (5 /20 min) OLN93 cells were harvested for protein extract preparation (Uberti et al., 1999). (B) Human neuroblastoma SHSY5Y cells were differentiated with 10 5 M retinoic acid for 1 week. The obtained neuron-like cells were triggered with 25 mM Ab25 35, or exposed to a brief pulse (5 min) of 1 mM H2O2. In both A and B panels 30 mg of proteins from the different cell cultures underwent electrophoresis on 10% SDS-PAGE, transferred to nitrocellulose paper and immunoblotted with anti-p53 monoclonal antibody.
neurons 12 h following 2 h focal cerebral ischemia (Chopp et al., 1992), while p53 null mice have been reputed to be more resistant to focal cerebral ischemia (Crumrine et al., 1994). Recently, the role of p53 in neuronal death was further confirmed by using animals models of human neurodegenerative diseases, including Alzheimer’s disease (AD). Transgenic mice for human APP gene carrying an AD-associated mutation, which overexpressed beta-amyloid 1/42 peptide (Ab1/42), displayed an intensive p53 immunorecativity in neurons that were TUNEL positive (LaFerla et al., 1996). We also demonstrated a correlation between p53 expression and Ab injury, using primary cultures of cortical neurons (Copani et al., 2001). In particular, Ab25 /35 neurotoxic peptide caused in these neurons an increase of p53 expression, which was observed from 8 h up to at least 20 h following the insult (Fig. 1). Herein we reported a series of data showing an accumulation of p53 protein in multiple cell population of central nervous system, following a diverse array of cellular insults (Fig. 1). p53 is upregulated in cerebellar granule neurons follow-
ing glutamate; in cortical neurons following Ab as well as in oligodendrocyte cell line exposed to an oxidative stress (Fig. 1, panel A). Furthermore in the neuron-like cells different stimuli, such as Ab and hydrogen peroxide (H2O2), lead to an increase of p53 expression (Fig. 1, panel B). It is interesting to note that p53 was very early induced (5 min) following H2O2 pulse, while longer time was required for detecting an increased p53 expression after glutamate or Ab exposure. These findings suggest that these divergent cellular insults may converge to a common pathway that initiate with elevation of p53 protein levels. In this regards, it should be noted that both glutamate and Ab may lead, although throughout different intracellular pathways, to the generation of free radicals as well as H2O2 is a reactive oxygen species (ROS) producer. Indeed, it is well established that ROS induce damage of cellular components such as lipids, proteins and DNA and that accumulation of DNA damage is a well known stimulus for elevating p53 protein levels and for activating p53-mediated signalling pathways.
102
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
3. Contribution of the mismatch repair system to neuronal degeneration The mismatch repair system (MMR) is an important member of the DNA checkpoint, that includes a number of proteins aimed to control genomic stability through cell cycle arrest, DNA repair and apoptosis. Repair of damaged genes is the prominent role of this system, that is shared with other mayor repair systems, i.e. nucleotide excision repair (NER) and base excision repair (BER). Several studies have led to a biochemical model for postreplication mismatch repair in Escherichia coli. Initiation of an MMR event occurs when MutS recognizes and binds mispaired nucleotides that result from polymerase misincorporation errors (Su and Modrich, 1986). In eukaryotic cells there are at least seven MutS homologues (MSH1 to MSH7) and four MutL homologues designated MLH1, MLH2, MLH3, PMS1 and PMS2. Eukaryotic MutS and MutL homologues typically form heterodimers. MSH heterodimers are specialized for different (but overlapping) classes of mismatches: the most important in mammals are MSH2 /MSH6 (MutSa) heterodimers, that generally recognize base mispairs (e.g. G/T, A/C) and short insertion /deletion loops (IDLs), and MSH2 /MSH3 (MutSb), that bind to IDLs. MutS proteins interact also with other DNA lesions: small alteration such as O6-methylguanine, 8-oxoguanine and thymine glycol, and major ultraviolet light photoproducts such as Cyclobutane pyrimidine dimers (CPDs) and 1,2 intrastrand G/G intrastrands crosslinks produced by cysplatin. PCNA is a further member of MMR system. This protein was originally thought to interact with DNA polymerases and increase their activity. Recently, it was found that yeast MSH3 and MSH6 contain N-terminal sequence motifs characteristics of proteins that bind to PCNA (Clark et al., 2000) and human PCNA co-immunoprecipitate in a complex containing MLH1 and PMS2 (Gu et al., 1998). These new insights suggest that PCNA may act in MMR before the DNA syntesis step, directly in mispairs recognition (Umar et al., 1996).
It has been recently shown that MMR system could play a further role in inducing apoptosis. When DNA damage is too extensive, these proteins, which are usually involved in DNA repair, may participate in the apoptotic pathway signalling (Bernstein et al., 2002). Moreover Hickman and Samson (1999) demonstrated that the MMR proteins might be involved in the molecular events required to initiate apoptosis in response to DNA base damage. On the other hand, hMutSa and hMutLa appeared to influence the phosphorylation of p53 in response to DNA methylator damage (Duckett et al., 1999). Human epithelial and mouse embryo fibroblast cell lines lacking MLH1 protein were more resistant to H2O2, likely via the disregulation of apoptosis. In fact, MLH1 proficient cells exposed to H2O2 underwent apoptotic death, involving mitochondrial permeability and caspase 3 activation (Hardman et al., 2001). Very little is know about the possible role of the MMR system in neurons. Marietta et al. (1998) show the expression of MutSa complex proteins in developing and adult rat brain tissues, suggesting that postmitotic cells also, and not only dividing cells, have the capacity to carry out DNA mismatch repair. Nuclear extracts from adult rat brain neurons were also found to be able to repair DNA mismatches (Brooks et al., 1996). In this contest, we previously demonstrated the modulation of MSH2 expression, a protein known to be involved in recognition of damaged DNA, following excitotoxic insult in adult rat brain. MSH2 immunoreactivity was found in several areas of the brain and in the spinal cord. Interestingly, MSH2 was detected only in neuronal cells with a nuclear localization, whereas glial cells were not positive (Belloni et al., 1999). Furthermore, the kainic acid injection caused a dose-dependent increase of MSH2 expression specifically in CA3/ CA4 hippocampal fields that was found damaged. MSH2 increased expression could be interpreted as a part of pro-apoptotic signalling. This is in line with recent data demonstrating overexpression of MSH2 and MLH1 proteins in cell undergoing apoptosis (Zhang et al., 1999). We further studied the possible involvement of MMR in the repair of H2O2 induced DNA base oxidation using SH-SY5Y human neuroblastoma
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
103
Fig. 2. MSH2 and PCNA expression in undifferentiated and differentiated cells following oxidative stress. Undifferentiated and retinoic acid-differentiated SH-SY5Y cells were exposed for 20 min to 2.5 mM H2O2. At different periods of time cells were harvested and subcellular fractionation was performed. 20 mg cytoplasmic and nuclear extracts were electrophoresed on 8 or 12% SDS-PAGE, and transferred to nitrocellulose paper. Filters were incubated with the monoclonal antibodies anti-MSH2 and anti-PCNA. Anti-actin monoclonal antibody was used as a control.
cell line. In order to compare the efficiency of this system in proliferating and quiescent cells, SHSY5Y cultures were differentiated with retinoic acid for 1 week to generate a neuron-like phenotype. Undifferentiated and differentiated cells were then exposed to a brief pulse of H2O2 and, at different time periods, evaluated for MSH2 and PCNA proteins expression and localization. Western blot analysis of cytosolic and nuclear extracts from cells untreated and challenged with H2O2 was performed using anti-MSH2 and anti-PCNA monoclonal antibodies (Fig. 2). A clear nuclear translocation of MSH2 and PCNA was observed only in differentiated cells after the oxidative insult. In particular, MSH2 translocated already 15 min after H2O2 pulse, while more time was required to detected PCNA protein in the nucleus (Fig. 1, panel B). Undifferentiated cells showed a high steady-state levels of MSH2 and PCNA in both nuclear and cytosolic compartment. No changes in MSH2 and PCNA cellular distribution were found after H2O2 treatment (Fig. 1, panel A). This differential behaviour observed in undifferentiated versus differentiated cells suggests that in neuronal cell line, the DNA repair systems can be activated only in suffering conditions, whereas in dividing cells the genetic sequences must be continuously monitored. Interestingly, PCNA translocation in differentiated cells (Fig. 1, panel B) could be interpreted in, at least, two different ways. In fact, this protein
could be implicated either in mispair recognition or act during the DNA synthesis step, or both.
4. Concluding remarks Historically, one of the most well known connections between abnormalities of the DNA damage response and neurodegeneration has been the human syndrome of ataxia telangiectasia. However, other syndromes associated with defective DNA damage response also include neurological symptoms as a primary feature of their phenotypes (Rolig and McKinnon, 2000). This argues that defects in the repair of, or response to, DNA damage impact significantly on brain function. We speculate that the list of neuropathologies associated with an impairment of the DNA damage repair system could be even larger including chronic and progressive neurodegenerative diseases like AD. In this regard, quantitative and histopathological markers of oxidative DNA damage have been found in the brain of AD patients (Pratico and Delanty, 2000) suggesting that the neuronal machinery devoted to restore damaged DNA might be impaired in AD brain. Also, in vitro studies support the view that p53 is implicated in the neurodegenerative program activated by b-amyloid deposition (Copani et al., 2001). Future studies focussing on these and other aspects of b-amyloid neurotoxicity will be impor-
104
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
tant for the molecular description of specific DNA processing and repair mechanisms in the brain as well as to define whether cancer and neurodegenerative disease share common genetic risk factors for the development and progression of the disease.
References Almong, N., Rotter, V., 1997. Involvement of p53 in cell differentiation and development. Biochem. Biophys. Acta 1333, F1 /F27. Amson, R., Lassalle, J.M., Halley, H., Prieur, S., Lethrosne, F., Roperch, J.P., Israeli, D., Gendron, M.C., Duyckaerts, C., Checler, F., Dausset, J., Cohen, D., Oren, M., Telerman, A., 2000. Behavioral alterations associated with apoptosis and down-regulation of presenilin 1 in the brains of p53-deficient mice. Proc. Natl. Acad. Sci. USA 97, 5346. Belloni, M., Uberti, D., Rizzini, C., Ferrari-Toninelli, G., Rizzonelli, P., Jiricny, J., Spano, P., Memo, M., 1999. Distribution and kainate-mediated induction of the DNA mismatch repair protein MSH2 in rat brain. Neuroscience 94, 1323 /1331. Bernstein, C., Bernstein, H., Payne, C.M., Garewal, H., 2002. DNA repair/pro-apoptotic dual role proteins in five major DNA repair pathways: fail-safe protection against carcinogenesis. Mutat. Res. 511, 145 /178. Brooks, P.J., Marietta, C., Goldman, D., 1996. DNA mismatch repair and DNA methylation in adult brain neurons. J. Neurosci. 16, 939 /945. Clark, A.B., Valle, F., Drotschmann, K., Gary, R.K., Kunkel, T.A., 2000. Functional interaction of proliferating cell nuclear antigen with MSH2 /MSH6 and MSH2 and MSH3 complexes. J. Biol. Chem. 275, 36498 /36501. Chopp, M., Li, Y., Zhang, Z.G., Freytag, S.O., 1992. p53 expression in brain after middle cerebral artery occlusion in the rat. Biochem. Biophys. Res. Commun. 182, 1201 /1207. Copani, A., Uberti, D., Sortino, H.A., Bruno, V., Nicoletti, F., Memo, M., 2001. Activation of cell-cycle-associated proteins in neuronal death: a mandatory or dispensable path? Trends Neurosci. 24, 25 /31. Crumrine, R.C., Thomas, A.L., Morgan, P.F., 1994. Attenuation of p53 expression protects against focal ischemic damage in transgenic mice. J. Cereb. Blood Flow Metab. 14, 887 /891. Donehower, L.A., Harvey, M., Slagle, B.T., McArthur, M.J., Montgomery, C.A., Butel, J.S., Bradley, A., 1992. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature 356, 215 /221. Duckett, D.R., Bronstein, S.M., Taya, Y., Modrich, P., 1999. hMutSa and hMutLa-dependent phosphorylation of p53 in response to DNA methylator damage. Proc. Natl. Acad. Sci. USA 96, 12384 /12388.
Giaccia, A.J., Kastan, M.B., 1998. The complexity of p53 modulation: emerging patterns from divergent signals. Genes Dev. 12, 2973 /2983. Grilli, M., Memo, M., 1999. Possible role of NF-kappaB and p53 in the glutamate-induced pro-apoptotic neuronal pathway. Cell Death Differ. 6, 22 /27. Gu, L., Hong, Y., McCulloch, S., Watanabe, H., Li, G., 1998. ATP-dependent interaction of human mismatch repair proteins and dual role of PCNA in mismatch repair. Nucleic Acid Res. 26, 1173 /1178. Hardman, R.A., Afshari, C.A., Barrett, J.C., 2001. Involvement of mammalian MLH1 in the apoptotic response to peroxide-induced oxidative stress. Cancer Res. 61, 1392 /1397. Hickman, M.J., Samson, L.D., 1999. Role of DNA mismatch repair and p53 in signaling induction of apoptosis by alkylating agents. Proc. Natl. Acad. Sci. USA 96, 10764 / 10769. Hughes, P.E., Alexi, T., Yoshida, T., Schreiber, S.S., Knusel, B., 1996. Excitotoxic lesion of rat brain with quinolinic acid induces expression of p53 messenger RNA and protein and p53-inducible genes Bax and Gadd-45 in brain areas showing DNA fragmentation. Neuroscience 74, 1143 /1160. Ko, L.J., Prives, C., 1996. p53: puzzle and paradigm. Genes Dev. 10, 1054 /1072. LaFerla, F.M., Hall, C.K., Ngo, L., Jay, G., 1996. Extracellular deposition of beta-amyloid upon p53-dependent neuronal cell death in transgenic mice. J. Clin. Invest. 98, 1626 /1632. Malkin, D., Li, F.P., Strong, L.C., Fraumeni, J.F., Jr, Nelson, C.E., Kim, D.H., Kassel, J., Gryka, M.A., Bischoff, F.Z., Tainsky, M.A., et al., 1990. Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 250, 1233 /1238. Marietta, C., Palombo, F., Gallinari, P., Jiricny, J., Brooks, P.J., 1998. Expression of long-patch and short-patch DNA mismatch repair proteins in the embryonic and adult mammalian brain. Mol. Brain Res. 53, 317 /320. Morrison, R.S., Kinoshita, Y., 2000. The role of p53 in neuronal cell death. Cell Death Diff. 7, 868 /879. Pratico, D., Delanty, N., 2000. Oxidative injury in diseases of the central nervous system: focus on Alzheimer’s disease. Am. J. Med. 7, 577 /585. Purdie, C.A., Harrison, D.J., Peter, A., Dobbie, L., White, S., Howie, S.E., Salter, D.M., Bird, C.C., Wyllie, A.H., Hooper, M.L., et al., 1994. Tumour incidence, spectrum and ploidy in mice with a large deletion in the p53 gene. Oncogene 9, 603 /609. Rolig, R.L., McKinnon, P.J., 2000. Linking DNA damage and neurodegeneration. Trends Neurosci. 23, 417 /424. Sakhi, S., Bruce, A., Sun, N., Tocco, G., Baudry, M., Schreiber, S.S., 1994. p53 induction is associated with neuronal damage in the central nervous system. Proc. Natl. Acad. Sci. USA 89, 12028 /12032. Scherer, S.J., Maier, S.M., Seifert, M., Hanselmann, R.G., Zang, K.D., Muller-Hermelink, H.K., Angel, P., Welter, C., Schartl, M., 2000. p53 and c-Jun functionally synergize in the regulation of the DNA repair gene hMSH2 in response to UV. J. Biol. Chem. 275, 37469 /37473.
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105 Schreiber, S.S., Sakhi, S., Dugich-Djordjevic, M.M., Nichols, N.R., 1994. Tumor suppressor p53 induction and DNA damage in hippocampal granule cells after adrenalectomy. Exp. Neurol. 130, 368 /377. Su, S.S., Modrich, P., 1986. Escherichia Coli mutS -encoded protein binds to mismatched DNA base pairs. Proc. Natl. Acad. Sci. 83, 5057 /5061. Uberti, D., Belloni, M., Grilli, M., Spano, P., Memo, M., 1998. Induction of tumour-suppressor phosphoprotein p53 in the apoptosis of cultured rat cerebellar neurones triggered by excitatory amino acids. Eur. J. Neurosci. 10, 246 /254. Uberti, D., Yavin, E., Gil, S., Ayasola, K.R., Goldfinger, N., Rotter, V., 1999. Hydrogen peroxide induces nuclear translocation of p53 and apoptosis in cells of oligodendroglia origin. Brain Res. Mol. Brain Res. 65, 167 /175.
105
Uberti, D., Grilli, M., Memo, M., 2000a. Induction of p53 in the glutamate-induced cell death program. Amino Acids 19, 253 /261. Uberti, D., Grilli, M., Memo, M., 2000b. Contribution of NFkappaB and p53 in the glutamate-induced apoptosis. Int. J. Dev. Neurosci. 18, 447 /454. Umar, A., Buermeyer, A.B., Simon, J.A., Thomas, D.C., Clark, A.B., Liskay, R.M., Kunkel, T.A., 1996. Requirement for PCNA in DNA mismatch repair at a step preceding DNA resynthesis. Cell 87, 65 /73. Zhang, H., Richards, B., Wilson, T., Lloyd, M., Cranston, A., Thorburn, A., Fishel, R., Meuth, M., 1999. Apoptosis induced by overexpression of hMSH2 or hMLH1. Cancer Res. 59, 3021 /3027.
Short communication
Involvement of DNA damage and repair systems in neurodegenerative process Daniela Uberti, Giulia Ferrari Toninelli, Maurizio Memo * Department of Biomedical Sciences and Biotechnologies, University of Brescia, Via Valsabbina, 19, 25124 Brescia, Italy
Abstract The study summarizes some recent data from our and other groups underlining the contribution to neurodegeneration of two transcription factors known to be involved in DNA damage sensing and repairing: the tumour suppressor gene p53 and the component of the DNA repair system MSH2. Both proteins participate in the cancer prevention machinery for the body as well as in the neurodegenerative process, suggesting that cancer and neurodegenerative disease may share common genetic risk factors for the development and progression of the disease. Here we show that, in neuronal cells, divergent cellular insults, i.e. the exposure to glutamate, b-amyloid (Ab) or H2O2, may converge to a common pathway that initiate with elevation of p53 protein levels. We also found that in SH-SY5Y neuronal cells H2O2 induced the activation of DNA repair system with the nuclear translocation of MSH2, and PCNA. Differently no changes in MSH2 and PCNA cellular distribution were found in undifferentiating SH-SY5Y cells exposed to H2O2. This argues that defects in the repair of, or response to, DNA damage impact significantly on brain function. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: p53 gene product; Excitotoxicity; DNA damage/repair; Apoptosis
1. Introduction Preservation of genomic stability is an essential biological function. Cells engage very efficiently mechanisms involving DNA surveillance/repair proteins that work to maintaining inherited nucleotide sequence of genomic DNA over time. After DNA damage, that can arise both physiologically and pathologically during duplication or after genotoxic stimuli, cells activate intracellular
* Corresponding author. Tel.: /39-030-371-7516; fax: /39030-370-1157 E-mail address: [email protected] (M. Memo).
pathways which are able to recognize the damage, to arrest cell cycle, to recruit DNA repair factors, to repair the damage or induce apoptosis. This definitely relevant process is finalized to prevent the generation and the persistence of impaired cells which may ultimately be detrimental to the organism. Very little is known about the role of DNA damage sensors and repair factors in terminally differentiated, not proliferating cells, like neurons. It is well recognized that mutation of genes related with DNA damage repair are associated with specific cancer-prone syndromes. Interestingly, as recently pointed out by Rolig and McKinnon (2000), many human pathological
0378-4274/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 4 2 7 4 ( 0 2 ) 0 0 4 2 3 - X
100
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
conditions with genetic defects in DNA damage responses are also characterized by neurological deficits. These neurological deficits can manifest themselves during many stages of development, suggesting an important role for DNA repair system during the development and maintenance of the brain. Here we summarize some recent data from our and other groups underlining the contribution to neurodegeneration of at least two transcription factors known to be involved in DNA damage sensing and repairing: the tumour suppressor gene p53 and the component of the DNA repair system MSH2. Both proteins participate in the cancer prevention machinery for the body as well as in the neurodegenerative process. Moreover, they interact with each other to orchestrate DNA repair functions and are transcriptionally self-regulated (Scherer et al., 2000).
2. Contribution of p53 to neuronal degeneration p53, is a transcription factor, whose main function is to control cell cycle progression and apoptotic process. p53 is also recognized to sense DNA damage and participate to DNA repair machinery (see review Almong and Rotter 1997). Loss or inactivation of p53 gene occurs in almost half of all human solid tumours, and is considered a fundamental predisposing event in the pathogenesis of many types of cancer (Purdie et al. 1994). Patients with Li-Fraumeni syndrome, characterized by germ line mutations in p53 gene, present high risk to develop a variety of tumours (Malkin et al., 1990), and mice deficient in p53 display precocious tumour development (Donehower et al., 1992). Interestingly, lack of p53 expression was associated with learning deficiency and behavioural alteration (Amson et al., 2000). p53 protein is upregulated in response to various cellular insults. Different cellular injury, including DNA damage, hypoxia, oxidative stress, ribonucleotide depletion and chemioterapeutic agents, were shown to stabilized p53 protein, which in turn can either cause growth arrest, permitting the induction of DNA repair processes, or alternatively, can direct cells to undergo apoptosis (Ko
and Prives, 1996; Giaccia and Kastan 1998). The involvement of p53 in sensing damaged DNA and in controlling its repair is supported by the observation that p53 directly transactivates the proliferating cell nuclear antigen (PCNA), that is involved in DNA replication and repair, and the GADD45 gene, whose product interacts with PCNA. Furthermore, p53 was shown to bind several transcription factor which are involved in DNA damage repair machinery. Up to now, it is well recognized that, in proliferating cells, versatility in p53 activity, as a response to exogenous or endogenous signals, serves as a control mechanism to preserve the genoma integrity. Remain an open question whether or not p53 exerts the same role in neuronal cells by regulating DNA repair process or inducing apoptosis. In the last decade an emerging body of evidence underscores the particular relevance of the tumour suppressor gene p53 in the central nervous system, because diverse form of neuronal damage have been associated with its induction (Morrison and Kinoshita, 2000). For example, damage resulting from neuronal stimulation by excitatory amino acid has been strongly associated with p53 accumulation. We previously demonstrated that glutamate-induced apoptosis in primary cultures of cerebellar granule cells occurred in a p53 dependent manner (Uberti et al., 1998; Grilli and Memo, 1999; Uberti et al., 2000a,b). In particular, we found that treatment of cerebellar granule cells with a p53 specific antisense oligonucleotide prevented both glutamate-induced p53 expression and apoptosis (Uberti et al., 1998). In vivo studies, using kainic acid systemic administration or quinolinic acid intracerebral injection, further support the involvement of p53 in neurodegeneration by showing an increased p53 mRNA levels in vulnerable brain regions, which displayed signs of apoptosis, including DNA fragmentation (Sakhi et al., 1994; Hughes et al., 1996). Similarly, elevated p53 mRNA levels have been also detected in adrenalectomy-induced degeneration of hippocampal dentate gyrus granule cells (Schreiber et al., 1994). Furthermore, with respect to ischemic cell death, enhanced p53 immunoreactivity has been observed in damaged cortical and striatal
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
101
Fig. 1. p53 expression in neuronal cell phenotypes following different cytotoxic insult. (A) Primary cultures of cerebellar granule cells were exposed to 100 mM glutamate for 15 min, and then returned to the original culture-conditioned media until protein extracts were performed. Primary cultures of cortical neurons were challenged with 25 mM Ab25 35 for different times (8 and 20 h). Oligodendroglialike cells, named OLN93, were treated with 1 mM H2O2 for 5 min and then returned to the growth medium. After designed time periods (5 /20 min) OLN93 cells were harvested for protein extract preparation (Uberti et al., 1999). (B) Human neuroblastoma SHSY5Y cells were differentiated with 10 5 M retinoic acid for 1 week. The obtained neuron-like cells were triggered with 25 mM Ab25 35, or exposed to a brief pulse (5 min) of 1 mM H2O2. In both A and B panels 30 mg of proteins from the different cell cultures underwent electrophoresis on 10% SDS-PAGE, transferred to nitrocellulose paper and immunoblotted with anti-p53 monoclonal antibody.
neurons 12 h following 2 h focal cerebral ischemia (Chopp et al., 1992), while p53 null mice have been reputed to be more resistant to focal cerebral ischemia (Crumrine et al., 1994). Recently, the role of p53 in neuronal death was further confirmed by using animals models of human neurodegenerative diseases, including Alzheimer’s disease (AD). Transgenic mice for human APP gene carrying an AD-associated mutation, which overexpressed beta-amyloid 1/42 peptide (Ab1/42), displayed an intensive p53 immunorecativity in neurons that were TUNEL positive (LaFerla et al., 1996). We also demonstrated a correlation between p53 expression and Ab injury, using primary cultures of cortical neurons (Copani et al., 2001). In particular, Ab25 /35 neurotoxic peptide caused in these neurons an increase of p53 expression, which was observed from 8 h up to at least 20 h following the insult (Fig. 1). Herein we reported a series of data showing an accumulation of p53 protein in multiple cell population of central nervous system, following a diverse array of cellular insults (Fig. 1). p53 is upregulated in cerebellar granule neurons follow-
ing glutamate; in cortical neurons following Ab as well as in oligodendrocyte cell line exposed to an oxidative stress (Fig. 1, panel A). Furthermore in the neuron-like cells different stimuli, such as Ab and hydrogen peroxide (H2O2), lead to an increase of p53 expression (Fig. 1, panel B). It is interesting to note that p53 was very early induced (5 min) following H2O2 pulse, while longer time was required for detecting an increased p53 expression after glutamate or Ab exposure. These findings suggest that these divergent cellular insults may converge to a common pathway that initiate with elevation of p53 protein levels. In this regards, it should be noted that both glutamate and Ab may lead, although throughout different intracellular pathways, to the generation of free radicals as well as H2O2 is a reactive oxygen species (ROS) producer. Indeed, it is well established that ROS induce damage of cellular components such as lipids, proteins and DNA and that accumulation of DNA damage is a well known stimulus for elevating p53 protein levels and for activating p53-mediated signalling pathways.
102
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
3. Contribution of the mismatch repair system to neuronal degeneration The mismatch repair system (MMR) is an important member of the DNA checkpoint, that includes a number of proteins aimed to control genomic stability through cell cycle arrest, DNA repair and apoptosis. Repair of damaged genes is the prominent role of this system, that is shared with other mayor repair systems, i.e. nucleotide excision repair (NER) and base excision repair (BER). Several studies have led to a biochemical model for postreplication mismatch repair in Escherichia coli. Initiation of an MMR event occurs when MutS recognizes and binds mispaired nucleotides that result from polymerase misincorporation errors (Su and Modrich, 1986). In eukaryotic cells there are at least seven MutS homologues (MSH1 to MSH7) and four MutL homologues designated MLH1, MLH2, MLH3, PMS1 and PMS2. Eukaryotic MutS and MutL homologues typically form heterodimers. MSH heterodimers are specialized for different (but overlapping) classes of mismatches: the most important in mammals are MSH2 /MSH6 (MutSa) heterodimers, that generally recognize base mispairs (e.g. G/T, A/C) and short insertion /deletion loops (IDLs), and MSH2 /MSH3 (MutSb), that bind to IDLs. MutS proteins interact also with other DNA lesions: small alteration such as O6-methylguanine, 8-oxoguanine and thymine glycol, and major ultraviolet light photoproducts such as Cyclobutane pyrimidine dimers (CPDs) and 1,2 intrastrand G/G intrastrands crosslinks produced by cysplatin. PCNA is a further member of MMR system. This protein was originally thought to interact with DNA polymerases and increase their activity. Recently, it was found that yeast MSH3 and MSH6 contain N-terminal sequence motifs characteristics of proteins that bind to PCNA (Clark et al., 2000) and human PCNA co-immunoprecipitate in a complex containing MLH1 and PMS2 (Gu et al., 1998). These new insights suggest that PCNA may act in MMR before the DNA syntesis step, directly in mispairs recognition (Umar et al., 1996).
It has been recently shown that MMR system could play a further role in inducing apoptosis. When DNA damage is too extensive, these proteins, which are usually involved in DNA repair, may participate in the apoptotic pathway signalling (Bernstein et al., 2002). Moreover Hickman and Samson (1999) demonstrated that the MMR proteins might be involved in the molecular events required to initiate apoptosis in response to DNA base damage. On the other hand, hMutSa and hMutLa appeared to influence the phosphorylation of p53 in response to DNA methylator damage (Duckett et al., 1999). Human epithelial and mouse embryo fibroblast cell lines lacking MLH1 protein were more resistant to H2O2, likely via the disregulation of apoptosis. In fact, MLH1 proficient cells exposed to H2O2 underwent apoptotic death, involving mitochondrial permeability and caspase 3 activation (Hardman et al., 2001). Very little is know about the possible role of the MMR system in neurons. Marietta et al. (1998) show the expression of MutSa complex proteins in developing and adult rat brain tissues, suggesting that postmitotic cells also, and not only dividing cells, have the capacity to carry out DNA mismatch repair. Nuclear extracts from adult rat brain neurons were also found to be able to repair DNA mismatches (Brooks et al., 1996). In this contest, we previously demonstrated the modulation of MSH2 expression, a protein known to be involved in recognition of damaged DNA, following excitotoxic insult in adult rat brain. MSH2 immunoreactivity was found in several areas of the brain and in the spinal cord. Interestingly, MSH2 was detected only in neuronal cells with a nuclear localization, whereas glial cells were not positive (Belloni et al., 1999). Furthermore, the kainic acid injection caused a dose-dependent increase of MSH2 expression specifically in CA3/ CA4 hippocampal fields that was found damaged. MSH2 increased expression could be interpreted as a part of pro-apoptotic signalling. This is in line with recent data demonstrating overexpression of MSH2 and MLH1 proteins in cell undergoing apoptosis (Zhang et al., 1999). We further studied the possible involvement of MMR in the repair of H2O2 induced DNA base oxidation using SH-SY5Y human neuroblastoma
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
103
Fig. 2. MSH2 and PCNA expression in undifferentiated and differentiated cells following oxidative stress. Undifferentiated and retinoic acid-differentiated SH-SY5Y cells were exposed for 20 min to 2.5 mM H2O2. At different periods of time cells were harvested and subcellular fractionation was performed. 20 mg cytoplasmic and nuclear extracts were electrophoresed on 8 or 12% SDS-PAGE, and transferred to nitrocellulose paper. Filters were incubated with the monoclonal antibodies anti-MSH2 and anti-PCNA. Anti-actin monoclonal antibody was used as a control.
cell line. In order to compare the efficiency of this system in proliferating and quiescent cells, SHSY5Y cultures were differentiated with retinoic acid for 1 week to generate a neuron-like phenotype. Undifferentiated and differentiated cells were then exposed to a brief pulse of H2O2 and, at different time periods, evaluated for MSH2 and PCNA proteins expression and localization. Western blot analysis of cytosolic and nuclear extracts from cells untreated and challenged with H2O2 was performed using anti-MSH2 and anti-PCNA monoclonal antibodies (Fig. 2). A clear nuclear translocation of MSH2 and PCNA was observed only in differentiated cells after the oxidative insult. In particular, MSH2 translocated already 15 min after H2O2 pulse, while more time was required to detected PCNA protein in the nucleus (Fig. 1, panel B). Undifferentiated cells showed a high steady-state levels of MSH2 and PCNA in both nuclear and cytosolic compartment. No changes in MSH2 and PCNA cellular distribution were found after H2O2 treatment (Fig. 1, panel A). This differential behaviour observed in undifferentiated versus differentiated cells suggests that in neuronal cell line, the DNA repair systems can be activated only in suffering conditions, whereas in dividing cells the genetic sequences must be continuously monitored. Interestingly, PCNA translocation in differentiated cells (Fig. 1, panel B) could be interpreted in, at least, two different ways. In fact, this protein
could be implicated either in mispair recognition or act during the DNA synthesis step, or both.
4. Concluding remarks Historically, one of the most well known connections between abnormalities of the DNA damage response and neurodegeneration has been the human syndrome of ataxia telangiectasia. However, other syndromes associated with defective DNA damage response also include neurological symptoms as a primary feature of their phenotypes (Rolig and McKinnon, 2000). This argues that defects in the repair of, or response to, DNA damage impact significantly on brain function. We speculate that the list of neuropathologies associated with an impairment of the DNA damage repair system could be even larger including chronic and progressive neurodegenerative diseases like AD. In this regard, quantitative and histopathological markers of oxidative DNA damage have been found in the brain of AD patients (Pratico and Delanty, 2000) suggesting that the neuronal machinery devoted to restore damaged DNA might be impaired in AD brain. Also, in vitro studies support the view that p53 is implicated in the neurodegenerative program activated by b-amyloid deposition (Copani et al., 2001). Future studies focussing on these and other aspects of b-amyloid neurotoxicity will be impor-
104
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105
tant for the molecular description of specific DNA processing and repair mechanisms in the brain as well as to define whether cancer and neurodegenerative disease share common genetic risk factors for the development and progression of the disease.
References Almong, N., Rotter, V., 1997. Involvement of p53 in cell differentiation and development. Biochem. Biophys. Acta 1333, F1 /F27. Amson, R., Lassalle, J.M., Halley, H., Prieur, S., Lethrosne, F., Roperch, J.P., Israeli, D., Gendron, M.C., Duyckaerts, C., Checler, F., Dausset, J., Cohen, D., Oren, M., Telerman, A., 2000. Behavioral alterations associated with apoptosis and down-regulation of presenilin 1 in the brains of p53-deficient mice. Proc. Natl. Acad. Sci. USA 97, 5346. Belloni, M., Uberti, D., Rizzini, C., Ferrari-Toninelli, G., Rizzonelli, P., Jiricny, J., Spano, P., Memo, M., 1999. Distribution and kainate-mediated induction of the DNA mismatch repair protein MSH2 in rat brain. Neuroscience 94, 1323 /1331. Bernstein, C., Bernstein, H., Payne, C.M., Garewal, H., 2002. DNA repair/pro-apoptotic dual role proteins in five major DNA repair pathways: fail-safe protection against carcinogenesis. Mutat. Res. 511, 145 /178. Brooks, P.J., Marietta, C., Goldman, D., 1996. DNA mismatch repair and DNA methylation in adult brain neurons. J. Neurosci. 16, 939 /945. Clark, A.B., Valle, F., Drotschmann, K., Gary, R.K., Kunkel, T.A., 2000. Functional interaction of proliferating cell nuclear antigen with MSH2 /MSH6 and MSH2 and MSH3 complexes. J. Biol. Chem. 275, 36498 /36501. Chopp, M., Li, Y., Zhang, Z.G., Freytag, S.O., 1992. p53 expression in brain after middle cerebral artery occlusion in the rat. Biochem. Biophys. Res. Commun. 182, 1201 /1207. Copani, A., Uberti, D., Sortino, H.A., Bruno, V., Nicoletti, F., Memo, M., 2001. Activation of cell-cycle-associated proteins in neuronal death: a mandatory or dispensable path? Trends Neurosci. 24, 25 /31. Crumrine, R.C., Thomas, A.L., Morgan, P.F., 1994. Attenuation of p53 expression protects against focal ischemic damage in transgenic mice. J. Cereb. Blood Flow Metab. 14, 887 /891. Donehower, L.A., Harvey, M., Slagle, B.T., McArthur, M.J., Montgomery, C.A., Butel, J.S., Bradley, A., 1992. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature 356, 215 /221. Duckett, D.R., Bronstein, S.M., Taya, Y., Modrich, P., 1999. hMutSa and hMutLa-dependent phosphorylation of p53 in response to DNA methylator damage. Proc. Natl. Acad. Sci. USA 96, 12384 /12388.
Giaccia, A.J., Kastan, M.B., 1998. The complexity of p53 modulation: emerging patterns from divergent signals. Genes Dev. 12, 2973 /2983. Grilli, M., Memo, M., 1999. Possible role of NF-kappaB and p53 in the glutamate-induced pro-apoptotic neuronal pathway. Cell Death Differ. 6, 22 /27. Gu, L., Hong, Y., McCulloch, S., Watanabe, H., Li, G., 1998. ATP-dependent interaction of human mismatch repair proteins and dual role of PCNA in mismatch repair. Nucleic Acid Res. 26, 1173 /1178. Hardman, R.A., Afshari, C.A., Barrett, J.C., 2001. Involvement of mammalian MLH1 in the apoptotic response to peroxide-induced oxidative stress. Cancer Res. 61, 1392 /1397. Hickman, M.J., Samson, L.D., 1999. Role of DNA mismatch repair and p53 in signaling induction of apoptosis by alkylating agents. Proc. Natl. Acad. Sci. USA 96, 10764 / 10769. Hughes, P.E., Alexi, T., Yoshida, T., Schreiber, S.S., Knusel, B., 1996. Excitotoxic lesion of rat brain with quinolinic acid induces expression of p53 messenger RNA and protein and p53-inducible genes Bax and Gadd-45 in brain areas showing DNA fragmentation. Neuroscience 74, 1143 /1160. Ko, L.J., Prives, C., 1996. p53: puzzle and paradigm. Genes Dev. 10, 1054 /1072. LaFerla, F.M., Hall, C.K., Ngo, L., Jay, G., 1996. Extracellular deposition of beta-amyloid upon p53-dependent neuronal cell death in transgenic mice. J. Clin. Invest. 98, 1626 /1632. Malkin, D., Li, F.P., Strong, L.C., Fraumeni, J.F., Jr, Nelson, C.E., Kim, D.H., Kassel, J., Gryka, M.A., Bischoff, F.Z., Tainsky, M.A., et al., 1990. Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 250, 1233 /1238. Marietta, C., Palombo, F., Gallinari, P., Jiricny, J., Brooks, P.J., 1998. Expression of long-patch and short-patch DNA mismatch repair proteins in the embryonic and adult mammalian brain. Mol. Brain Res. 53, 317 /320. Morrison, R.S., Kinoshita, Y., 2000. The role of p53 in neuronal cell death. Cell Death Diff. 7, 868 /879. Pratico, D., Delanty, N., 2000. Oxidative injury in diseases of the central nervous system: focus on Alzheimer’s disease. Am. J. Med. 7, 577 /585. Purdie, C.A., Harrison, D.J., Peter, A., Dobbie, L., White, S., Howie, S.E., Salter, D.M., Bird, C.C., Wyllie, A.H., Hooper, M.L., et al., 1994. Tumour incidence, spectrum and ploidy in mice with a large deletion in the p53 gene. Oncogene 9, 603 /609. Rolig, R.L., McKinnon, P.J., 2000. Linking DNA damage and neurodegeneration. Trends Neurosci. 23, 417 /424. Sakhi, S., Bruce, A., Sun, N., Tocco, G., Baudry, M., Schreiber, S.S., 1994. p53 induction is associated with neuronal damage in the central nervous system. Proc. Natl. Acad. Sci. USA 89, 12028 /12032. Scherer, S.J., Maier, S.M., Seifert, M., Hanselmann, R.G., Zang, K.D., Muller-Hermelink, H.K., Angel, P., Welter, C., Schartl, M., 2000. p53 and c-Jun functionally synergize in the regulation of the DNA repair gene hMSH2 in response to UV. J. Biol. Chem. 275, 37469 /37473.
D. Uberti et al. / Toxicology Letters 139 (2003) 99 /105 Schreiber, S.S., Sakhi, S., Dugich-Djordjevic, M.M., Nichols, N.R., 1994. Tumor suppressor p53 induction and DNA damage in hippocampal granule cells after adrenalectomy. Exp. Neurol. 130, 368 /377. Su, S.S., Modrich, P., 1986. Escherichia Coli mutS -encoded protein binds to mismatched DNA base pairs. Proc. Natl. Acad. Sci. 83, 5057 /5061. Uberti, D., Belloni, M., Grilli, M., Spano, P., Memo, M., 1998. Induction of tumour-suppressor phosphoprotein p53 in the apoptosis of cultured rat cerebellar neurones triggered by excitatory amino acids. Eur. J. Neurosci. 10, 246 /254. Uberti, D., Yavin, E., Gil, S., Ayasola, K.R., Goldfinger, N., Rotter, V., 1999. Hydrogen peroxide induces nuclear translocation of p53 and apoptosis in cells of oligodendroglia origin. Brain Res. Mol. Brain Res. 65, 167 /175.
105
Uberti, D., Grilli, M., Memo, M., 2000a. Induction of p53 in the glutamate-induced cell death program. Amino Acids 19, 253 /261. Uberti, D., Grilli, M., Memo, M., 2000b. Contribution of NFkappaB and p53 in the glutamate-induced apoptosis. Int. J. Dev. Neurosci. 18, 447 /454. Umar, A., Buermeyer, A.B., Simon, J.A., Thomas, D.C., Clark, A.B., Liskay, R.M., Kunkel, T.A., 1996. Requirement for PCNA in DNA mismatch repair at a step preceding DNA resynthesis. Cell 87, 65 /73. Zhang, H., Richards, B., Wilson, T., Lloyd, M., Cranston, A., Thorburn, A., Fishel, R., Meuth, M., 1999. Apoptosis induced by overexpression of hMSH2 or hMLH1. Cancer Res. 59, 3021 /3027.