Xiusheng Qin

DNA repair and cancer: Lessons from mutant mouse models

DNA damage, if the repair process, especially nucleotide excision repair (NER), is compromised or the lesion is repaired by some other error‐prone mechanism, causes mutation and ultimately contributes to neoplastic transformation. Impairment of components of the DNA damage response pathway (e.g., p53) is also implicated in carcinogenesis. We currently have considerable knowledge of the role of DNA repair genes as tumor suppressors, both clinically and experimentally. The deleterious clinical consequences of inherited defects in DNA repair system are apparent from several human cancer predisposition syndromes (e.g., NER‐compromised xeroderma pigmentosum [XP] and p53‐deficient Li‐Fraumeni syndrome). However, experimental studies to support the clinical evidence are hampered by the lack of powerful animal models. Here, we review in vivo experimental data suggesting the protective function of DNA repair machinery in chemical carcinogenesis. We specifically focus on the three DNA repair genes, O6‐methylguanine‐DNA methyltransferase gene (MGMT), XP group A gene (XPA) and p53. First, mice overexpressing MGMT display substantial resistance to nitrosamine‐induced hepatocarcinogenesis. In addition, a reduction of spontaneous liver tumors and longer survival times were evident. However, there are no known mutations in the human MGMT and therefore no associated cancer syndrome. Secondly, XPA mutant mice are indeed prone to spontaneous and carcinogen‐induced tumorigenesis in internal organs (which are not exposed to sunlight). The concomitant loss of p53 resulted in accelerated onset of carcinogenesis. Finally, p53 null mice are predisposed to brain tumors upon transplacental exposure to a carcinogen. Accumulated evidence in these three mutant mouse models firmly supports the notion that the DNA repair system is vital for protection against cancer.

Importance of DNA repair in carcinogenesis: evidence from transgenic and gene targeting studies.

We have generated transgenic mice by introducing copies of the E. coli O6-methylguanine-DNA methyltransferase gene, ada. Liver extracts from homozygotes demonstrate about three times the control enzyme activity and increase up to about eight-fold can be induced by treatment with zinc, since the metal-responsive metallothionein promoter is attached to the ada gene. Furthermore, studies of liver carcinogenesis in our transgenic mice demonstrated significantly reduced rates of development of hepatocellular tumors after treatment with dimethylnitrosamine or diethylnitrosamine. It is well known that xeroderma pigmentosum (XP) patients are deficient in DNA repair. The availability of XPA (XP group A complementing) knockout mice has enabled us to investigate the functional role of the XPA nucleotide excision repair gene in carcinogenesis in vivo, first using the mouse skin as a model system. XPA-/- mice demonstrated skin ulcers 5-7 days after 7,12-dimethylbenz[a]anthracene (DMBA) treatment and papilloma development within 4 weeks prior to promotion, skin tumor incidence being also much higher than in heterozygous and wild-type mice. Experiments targeting the lung, liver and tongue have also been conducted to answer the question of whether the internal organs of these mice are also susceptible to chemical carcinogens. For lung carcinogenesis, mice were instilled intratracheally with a small dose of benzo[a]pyrene. The pulmonary tumor incidence in XPA-/- mice was significantly higher than in XPA+/- and XPA+/+ mice. XPA-/- mice were also found to be have enhanced sensitivity to aflatoxin B1 regarding liver tumor induction. In addition, administration of 4-nitroquinoline-1-oxide in drinking water for 50 weeks resulted in tongue tumors only in XPA-/- mice. These studies, thus, provided convincing evidence that XPA mice are also sensitive to carcinogenesis in organs other than the skin.

Detection of active UV-photoproduct repair in monkey skin in vivo by quantitative immunohistochemistry

Ultraviolet-induced cyclobutane pyrimidine dimers (CPDs) and pyrimidine-pyrimidone (6-4)photoproducts in DNA were quantitatively measured in monkey skin using an immunohistochemical method with two specific monoclonal antibodies. The skins of Cynomolgus monkeys (Macaca fascicularis) were irradiated with UV light and processed for preparation of conventional formalin-fixed, paraffin-embedded histological sections. Both of the photoproducts were detectable in the nuclei of epidermal cells at doses of 500 J/m2 for UVB and 50 J/m2 for UVC, respectively, nuclear staining being clearly dose-dependent. Time course studies also showed a statistically significant decrease in nuclear staining with time after exposure to either UVB or UVC irradiation. Although only 30% of CPDs were removed from DNA in the first 24 h, about half of the (6-4) photoproducts were repaired within 3 h post-UV irradiation. Staining completely disappeared by 48 h in the (6-4) photoproduct case and by 72 h in the case of CPDs. The results suggest that epidermal cells of monkey skin can efficiently repair UV-photoproducts in DNA, but that the capacity is slightly less than in man.

Detection of ultraviolet photoproducts in mouse skin exposed to natural sunlight.

In the present study, we for the first time investigated the formation of ultraviolet (UV) photoproducts, cyclobutane pyrimidine dimers (CPDs), pyrimidine‐pyrimidone (6–4) photoproducts (64PPs) and Dewar isomers, in vivo in shaved and depilated C3H/HeN mouse skin exposed to natural sunlight (NSL) at noon for 5 min to 1 h in mid‐summer, using a highly sensitive immunohistochemical method. This method permits the quantitative analysis of UV‐photoproducts in formalin‐fixed, paraffin‐embedded sections with specific antibodies against CPDs, 64PPs and Dewar isomers. We demonstrated that the induction of CPDs in vivo in mouse skin by NSL was exposure time‐dependent, but the accumulation of 64PPs or Dewar isomers was comparatively low in the skin sections from mice exposed to NSL in vivo. The results indicate that CPDs are the main photoproducts in vivo induced by sunlight and that their formation and repair may be important in connection with carcinogenesis in sun‐exposed areas of human skin.

Protection against malignant progression of spontaneously developing liver tumors in transgenic mice expressing O6-methylguanine-DNA methyltransferase

To study the effect of O6‐methylguanine‐DNA methyltransferase (MGMT) on carcinogenesis, we have previously generated MGMT transgenic mice overexpressing the bacterial MGMT gene, ada, and demonstrated that high MGMT levels in the liver suppress induction of liver tumors after treatment with an alkylating hepatocarcinogen. To examine the effects of life‐long elevation of MGMT activity on mouse spontaneous liver tumor development, ada‐transgenic and control nontransgenic mice were compared. We also examined mutations at codon 61 of the H‐ras oncogene, reported as a hot spot in mouse liver tumors, using a direct DNA sequencing method. The results revealed no significant difference in tumor incidence or mutation spectrum, but interestingly, ada‐transgenic mice were found to have fewer malignant tumors and survived longer, indicating a possible protective role of MGMT against malignant conversion.