Mutational hotspot in the p53 gene in human hepatocellular carcinomas

Tumor suppressor genes

Tumor suppressor genes are negative regulators of cell growth. When their normal function is compromised, absence of their inhibitory effects can lead to unrestrained cell cycling and growth. Strong evidence now confirms that loss of proper function of these genes is a common occurrence leading to cancer. Their failure can be caused by alterations in the gene DNA or malfunction of their protein products. The recent extraordinary accumulation of knowledge about these genes reveals that normal carcinogenesis represents breakdown of normal regulatory processes.

p53 gene mutations in gastric and esophageal cancers

The presence of point mutation of the p53 gene in exons 5, 6, 7, and 8 was examined in 10 cases of gastric adenocarcinoma and 5 cases of esophageal squamous cell carcinoma by polymerase chain reaction and direct nucleotide sequencing. Mutations of the p53 gene were found in 5 cases of gastric cancer and 4 cases of esophageal cancer. The mutations in the stomach cancers consisted of four missence mutations (exons 5 and 8) and one frame shift (exon 7). In the esophageal cancers, three missence mutations (exons 6, 7, and 8) and one point mutation within the splice donor site of intron 5 were found. Of the seven missence mutations in the two cancers, five showed the transition from G to A and two from G to T. All these changes occurred in the highly conserved region of the p53 protein. These results suggest that mutations of the p53 gene are genetic events in the pathogenesis of gastric adenocarcinoma and esophageal squamous cell carcinoma.

Singlet oxygen induced mutation spectrum in mammalian cells

In order to characterize the molecular nature of singlet oxygen (1O2) induced mutations in mammalian cells, a SV40-based shuttle vector (pi SVPC13) was treated with singlet oxygen arising from the thermal decomposition of the water-soluble endoperoxide of 3,3'-(1,4-naphthylidene) dipropionate (NDPO2). After the passage of damaged plasmid through monkey COS7 cells, the vector was shuffled into E. coli cells, allowing the screening of supF mutants. The mutation spectrum analysis shows that single and multiple base substitutions arose in 82.5% of the mutants, the others being rearrangements. The distribution of mutations within the supF gene is not random and some hotspots are evident. Most of the point mutations (98.4%) involve G:C base pairs and G:C to T:A transversion was the most frequent mutation (50.8%), followed by G:C to C:G transversion (32.8%). These results indicate that mutagenesis in mammalian cells, mediated by 1O2-induced DNA damage, is targeted selectively at guanine residues.

Prognostic implications of p53 protein, epidermal growth factor receptor, and Ki-67 labelling in brain tumours

The expression of p53 protein, epidermal growth factor receptor (EGFR), and Ki-67 nuclear antigen was examined by immunohistochemistry in biopsies of 16 types of human brain tumours, including 43 astrocytomas. P53 protein, almost certainly its mutant form, was expressed in seven of the 16, and EGFR in 11 of the 16 types of tumours. In astrocytomas both the proportion of tumours which expressed p53 or EGFR increased with grade of malignancy as did the mean Ki-67 labelling index (LI): p53-0% in grade 1, 17% in grade 2, 38% in grade 3, 65% in grade 4; EGFR-0% in grade 1, 33% in grade 2, 85% in grade 3, 95% in grade 4; mean Ki-67 L1-1.1% in grades 1 and 2, 8.3% in grade 3, and 13.4% in grade 4. Astrocytomas which expressed p53 or EGFR had a significantly higher Ki-67 LI at P less than 0.05 (11.8% and 10.7%, resp.) than those that did not (6.2% or 4.1%, resp.). Patients with astrocytomas expressing p53 or EGFR had a significantly reduced survival (P = 0.035 and P = 0.007, resp.): only 11% of the p53 + ve and 13% of the EGFR + ve patients were alive at 100 weeks following diagnosis compared to 36% of p53-ve or 60% of EGFR-ve patients. Patients with Ki-67 LI greater than 5% had a reduced survival (P less than 0.0001)--none survived beyond 86 weeks following diagnosis, whilst 63% of patients with less than 5% positive cells were still alive at 100 weeks. The univariate analysis showed that in astrocytomas expression of p53 mutants, EGFR protein, and Ki-67 greater than 5% are associated with malignant progression and poor prognosis. The multivariate analysis revealed that only tumour grade and Ki-67LI were independent prognostic factors for survival.

Evolutionary consequences of nonrandom damage and repair of chromatin domains

Some evolutionary consequences of different rates and trends in DNA damage and repair are explained. Different types of DNA damaging agents cause nonrandom lesions along the DNA. The type of DNA sequence motifs to be preferentially attacked depends upon the chemical or physical nature of the assaulting agent and the DNA base composition. Higher-order chromatin structure, the nonrandom nucleosome positioning along the DNA, the absence of nucleosomes from the promoter regions of active genes, curved DNA, the presence of sequence-specific binding proteins, and the torsional strain on the DNA induced by an increased transcriptional activity all are expected to affect rates of damage of individual genes. Furthermore, potential Z-DNA, H-DNA, slippage, and cruciform structures in the regulatory region of some genes or in other genomic loci induced by torsional strain on the DNA are more prone to modification by genotoxic agents. A specific actively transcribed gene may be preferentially damaged over nontranscribed genes only in specific cell types that maintain this gene in active chromatin fractions because of (1) its decondensed chromatin structure, (2) torsional strain in its DNA, (3) absence of nucleosomes from its regulatory region, and (4) altered nucleosome structure in its coding sequence due to the presence of modified histones and HMG proteins. The situation in this regard of germ cell lineages is, of course, the only one to intervene in evolution. Most lesions in DNA such as those caused by UV or DNA alkylating agents tend to diminish the GC content of genomes. Thus, DNA sequences not bound by selective constraints, such as pseudogenes, will show an increase in their AT content during evolution as evidenced by experimental observations. On the other hand, transcriptionally active parts may be repaired at rates higher than inactive parts of the genome, and proliferating cells may display higher repair activities than quiescent cells. This might arise from a tight coupling of the repair process with both transcription and replication, all these processes taking place on the nuclear matrix. Repair activities differ greatly among species, and there is a good correlation between life span and repair among mammals. It is predicted that genes that are transcriptionally active in germ-cell lineages have a lower mutation rate than bulk DNA, a circumstance that is expected to be reflected in evolution. Exception to this rule might be genes containing potential Z-DNA, H-DNA, or cruciform structures in their coding or regulatory regions that appear to be refractory to repair.(ABSTRACT TRUNCATED AT 400 WORDS)

An infrequent point mutation of the p53 gene in human nasopharyngeal carcinoma

Point mutations in the p53 gene have been detected in a variety of human cancers; the mutations are clustered in four "hot-spots" located in the coding region of exons 5, 7, and 8, which coincide with the four most highly conserved regions of the gene. We report the finding of a heterozygous G----C mutation at codon 280 (exon 8), position 2, of the p53 gene in a nasopharyngeal carcinoma (NPC) cell line, originating from Guangdong, a province in the People's Republic of China that leads the world in NPC incidence. A survey of nasopharyngeal tissues and NPC biopsies revealed that 1 out of 12 NPC samples from Hunan, another province in the People's Republic of China with high NPC incidence, had the same heterozygous mutation at codon 280 of p53, and none of 10 biopsies from Taiwan showed a mutation within exons 5-8 of the p53 gene. No other alteration of gene structure, including gross rearrangement or loss of heterozygosity or abnormality of gene expression was detected in NPC cell lines or NPC biopsies. We conclude from this study that mutational or other alterations of the p53 gene are not common in nasopharyngeal carcinogenesis and that a codon-280 mutation of p53 may be involved in less than 10% of NPC cases. This result contrasts with the relatively high frequency of p53 mutations associated with several other human carcinomas and suggests the importance of other genes in NPC genesis.

Aflatoxin exposure, malaria and hepatitis B infection in rural Gambian children

Aflatoxin-albumin adduct levels were measured in serum samples obtained from a group of Gambian children. The relationships between exposure to aflatoxin and the prevalence of malaria, between exposure and humoral and cellular responses in vitro to defined malaria antigens and, amongst children with evidence of exposure to hepatitis B infection, between aflatoxin and carriage of the hepatitis B surface antigen (HBsAg), were assessed. Aflatoxin-albumin adduct was found in nearly all serum samples collected during a survey performed at the end of the dry season and levels of adduct were generally high (up to 720 pg aflatoxin-lysine equivalent/mg albumin). Higher levels of aflatoxin-albumin adduct were detected in Wollof children than in children of other ethnic groups and marked variation in mean adduct levels between villages was observed. Aflatoxin-albumin adduct levels were higher in children who were HbsAg positive and in children with Plasmodium falciparum parasitaemia than in controls. However, levels of adduct had no consistent effect on either malaria-specific antibody responses, lymphoproliferative responses in vitro, or morbidity from malaria during the subsequent rainy season. Much lower levels of aflatoxin-albumin adduct were detected in repeat samples obtained at the end of the rainy season. There was poor correlation between dry and rainy season levels of adduct in individual children. We have shown that Gambian children are exposed to high levels of aflatoxin. The seasonal variation of aflatoxin-albumin adduct and marked fluctuation of adduct with time in individual children need to be considered in the future planning of epidemiological studies using this marker of exposure.(ABSTRACT TRUNCATED AT 250 WORDS)

Hepatitis B virus infection and primary hepatocellular carcinoma

For many years, epidemiological studies have demonstrated a strong link between chronic hepatitis B virus (HBV) infection and the development of primary hepatocellular carcinoma (PHC). Other hepatocarcinogens such as hepatitis C virus and aflatoxin also contribute to hepatocarcinogenesis either in conjunction with HBV infection or alone. Cellular and molecular biological studies are providing explanations for the HBV-PHC relationship, and models are now being formulated to further test the relative importance of various factors such as viral DNA integration, activation of oncogenes, genetic instability, loss of tumor suppressor genes, and trans-activating properties of HBV to the pathogenesis of PHC. Further research will probably define more than a single mechanism whereby chronic HBV infection results in PHC.

Rat p53 gene mutations in primary Zymbal gland tumors induced by 2-amino-3-methylimidazo[4,5-f]quinoline, a food mutagen

There are reports of p53 gene mutations in various human cancers but not in rat tumor cell lines or rat primary tumor tissue. We found a p53 gene mutation in a cell line of a spontaneous squamous cell carcinoma of the rat Zymbal gland, SCC131, at codon 171 by direct sequencing of cDNA fragments amplified by PCR. We tested for p53 gene mutations in 15 primary Zymbal gland tumors induced by 2-amino-3-methylimidazo[4,5-f]quinoline by single-strand conformation polymorphism analysis of the PCR-amplified cDNA products. Samples of four tumors showed mobility shifts. Direct sequencing revealed that all these tumors had mutations in conserved regions or in scattered conserved residues. Single-strand conformation polymorphism analysis of cDNA suggested that mRNA from the wild-type allele of the p53 gene was not present in tumor cells of three of four positive cases, although genomic DNA analysis indicated that the wild-type allele was retained in all the cases. All mutations were found at a guanine base: three mutations were guanine----pyrimidine transversions and one was a deletion of a guanine base within a G+C-rich sequence. These findings indicate that 2-amino-3-methylimidazo[4,5-f]quinoline may be directly involved in induction of these mutations by forming DNA adducts at various sites in the p53 gene.

Mutations in the P53 tumour suppressor gene in primary lung cancer in Japan

The reverse transcription-polymerase chain reaction-single strand conformation polymorphism (RT-PCR-SSCP) analysis and sequencing were used to examine p53 gene alterations in 18 surgical specimens of primary lung cancers obtained in Japan. Somatic mutations resulting in amino acid changes were found in eight of the 18 cases (44%). Seven missense mutations were located in amino acid-conserved domains or their vicinities (codons 110 to 307). Most mutations were found at G-C pairs, suggesting that specific carcinogens are involved in the etiology of lung cancer. The p53 mutations showed a significant association with a history of smoking (P = 0.0294). We suggest that the p53 mutations may be associated with smoking-induced lung carcinogenesis.

Induction of apoptosis by wild-type p53 in a human colon tumor-derived cell line

A wild-type p53 gene under control of the metallothionein MT-1 promoter was stably transfected into human colon tumor-derived cell line EB. Repeated inductions of the metallothionein wild-type p53 gene with zinc chloride results in progressive detachment of wild-type p53 cells grown on culture dishes. Examination at both the light and electron microscopic level revealed that cells expressing wild-type p53 developed morphological features of apoptosis. DNA from both attached and detached cells was degraded into a ladder of nucleosomal-sized fragments. Expression of wild-type p53 inhibited colony formation in soft agar and tumor formation in nude mice. Furthermore, established tumors in nude mice underwent regression if wild-type p53 expression was subsequently induced. Regressing tumors showed histological features of apoptosis. Thus, regression of these tumors was the result of apoptosis occurring in vivo. Apoptosis may be a normal part of the terminal differentiation program of colonic epithelial cells. Our results suggest that wild-type p53 could play a critical role in this process.

Clonal p53 mutation in primary cervical cancer: association with human-papillomavirus-negative tumours

Analyses of cancer cell lines and of anal cancers suggest an inverse correlation between infection with human papillomavirus (HPV) and somatic mutation of the p53 tumour-suppressor gene. We have investigated this association in primary cervical tumours. Tumour-tissue samples from 28 women with primary cancer of the cervix were analysed for presence of HPV sequences and for somatic mutations of the p53 gene. Southern blot analysis and the polymerase chain reaction (PCR) showed that 25 of the tumours contained HPV sequences; 20 were HPV16 positive and 5 HPV18 positive. 17 tumours subjected to restriction fragment length polymorphism analysis for the short arm of chromosome 17 showed no evidence of allelic deletion. Sequencing of the entire coding region of the p53 gene by asymmetric PCR detected heterozygous point mutations in only 3 HPV-negative tumours. By contrast, in 21 HPV-positive cancers the p53 sequence was wild-type throughout. Our data indicate that loss of wild-type p53 function is important in the pathology of cervical cancer and that in the absence of an HPV-encoded gene product that mediates loss of p53 function, somatic mutation of the gene is required. This pattern of p53 mutation may partly explain the apparently worse prognosis of HPV-negative cervical cancers.

Two-dimensional electrophoretic analysis of transformation-sensitive polypeptides during chemically, spontaneously, and oncogene-induced transformation of rat liver epithelial cells

Recently, we described the establishment of a computerized database of rat liver epithelial (RLE) cellular polypeptides (Wirth et al., Electrophoresis, 1991, 12, 931-954). This database has now been expanded to include the analysis of cellular polypeptide alterations during chemically (aflatoxin B1; AFB), spontaneously, and oncogene (v-Ha-ras, v-raf, and v-myc/v-raf)-induced transformation of RLE cells. Two-dimensional mapping of [35S]methionine-labeled whole cell lysate, cell-free in vitro translation products and [32P]orthophosphate-labeled polypeptides revealed subsets of polypeptides specific for each transformation modality. A search of the RLE protein database indicated the specific subcellular location for the majority of these transformation-sensitive proteins. Significant alterations in the expression of the extracellular matrix protein, fibronectin, as well as tropomyosin- and intermediate filament-related polypeptides (vimentin, beta-tubulin, the cytokeratins, and actin) were observed among the various transformant cell lines. Immunoprecipitation and Western immunoblot analysis of tropomyosin expression in four individual AFB-, as well as four spontaneously induced, and each of the oncogene-transformed cell lines indicated that five major tropomyosin (Tm 1-5) isoforms were variably expressed in the various cell lines, including one polypeptide tentatively identified as Tm6. Whereas alterations in tropomyosin expression appeared to be transformation-specific, alterations in the individual intermediate filament polypeptides were related more to the differentiation state of the individual cell lines rather than to the transformation phenotype. These studies extend our earlier efforts toward the establishment of a comprehensive computerized database of RLE cellular proteins and demonstrates how such a database may serve as a useful source for studies concerning the regulation of growth and differentiation as well as transformation of RLE cells.

Comparative analysis of p53 protein immunoreactivity in prostatic, lung and breast carcinomas

In this study we analysed the expression of p53 protein in a total of 143 carcinomas immunohistochemically. These consisted of 34 prostatic adenocarcinomas, 59 lung and 50 breast carcinomas. In 28 cases, an average of 2-3 additional sections from different tumour areas were analysed. Forty-nine of the 143 carcinomas (34%) showed typical nuclear immunoreactivity by immunohistochemical staining with the p53 antibody CM-1. Two of the 34 prostatic carcinomas (6%) were p53 positive while 25 of the 59 lung carcinomas (43%) and 22 of the 50 breast carcinomas (44%) showed positivity for p53. By grade: 49% of grade III tumours, 36% of grade II and 5% of grade I tumours were p53 positive. There were significantly more p53-positive cases in grade II-III tumours than in grade I tumours (P = 0.001) when all tumours were taken into account. Further, there were significantly more p53-positive cases in grade III than in grade I-II tumours (P = 0.001). In lung tumours there were significantly more p53-positive cases in grade II-III tumours than in grade I tumours (P = 0.018). Similarly, there were significantly more p53-positive tumours in grade III breast tumours than in grade I-II tumours (P = 0.003). The low incidence of p53 positivity in prostate carcinomas suggests that mutations of the p53 gene are not as frequent in the neoplastic transformation of these tumours as in lung or breast carcinomas. The association of p53 positivity with tumours of higher grade suggests that p53 mutations lead to tumours of a more aggressive type. The analysis of tumours by multiple sections indicates that p53 positivity is not evenly distributed in tumour tissue. Therefore, analysis of additional tumour areas may reveal positivity some cases, which is not evident if only one section is studied.

p53 expression and prognosis in gastric carcinoma

Abnormalities of the p53 gene have been identified in many malignancies, with reports of aberration in over half of colorectal, lung, breast and hepatocellular carcinoma cases. The normal gene acts as a recessive oncogene, while mutations change the apparent function to that of a dominant oncogene. In this investigation a 3-layered immunoperoxidase technique was applied to routinely fixed and paraffin-embedded tissue sections from 125 gastric carcinomas, using a polyclonal anti-p53 antibody (CM-I). We found that 57% of these carcinomas expressed high levels of p53 protein (positive nuclear staining). Survival analysis revealed a strong association between p53 status of the tumour and patient survival time after diagnosis (p = 0.02, Mantel-Cox Test); odds ratio of death, 2.09 (95% confidence interval 1.02 to 4.25). The 5-year survival of patients with p53-expressing tumours was 24%, compared with 56% for those non-p53-expressing tumours (the median survival times were 13 and 102 months, respectively).

Growth inhibition by anchorage-deficiency is associated with increased level but reduced phosphorylation of mutant p53

Human breast carcinoma MCF-7 cells seeded on type I collagen-coated dishes were provided with an anchor via the collagen receptor, integrin, and grew as actively as those in plastic tissue culture dishes. In contrast, cells seeded on a layer of soft agar became anchorage-deficient and their growth was significantly inhibited, although the cell viability and the cell cycle distribution were unaffected. Immunoprecipitation analysis revealed that mutant p53 was phosphorylated at tyrosine in the anchorage-provided cells. In contrast, the p53 in the anchorage-deficient cells was present in 2-fold greater amount, but was phosphorylated to a lesser extent. Addition of a potent protein-tyrosine kinase inhibitor, herbimycin A, to the anchorage-provided cells caused an elevated level of p53, and inhibitions of cell proliferation and p53 phosphorylation, without interfering with the cell adhesion to the substratum. These results demonstrated that the growth inhibition by anchorage-deficiency or by herbimycin A is associated with an elevated p53 level and reduced p53 phosphorylation at tyrosine.

Mutations of p53 and ras genes in radon-associated lung cancer from uranium miners

Radon increases the risk of lung cancer in smoking and non-smoking underground miners. To investigate the mutational spectrum associated with exposure to high levels of radon, we sequenced exons 5-9 of the p53 tumour suppressor gene and codons 12-13 of the Ki-ras protooncogene in 19 lung cancers from uranium miners exposed to radon and tobacco smoke. Mutations were not found in Ki-ras, but 9 p53 mutations, including 2 deletions, were found in 7 patients by direct DNA sequencing after polymerase chain reaction amplification of DNA from formalin-fixed, paraffin-embedded tissue. In tumours from 5 patients, the mutation produced an aminoacid change and an increased nuclear content of p53 protein. The tumours with either a stop codon or frame-shift deletion in the p53 gene were negative by immunohistochemistry. None of the mutations were G:C to T:A transversions in the coding strand of the p53 gene, which are the most frequent base substitutions associated with tobacco smoking, and none were found at the hotspot codons described in lung cancer. The observed differences from the usual lung cancer mutational spectrum may reflect the genotoxic effects of radon.

Detection of a Ha-ras point mutation by polymerase chain reaction-single strand conformation polymorphism analysis in 2-amino-3,4-dimethylimidazo[4,5-f]quinoline-induced mouse forestomach tumors

Ha-ras activation in forestomach squamous cell carcinomas of CDF1 mice induced by 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ), one of the heterocyclic amines isolated from broiled sardine was analyzed. Mutations were detected in two of three primary original carcinomas and two of four cell lines derived from other independent carcinomas by the polymerase chain reaction followed by analysis of single strand conformation polymorphism and direct sequencing. All the mutations detected were G----T transversions at the second letter of codon 13 resulting in an amino acid change from Gly to Val. This finding together with the previous reports on squamous cell carcinomas of the rat Zymbal gland suggest that MeIQ induces a specific type of mutation at a specific site of the Ha-ras gene during squamous cell carcinogenesis, in a species-independent manner.

p53 mutations are not selected for in simian virus 40 T-antigen-induced tumors from transgenic mice

Many diverse tumors contain cells that select for mutations at the p53 gene locus. This appears to be the case because the p53 gene product can act as a negative regulator of cell division or a tumor suppressor. These mutations then eliminate this activity of the p53 gene product. The simian virus 40 (SV40) large T antigen binds to p53 and acts as an oncogene to promote cellular transformation and initiate tumors. If the binding of T antigen to the p53 protein inactivated its tumor suppressor activity, there would be no selection pressure for p53 mutants to appear in tumors. To test this idea, transgenic mice that carried and expressed the SV40 large T-antigen gene were created. Expression of the T antigen was directed to the liver, using the albumin promoter, and the choroid plexus, using the SV40 enhancer-promoter. A large number of papillomas (indicated in parentheses) of the choroid plexus (14), hepatocellular carcinomas (5), liver adenomas (10), and tumors of clear-cell foci (5) were examined for mutant and wild-type p53 genes and gene products. In all cases, the tumor extracts contained readily detectable T-antigen-p53 protein complexes. A monoclonal antibody specifically recognizing the wild-type p53 protein (PAb246) reacted with p53 in every tumor extract. A monoclonal antibody specifically recognizing mutant forms of the p53 protein (PAb240) failed to detect p53 antigens in these extracts. Finally, p53 partial cDNAs were sequenced across the regions of common mutations in this gene, and in every case only the wild-type sequence was detected. These results strongly support the hypothesis that T antigen inactivates the wild-type p53 tumor-suppressing activity and there is no need to select for mutations at the p53 locus.

p53 point mutations in primary human gastric carcinomas

p53 point mutations in primary gastric carcinomas were analyzed by performing cDNA deoxynucleotide sequencing of the gene. Out of 16, 9 (56.3%) primary gastric carcinoma cases, including early cancer, showed one or more p53 point mutations in their open-reading frame, and 4 out of 9 cases had a p53 point mutation within highly conserved domains. The characteristics of the p53 mutation spectrum observed in primary tumors were (a) frequent mutation at an A:T pair (50%, 7 out of 14 mutations), (b) high transversion incidence (29%, 4 out of 14 mutations), (c) no transition at CpG, and (d) no G:C to T:A transversion. Our results suggest that p53 mutation is a common event in gastric carcinoma occurring from the early stage of progression with its specific mutation spectrum.

Hereditary and acquired p53 gene mutations in childhood acute lymphoblastic leukemia

The p53 gene was examined in primary lymphoblasts of 25 pediatric patients with acute lymphoblastic leukemia by the RNase protection assay and by single strand conformation polymorphism analysis in 23 of 25 cases. p53 mutations were found to occur, but at a low frequency (4 of 25). While all four mutations were identified by single strand conformation polymorphism, the comparative sensitivity of RNase protection was 50% (2 of 4). Heterozygosity was retained at mutated codons in 3 of 4 cases. One pedigree was consistent with the Li-Fraumeni syndrome, and bone marrow from both diagnosis and remission indicated a germline G to T transversion at codon 272 (valine to leucine). Although members of another family were affected with leukemia, a 2-bp deletion in exon 6 was nonhereditary. The other two nonhereditary p53 mutations included a T to G transversion at codon 270 (phenylalanine to cysteine) and a G to C transversion at codon 248 (arginine to proline). These data support the role of both hereditary and acquired p53 mutations in the pathogenesis and/or progression of some cases of childhood acute lymphoblastic leukemia.

Oncogenes and cancer suppressor genes

Cancer is caused by the malfunction of genes that regulate cell proliferation. Two kinds of regulatory genes have been discovered in the search for cancer genes: those that promote growth, called oncogenes, and those that suppress growth, called anti-oncogenes or cancer suppressor genes. The retroviruses that cause animal cancers contain oncogenes coding for growth-promoting signals. These retroviruses rarely cause human cancer but study of their oncogenes has allowed identification of many human cancer genes. These genes code for growth factors, growth factor receptors, cytoplasmic proteins, and nuclear proteins. The complete sequence of cellular growth control begins when a growth factor binds to its receptor and acts directly or indirectly through a G protein and second messenger to induce phosphorylation (activation) of an intracellular protein that ultimately alters the expression of the genes necessary to initiate cell division. At each step in the complex sequence that up-regulates cell division, there is an opposite down-regulating activity produced by the protein products of anti-oncogenes or cancer suppressor genes. These proteins do this by binding to and inactivating transcription factors that initiate DNA synthesis or by directly inactivating the molecules activated by the oncogene products. When this carefully orchestrated and regulated cell control process goes awry because one or more of the proteins in the sequence has been altered by a mutated gene, the cell divides in an uncontrolled manner and malignancy results. It is thought that most human cancers result from a combination of genetic changes that must include both the absence of the protein products of cancer suppressor genes and the presence of abnormal products of oncogenes. The work of Volgelstein and coworkers at Johns Hopkins University has provided the best insight so far into the complex pathogenesis of a common tumor, colon cancer. Carcinogenesis in colon cancer requires a sequence of events that involves more than five genes. Understanding of these pathogenic mechanisms should improve cancer diagnosis and treatment.

Alterations in the structural gene and the expression ofp53 in rat liver tumors induced by aflatoxin B1

Rat hepatocellular carcinomas (HCCs) induced by aflatoxin B1 (AFB) treatment were examined for changes in the p53 tumor suppressor gene and in p53 suppressor gene expression. A high proportion of HCCs (nine of 11 tumors in six of eight animals) exhibited new p53 restriction fragments, indicating genomic alterations of one of the p53 alleles. Each tumor with an altered p53 restriction-fragment pattern exhibited a new fragment in one of two size classes (3 kb or 7 kb with EcoRI digestion) that were missing portions of the 3' end of the p53 gene. These findings indicate that apparently similar genomic rearrangements or deletions occurred independently in AFB-induced tumors. When compared with nontumor liver tissue from the same animal, the tumors with p53 gene alterations showed dramatically reduced levels of p53 mRNA and protein and greatly increased levels of histone H2B and retinoblastoma tumor suppressor (Rb) mRNA. In two HCCs showing no evidence of p53 restriction-fragment alterations, mutant p53 protein was detected. Mutant protein was also detected in two liver samples containing an adenoma and altered foci. These data suggest that alterations of the p53 tumor suppressor gene are involved in the induction of rat HCC by AFB.

Tumours of the liver

One of the major debates in hepatocellular carcinogenesis at present is whether the hepatitis-B and -C viruses are directly carcinogenic or exert their effect indirectly by causing chronic necro-inflammatory hepatic disease, which in turn is responsible for malignant transformation of hepatocytes. This debate has been fueled by the observation that hepatitis C virus is a single-stranded RNA virus with no precedent for inducing cancer but with a marked propensity to cause chronic necro-inflammatory hepatic disease and by the findings in Chisari's transgenic mouse model, which suggest that severe and prolonged hepatocellular injury per se induces a proliferative response that progresses to tumour formation. Recent reports of a guanine to thymine mutation of the third base of codon 249 of the tumour suppressor gene, p53, in 50% of patients with hepatocellular carcinoma in regions of high aflatoxin exposure, and mutagenic experiments showing that aflatoxin B1 binds particularly to guanine residues in G-C-rich domains and that codon 249 is a preferred target have suggested a mechanism whereby aflatoxin might induce malignant transformation.

Kinds of mutations induced by aflatoxin B1 in a shuttle vector replicating in human cells transiently expressing cytochrome P4501A2 cDNA

Transient expression of rat liver cytochrome P450lA2 cDNA was combined with the use of a shuttle vector as a mutational target to determine the frequency and types of mutation caused by the conversion of aflatoxin B1 into genotoxic metabolites within human cells. Ad293 cells were first transfected with p91-lA2, a rat liver P450lA2 cDNA expression vector, or with p91-lA2(i) (a control vector that has the P450 cDNA in the inverted orientation) and incubated for 24 h to permit P450lA2 accumulation. Cells were then transfected with the pS189 shuttle-vector plasmid, which carries the Escherichia coli supF gene as a mutational target, and incubated for a further 24 h in the presence of aflatoxin B1 to permit promutagen activation and pS189 replication. In shuttle vectors replicated in p91-lA2-transfected cells, the supF point-mutation frequency increased with increasing concentration of aflatoxin B1. This frequency was nine to 23 times greater than the background point-mutation frequency obtained with aflatoxin B1-treated control (p91-lA2(i)-transfected) cells. The large majority of the aflatoxin B1-induced supF point mutations were base substitutions, mostly G:C----T:A transversions. This mutagenesis system permits the molecular analysis of mutations induced by specific P450/promutagen pairs in a shuttle vector replicating in human cells and will permit the investigation of host cell mechanisms involved in the generation of these mutations.

TP53 tumor suppressor gene: a model for investigating human mutagenesis

More than 350 independent point mutations of the TP53 gene, found in a wide variety of human cancers, were compiled and analysed. From this study, we confirm the presence of four hot-spot regions which colocalize with some highly conserved domains of the protein. We also define a new hot-spot region which is observed predominantly in lung tumors. Analysis of the mutational events suggests the direct involvement of environmental carcinogens in lung tumors and hepatocarcinomas, and spontaneous mutagenesis generating essentially CpG transitions in most of the remaining ones. Furthermore, we demonstrate in this work that the TP53 gene is an informative model with which to study the molecular mechanisms of mutagenesis in the human genome.

p53 mutations are absent from carcinogen-induced mouse liver tumors but occur in cell lines established from these tumors

Mutations in the p53 gene are frequent genetic alterations in human hepatocellular carcinomas. We have examined, by single-strand conformation polymorphism analysis of polymerase chain reaction products, a total of 93 carcinogen-induced liver tumors from mice of three different strains (C3H/He, C57BL/6J, and B6C3F1) for the presence of p53 aberrations. Hepatoma lines, established from 12 liver tumors, were also included in the analysis. While structural aberrations of the p53 gene were not detected in any of the primary mouse liver tumors analyzed, single-base substitutions occurred at different locations within the p53 gene in three of the cell lines during in vitro propagation. One hepatoma line carried two point mutations on separate alleles. All four mutations were either G:C----C:G or C:G----G:C transversions. Mutations at codon 61 of the c-Ha-ras gene, which were frequent in primary liver tumors from C3H/He and B6C3F1 mice, were not detected in the hepatoma lines. Our data indicate (i) that c-Ha-ras but not p53 mutations play an important role during the early stages of mouse hepatocarcinogenesis and (ii) that p53 mutations confer a selective growth advantage to the mutated hepatoma cells in vitro.

p53 gene mutations in human endometrial carcinoma

Although carcinoma of the uterine endometrium is the most frequently diagnosed malignancy of the female reproductive tract, the molecular genetic features of this tumor have yet to be described in significant detail. Since mutations of the p53 tumor suppressor gene are the single most common genetic alteration found in human malignancies, we examined the hypothesis that p53 mutations occur in human endometrial carcinoma. Sequencing analysis of exons 5-8 revealed point mutations in 3 of 21 (14%) tumors; one mutation was an unusual single-base insertion at codons 176-177, resulting in a premature stop codon, whereas the other two were CGG----TGG transitions at codon 248. Two of these tumors showed reduction to homozygosity at the p53 allele, but one tumor apparently retained heterozygosity. These data indicate that p53 mutations occur in human endometrial carcinoma, although relatively infrequently, and that loss of the normal p53 allele does not necessarily occur with point mutation of the p53 gene in this tumor type.

Long-term survival following treatment of hepatocellular carcinoma in Singapore: evaluation of Wellferon in the prophylaxis of high-risk pre-cancerous conditions

The present paper reviews several studies performed between 1977 and 1986 in Singapore on the 10-year survival outcome of treatment for stage I and II hepatocellular carcinoma (HCC). Of 801 HCC patients evaluated, only 2 survivors (0.3%) remained in complete remission for 13 and 14 years, respectively. One had received four weekly cycles of prednisolone, Adriamycin, vincristine and 5-fluorouracil for an inoperable HCC with a 10-cm diameter, and the other had received localised synchronised hepatic irradiation and Adriamycin. As follow-up, the use of localised hepatic irradiation consisting of 131I-labeled (30 mCi) iodised oil in lipiodol infused via the hepatic artery appeared to benefit patients with small residual tumours but did not affect larger tumours measuring 2 cm in diameter. Prophylactic, intermittent long-term administration of lymphoblastoid interferon-alpha (Wellferon) was carried out in pre-cancerous, high-risk hepatitis B surface antigen (HBsAg)-positive patients with cirrhosis, in immediate male relatives of liver cancer patients, and in persons who had undergone hepatic resection. In the untreated group, 10/162 (6%) cirrhotics, 3/18 (17%) male family members, and 6/10 (60%) post-resection cases developed single or multiple HCCs within 1 year of screening done at 3-month intervals on the basis of alpha-fetoprotein (AFP) levels and real-time hepatic ultrasonography. In contrast, none of the Wellferon-treated group consisting of 518 cirrhotic patients, 82 male relatives of HCC patients and 20 post-resection cases developed HCC. Two HBsAg-positive individuals who had not been treated with interferon (IFN) developed hepatic nodules which that showed dysplasia, AFP elevation and chromosomal changes. These studies demonstrate the poor results of late diagnosis and show that early intervention and prophylaxis with Wellferon can reduce the incidence of HCC in high-risk persons. In addition, transhepatic chemoembolisation and liver resection are suitable methods for treating small HCCs (single or multiple) that are detected by screening. However, some of these early-detected HCCs remain highly malignant. Prophylactic treatment of pre-cancerous conditions appears to be a better option as a long-term programme for HCC.

Murine p53 intron sequences 5-8 and their use in polymerase chain reaction/direct sequencing analysis of p53 mutations in CD-1 mouse liver and lung tumors

Inactivating point mutations and small deletions in the p53 tumor suppressor gene have been found in human liver and lung tumor--derived cell lines and tumors. However, little evidence has been reported concerning inactivation or mutation of the p53 gene in mouse primary tumors. To examine CD-1 mouse liver and lung tumors for mutations in the p53 gene, we first sequenced p53 introns 5-8 so that polymerase chain reaction amplification and sequencing primers located within the introns could be prepared. Use of these primers prevented amplification of the mouse p53 pseudogene and allowed sequencing of exons 5-8 in their entirety as well as their intron-exon junctions. DNA isolated from CD-1 mouse tumors was amplified and directly sequenced using nested primers. Nine spontaneous hepatocellular carcinomas (HCCs) and 34 chemically induced HCCs (induced by single intraperitoneal injections of N-nitrosodiethylamine [DEN] [8 HCCs], 7,12-dimethylbenz[a]anthracene [DMBA] [8 HCCs], 4-aminoazobenzene [8 HCCs], and N-OH-2-acetylaminofluorene [10 HCCs]) were examined for mutations in exons 5-8 of the p53 gene. In addition, 12 spontaneous, 10 DMBA-induced, and 13 DEN-induced lung adenocarcinomas or adenomas were analyzed for mutations. No mutations were found in any of the tumors examined. However, a mutation was demonstrated at codon 135 in the positive-control plasmid LTRp53cG(val). The results of this study suggest that inactivation of p53 is unlikely to play a major role in murine lung or liver carcinogenesis. However, inactivation of p53 may occur at a very low frequency, or it may occur as a late event and therefore be present in only a very small number of the tumor cells, rendering it undetectable by this method. Lastly, although few p53-inactivating mutations are found outside of exons 5-8 in human tumors, it is possible that these murine tumors contained mutations outside of this region and were therefore missed by our approach.

Metabolic activation of carcinogens

Most chemical carcinogens are not active in themselves but require bioactivation to electrophiles that bind covalently to DNA and often act by producing mutations. In recent years it has been realized that mutations can be important at many stages of carcinogenesis. A variety of different enzymes are involved in bioactivation reactions, which include oxidation, reduction, thiol conjugation, acetyl transfer, sulfur transfer, methyl transfer, glucuronosyl transfer, and epoxide hydrolysis. These processes often occur in concert with a single carcinogen. Humans vary considerably in activities of these enzymes and this variation may contribute to differences in risk.

Suppression of human nasopharyngeal carcinoma cell growth in nude mice by the wild-type p53 gene

Wild-type and mutant human p53 genes were transfected into the nasopharyngeal carcinoma (NPC) cell line CNE-3. Tumorigenicity in nude mice showed that the tumor resulting from the cells transfected with the wild-type p53 gene grew more slowly and was smaller than that from the cells transfected with mutant p53 gene and that from control CNE-3 cells. In contrast, the tumor from the cells transfected with the mutant p53 gene grew faster than that produced by cells transfected with the wild-type p53 gene and that produced by control CNE-3 cells. The results demonstrate that the wild-type p53 gene could inhibit the NPC cell growth in nude mice and the mutant p53 gene could enhance the NPC cell growth in nude mice. The p53 gene may also play an important role in the pathogenesis of NPC.

Methylation, mutation and cancer

The fifth base in human DNA, 5-methylcytosine, is inherently mutagenic. This has led to marked changes in the distribution of the CpG methyl acceptor site and an 80% depletion in its frequency of occurrence in vertebrate DNA. The coding regions of many genes contain CpGs which are methylated in sperm and serve as hot spots for mutation in human genetic diseases. Fully 30-40% of all human germline point mutations are thought to be methylation induced even though the CpG dinucleotide is under-represented and efficient cellular repair systems exist. Importantly, tumor suppressor genes such as p53 also contain methylated CpGs and these serve as hot spots for mutations in some, but not all, human cancers. Comparison of the spectrum of mutations present in this gene in different human cancers allows for predictions to be made on the molecular mechanisms of tumorigenesis.