Multistep carcinogenesis: a 1992 perspective

Cancer prevention: past, present, future

There are many examples of different types of cancers that have been prevented by appropriate measures in the past. Most of them were related to occupational, iatrogenic or accidental factors, often as the outcome of heavy exposure of humans to specific carcinogenic agents. Cancer is a disease of DNA, and is generally associated with multiple genetic alterations, these being produced in the typical case by exposure to various carcinogens, each of which exists at minute concentrations. Thus, the impact of carcinogenic factors, xenobiotics and autobiotics, is due to their actions in concert. However, a single mutation yielding genomic instability exerts a disproportionately large influence by resulting in a large number of secondary mutational events. Epigenetic changes can also not be disregarded especially from the view point of prevention of neoplasia. The occurrence of multiple primary cancers among survivors of initial primaries, and the presence of hereditary groups with a high risk of cancer development provide a strong stimulus for establishment of effective approach for cancer prevention, which should be, in principle, multi-faceted. Therefore, a holistic approach is essential with improvement in life style including choosing a balanced diet and avoidance of cigarette smoking and other sources of carcinogens, as integral elements.

Inhibitory effects of antioxidants on formation of heterocyclic amines

It is important to search for effective antioxidants to suppress formation of mutagenic and carcinogenic heterocyclic amines (HCAs), like 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), because these HCAs are considered to be probable human carcinogens. The effects of various food-derived antioxidants on MeIQx formation were examined by their addition (0.2 mmol each) to mixtures of creatine (0.4 mmol), glycine (0.4 mmol) and glucose (0.2 mmol), and heating at 128 degreesC for 2 h. Glycine was replaced by l-phenylalanine in the case of PhIP formation. Among the 14 kinds of antioxidants tested, green tea catechins and the major component [(-)-epigallocatechin gallate], two flavonoids (luteolin and quercetin) and caffeic acid were found to clearly suppress the formation of both MeIQx and PhIP, being 3.2-75% of the level of the controls. These phenolic antioxidants also reduced the total mutagenicity of the heated mixtures. The results suggest that foodstuffs containing catechins, flavonoids and caffeic acid may suppress the formation of HCAs in cooked foods.

No direct correlation between mutant frequencies and cancer incidence induced by MeIQ in various organs of Big Blue mice

The carcinogenicity of 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ) was examined in Big Blue female mice with the genetic background of C57BL/6N. With the administration of 300 ppm of MeIQ in their diet for 92 weeks, the Big Blue female mice developed intestinal tumors and hepatocellular carcinomas. The incidences of adenocarcinomas were 42% (8/19) in the colon and 68% (13/19) in the cecum. The incidence of hepatocellular carcinomas was 84% (16/19). No carcinomas of the intestine or the liver were induced in the control group. As we previously reported, administration of 300 ppm of MeIQ in a diet for 12 weeks induced lacI mutants at the highest frequency in colonocytes, and at only less than one-tenth of the colon in cells of the liver, forestomach and bone marrow, indicating no direct correlation between the lacI mutant frequency (MF) and cancer incidence (CI). The fate of cells with lacI mutation in each organ should be taken into consideration to validate MF as an indicator of carcinogenic potency of a chemical in different organs.

pp60c-src activation in hepatocellular carcinoma of humans and LEC rats

For the related Src kinases, a close correlation exists between elevated tyrosine kinase activity and cell transformation. However, the involvement of pp60c-src in hepatocellular carcinoma (HCC) remains obscure. The aim of this study was to evaluate whether pp60c-src tyrosine kinase activity is elevated in HCC. We analyzed the kinase activity of pp60c-src in normal liver tissue, chronic hepatitis liver tissue, and tumorous and adjacent nontumorous portions of HCC tissue from patients and Long-Evans cinnamon (LEC) rats that are known to develop liver cancer spontaneously. The kinase activity of pp60c-src was rarely detected in the normal human liver tissue and chronic hepatitis liver tissue, but it was elevated in tumorous and nontumorous portions of HCC tissue. Furthermore, the kinase activity of pp60c-src was significantly elevated in tumorous tissues compared with nontumorous tissues. The kinase activity of pp60c-src was also higher in poorly differentiated HCC. In addition, the kinase activity of pp60c-src increased proportionately with the development of HCC of LEC rats. Our results suggest that activation of the protooncogene product pp60c-src may play an important role in the malignant transformation of hepatocytes in human and LEC rats, and that it may be closely related to the histopathological grading of human HCC.

Li-Fraumeni syndrome and the role of the p53 tumor suppressor gene in cancer susceptibility

Mutation of the tumor suppressor gene p53 is a molecular genetic event frequently observed in human cancer, and inactivating missense mutations usually are accompanied by the resultant overexpression of mutant p53 protein. In gynecologic cancers, p53 is also often altered; the frequency varies depending on types of cancers and where they develop. Further, human papillomavirus oncoproteins that inactivate p53 and Rb proteins play important roles in the development of several gynecologic cancers. Individuals who are heterozygous for germline mutations of the p53 gene are strongly predisposed to a variety of cancers. The identification of these individuals may have profound value in the future when therapies or chemopreventive agents specific for the p53 alteration are available. The role of p53 tumor suppressor gene in gynecologic cancers and heritable cancer susceptibility syndromes including Li-Fraumeni and Lynch II syndromes is an active and important area of study.

Experimental chemical carcinogenesis in the stomach and colon

Experimental chemical carcinogenesis in the digestive tract is reviewed, mainly on the basis of information obtained in the laboratories of the National Cancer Center Research Institute. It is generally accepted that cancer is the outcome of DNA damage, resulting in mutation, loss, amplification and recombination of genes. Gastric cancer is no exception. It was shown very early that cancer of the glandular stomach can be produced in rats by administration of N-methyl-N'-nitro-N-nitrosoguanidine (MNNG), a widely used mutagen. However, this depends on the genotype. Whereas the ACI rat is susceptible to MNNG, the Buffalo rat is resistant and this is a dominantly inherited trait. Genes responsible for the sensitivity to gastric cancer induction are at present under investigation by linkage analysis of rat genome markers. With regard to cancer in humans, our finding that cooked proteinaceous foods can give rise to a series of heterocyclic amines (HCAs) is of major significance. 2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), one of the most abundant, causes colon cancers in male rats, whereas in females it induces breast cancers. The colon cancers induced by PhIP feature a deletion of G as represented by 5-GGGA-3-->5-GGA-3 in the Apc gene, resulting in a truncated Apc molecule. Microsatellite mutations have also been found in PhIP-induced colon tumors, as in human hereditary non-polyposis colorectal cancer cases. Similarly to the case of gastric cancer production by MNNG, there is a genetic component and F344 rats are more susceptible to PhIP colon carcinogenesis than the ACI/N strain and the gene responsible is being sought. Since carcinogenesis proceeds with accumulation of genetic alteration, often involving genomic instability, exposure to any kind of carcinogenic substances, either xeno- or autobiotics, needs to be reduced as far as possible, taking account of inconvenience at the individual and socio-economical levels.

Aberrant FHIT transcripts in hepatocellular carcinomas

To study abnormalities of the FHIT gene in human hepatocellular carcinoma (HCC), eight liver cancer cell lines, 18 matched tumorous and non-tumorous tissues from patients with HCC and three normal liver tissues were analysed by microsatellite polymorphism analysis and reverse transcription of FHIT mRNA followed by polymerase chain reaction (PCR) amplification and sequencing of the products. No loss of heterozygosity at chromosome 3p14.2 as defined by markers D3S1300 and D3S1312 was detected in any of the specimens. In addition, a normal transcript of the gene without any sequence change was found to be expressed in all the cell lines, 17 of the 18 tumorous and all 21 non-tumorous liver tissues tested. Although five out of eight liver cancer cell lines (62.5%), 12 out of 18 HCC tissues (66.7%) and 8 out of 18 paired non-tumorous liver tissues (44.4%) displayed abnormal faint bands of smaller size, sequence analysis revealed that they were aberrant FHIT transcripts lacking three or more exons and might represent alternatively spliced transcripts only. In conclusion, these studies indicate that abnormalities of the FHIT gene transcripts occur in a fairly high frequency of tumorous and non-tumorous liver tissues. However, it might not be causally related to the hepatocarcinogenesis.

Inhibitory effects of diallyl disulfide or aspirin on 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine-induced mammary carcinogenesis in rats

Modifying effects of diallyl disulfide (DAD), aspirin or DL-alpha-difluoromethylornithine (DFMO) on 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP)-induced mammary carcinogenesis in SD rats were investigated. A total of 166 female rats, 6 weeks old, were divided into 8 groups. They were fed a high fat diet throughout the experiment. Starting at 7 weeks of age, groups 1-4 were given PhIP (85 mg/kg body weight in corn oil) by gavage 8 times in 10 days, and groups 5-8 were given corn oil alone. For the beginning 4 weeks, groups 2 and 5 were given DAD at 200 ppm in diet. Similarly groups 3 and 6, and groups 4 and 7 were given aspirin (400 ppm) and DFMO (400 ppm), respectively. Mammary carcinomas were only recognized in groups 1-4 at the termination (25 weeks after the start of experiment). Multiplicity (mean number/rat) of neoplasms in group 2 (PhIP+DAD, 0.90/rat) and group 3 (PhIP+aspirin, 1.37/rat) was significantly smaller than that in group 1 (PhIP alone, 2.45/ rat) (P < 0.005 and P < 0.05, respectively). These results indicate that dietary intake of DAD or aspirin during the time corresponding to initiation phase has chemopreventive potential on PhIP-induced mammary carcinogenesis in rats.

Opposite effect of stable transfection of bioactive transforming growth factor-beta 1 (TGF beta 1) versus exogenous TGF beta 1 treatment on expression of 92-kDa type IV collagenase in mouse skin squamous cell carcinoma CH72 cells

We have previously shown that transforming growth factor-beta 1 (TGF beta 1) mRNA is consistently overexpressed in squamous cell carcinomas relative to normal mouse skin. Here we show that 92-kDa type IV collagenase (matrix metalloproteinase) (MMP-9) mRNA was likewise progressively overexpressed during mouse skin carcinogenesis. To determine if overexpression of MMP-9 and TGF beta 1 are linked, we stably transfected a bioactive TGF beta 1 into a mouse skin squamous cell carcinoma cell line (CH72), which resulted in about twofold to three-fold higher levels of secreted active TGF beta 1. Active TGF beta 1-transfected cells grew only slightly, but not significantly, more slowly in vitro and in vivo than vector-only transfectants. Two clones overexpressing active TGF beta 1 secreted much reduced levels of MMP-9 activity, as determined by zymogram analyses. However, treatment of these clones with 40 pM exogenous TGF beta 1 for 48 h enhanced secretion of MMP-9 activity. Constitutive mRNA expression of MMP-9 was reduced twofold to 70-fold in five untreated active TGF beta 1-transfected clones relative to the other transfectants. In contrast, treatment with 40 pM exogenous TGF beta 1 induced MMP-9 mRNA expression in a time-dependent fashion, from twofold to fourfold after 4 h to a maximum of 12- to 19-fold after 24-48 h. Induction of MMP-9 mRNA was dose dependent at TGF beta 1 concentrations of 4-400 pM. Thus, stable transfection of bioactive TGF beta 1 downregulated whereas exogenous TGF beta 1 treatment upregulated MMP-9 activity and expression. Treatment of transfectants with a neutralizing TGF beta 1 antibody slightly downregulated constitutive MMP-9 mRNA (20-30%) but completely blocked induction by exogenous TGF beta 1. Thus, the effect of TGF beta 1 transfection was not due to secreted TGF beta 1 but may have been a secondary effect.

Carcinogenicity of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in the rat

A total of 10 highly-mutagenic heterocyclic amines have been identified to be carcinogenic in rodents. Among these, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), generally the most abundant with normal cooking procedures, induces mammary and colon carcinomas in rats in a clear dose-dependent manner. In a two-generation exposure (transplacental and trans-breast milk) experiment using Sprague-Dawley rats, an increased risk of mammary adenocarcinoma development was found in the second generation. Excretion of PhIP into the milk and transfer of PhIP to fetuses and neonates with resultant hepatic PhIP-DNA adduct formation were also confirmed. On the other hand, PhIP mammary carcinogenesis was significantly inhibited by coadministration of chlorophyllin or a synthetic antioxidant, 1-O-hexyl-2,3,5-trimethylhydroquinone, in long-term experiments using female F344 rats. The available findings strongly suggest that this food-derived carcinogen might be of importance as an environmental factor in the production of human cancers and that its carcinogenicity could be largely avoided by reducing intake of such compounds or by adoption of appropriate chemopreventive measures.

The role of xenobiotic metabolizing enzymes in arylamine toxicity and carcinogenesis: functional and localization studies

In both animal models and humans, the first and obligatory step in the activation of arylamines is N-hydroxylation. This pathway is primarily mediated by the phase-I enzymes CYP1A1, CYP1A2 and CYP4B1. In the presence of flavonoids such as alpha-naphthoflavone and flavone, both CYP3A4 and CYP3A5 have also been shown to play a minor role in the activation of food-derived heterocyclic amines. The further activation of N-hydroxyarylamines by phase-II metabolism can involve both N, O-acetylation and N, O-sulfonation catalyzed by N-acetyltransferases (NAT1 and NAT2) and sulfotransferases, respectively. Using an array of techniques, we have been unable to detect constitutive CYP1A expression in any segments of the human gastrointestinal tract. This is in contrast to the rabbit where CYP1A1 protein was readily detectable on immunoblots in microsomes prepared from the small intestine. In humans, CYP3A3/3A4 expression was detectable in the esophagus and all segments of the small intestine. Northern blot analysis of eleven human colons showed considerable heterogeneity in CYP3A mRNA between individuals, with the presence of two mRNA species in some subjects. Employing the technique of hybridization histochemistry (also known as in situ hybridization), CYP4B1 expression was observed in some human colons but not in the liver or the small intestine. Hybridization histochemistry studies have also demonstrated variable NAT1 and NAT2 expression in the human gastrointestinal tract. NAT1 and NAT2 mRNA expression was detected in the human liver, small intestine, colon, esophagus, bladder, ureter, stomach and lung. Using a general aryl sulfotransferase riboprobe (HAST1), we have demonstrated marked sulfotransferase expression in the human colon, small intestine, lung, stomach and liver. These studies demonstrate that considerable variability exists in the expression of enzymes involved in the activation of aromatic amines in human tissues. The significance of these results in relation to a role for heterocyclic amines in colon cancer is discussed.

Human exposure to mutagenic/carcinogenic heterocyclic amines and comutagenic beta-carbolines

Various kinds of mutagenic and carcinogenic heterocyclic amines (HCAs) are produced by heating protein-rich foods, such as meat and fish. To evaluate the risk of these HCAs in terms of human cancer development, exposure levels must be measured. We therefore analyzed their amounts in various kinds of cooked foods and in urine samples of healthy volunteers living in Tokyo. Based on the obtained quantitative data, daily exposure levels to 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) were calculated to be 0.3-3.9 and 0.005-0.3 microgram per person, respectively. Moreover, human DNA samples were analyzed with the 32P-postlabeling method, and colon, rectum and kidney tissues were found to contain an adduct spot corresponding to the standard 5'-pdG-C8-MeIQx by TLC and HPLC, at levels of 14, 18 and 1.8 per 10(10) nucleotides, respectively. The beta-carboline compound, norharman, is produced by heating L-tryptophan, and is known to be present in cooked foods and in cigarette smoke at higher levels than mutagenic and carcinogenic HCAs. While norharman is not itself mutagenic to Salmonella, it does become mutagenic to S. typhimurium TA98 with S9 mix in the presence of non-mutagenic aromatic amines like aniline and o-toluidine. When we examined whether DNA adducts are formed in the DNA of S. typhimurium TA98 by treatment with norharman and aromatic amines using 32P-postlabeling analysis, DNA adduct formation by norharman with aromatic amines was found to be related to the appearance of mutagenicity by norharman with aromatic amines.

Activation of phospholipase D by growth factors and oncogenes in murine fibroblasts follow alternative but cross-talking pathways

Phospholipase D (PLD) is activated by a variety of stimuli, including mitogenic stimulation by growth factors and oncogene transformation. Activation of PLD by growth factors requires protein kinase C (PKC) since depletion of the enzyme by down-regulation or direct inhibition by specific drugs completely abrogates this effect. Transformation by the ras and src oncogenes is also associated with an increase in basal PLD activity. However, this effect is not dependent on PKC, suggesting that growth factors and oncogenes may activate PLD by two independent mechanisms. Here we demonstrate that activation of PLD by phorbol esters is greatly enhanced in ras-transformed cells, suggesting synergistic activation of PLD by ras oncogenes and PKC. Also, ras-transformed cells showed a dramatic attenuation of the PLD activation induced by growth factors, although receptor function was still detectable. This attenuation paralleled the specific uncoupling of the phosphatidylinositol-specific phospholipase C (PI-PLC) pathway, indicating that activation of PLD by growth factors may be mediated by PI-PLC and PKC activation. Attenuation of PLD activation by platelet-derived growth factor was also observed in several oncogene-transformed cells, as well as the uncoupling of the PI-PLC pathway. Neither the co-operation with PKC activation nor the attenuation of the PLD response to growth factors in ras-transformed cells was a general consequence of cell transformation, since cells transformed by other oncogenes showed a normal response to either treatment. These results support the existence of at least two alternative signalling routes for the activation of PLD, one mediated by the PI-PLC/diacylglycerol/PKC pathway and a second one mediated by several oncogenes, independent of the PKC pathway, which synergizes with the PI-PLC/PKC-dependent pathway.

A childhood fibrolamellar hepatocellular carcinoma with increased aromatase activity and a near triploid karyotype

We report a 15-year-old boy with hepatocellular carcinoma (HCC) of the fibrolamellar type. He presented with advanced disease and a non-resectable tumor. Clinical features included marked gynecomastia which had been present for 3 years, failure to enter puberty, and failure to thrive. These features might have been due to a high aromatase activity of the tumor. The course of the illness suggested that the tumor had been present for at least 3 years prior to diagnosis. At diagnosis the patient had multiple metastases which included infiltrated ascites. Cytogenetic analysis of the ascites revealed a near triploid karyotype with cell-to-cell variation and an abnormality of chromosome 1 q. This to our knowledge is the first karyotype report of fibrolamellar HCC in a child.

Allelic loss on chromosomes 3p, 5q and 17p in renal cell carcinomas

Loss-of-heterozygosity (LOH) has been studied on 3p (von Hippel-Lindau gene locus), 5q and 17p (p53 gene locus) by a polymerase chain reaction (PCR)-based strategy in 42 sporadic renal cell carcinomas (RCC). LOH at seven microsatellite loci on 5q was investigated because a tumor suppressor gene on 5q involved in the development and/or progression of RCC has not yet been identified. LOH was found in seven (17%) RCC at single or multiple loci on 5q, 38% (11/29 informative cases) on 3p, and 6% (2/35 informative cases) on 17p. Replication error (RER) was present in 10% (4/42) RCC at single or multiple loci. The minimum region of deletion on 5q to account for LOH was mapped to 5q31.1 (interferon regulatory factor-1; IRF-1 locus), where LOH was detected in 23% (6/26 informative cases). LOH on 3p and 5q occurred in both stage 2 and more advanced (stage 3 and 4) tumors at similar incidences (41 and 33% on 3p; and 24 and 22% on 5q, respectively), suggesting that LOH on these chromosomes is an early genetic event. All RCC exhibiting LOH on 3p or 5q (IRF-1 locus) were the clear cell or the mixed clear and granular cell types. These findings suggest that LOH on 3p and 5q plays an important role in the genesis of clear cell RCC. In addition, only one tumor exhibited LOH on both 3p and 5q, which suggests that LOH occurs not sequentially but independently.

Heterocyclic aromatic amine content in pre-processed meat cuts produced in Canada

In an ongoing survey, the presence of heterocyclic aromatic amines (HAAs) was determined in processed, ready-to-eat meat products sold as 'meat cuts'. HAAs are a group of recently recognized mutagenic/carcinogenic contaminants in foods that are produced during the heat processing of meat. 16 samples of meat cuts (e.g. turkey breast, salami, chicken loaf, cooked ham, all beef meat, pepperoni, etc.), randomly purchased from supermarkets and specialty food stores in the Ottawa area, were analysed for the presence of eight HAAs. The isolation of HAAs was based on sequential liquid-liquid extraction procedures of the samples at both acidic and basic pH values. The mutagenic activity of these samples was determined using the Ames/Salmonella microsome assay with the strain TA98 plus rat liver S-9 metabolic activation. The mutagenicity of these samples ranged from undetectable to slightly active. The highest mutagenic activity, 141 induced revertants/g, was found in a smoked turkey breast sample. 11 samples were not mutagenic, including two that indicated a tendency for inhibition of the spontaneous revertants. The remaining four samples exhibited very low mutagenic activity. For chemical analysis, the extracts were purified with two solid phase extraction cartridges. Quantitative analysis was performed by using liquid chromatography for separation and mass spectrometry for detection. With the exception of trace amounts (0.4 ng/g) of 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) in the sample with highest mutagenic activity, the chemical analysis did not detect the presence of any of the eight most frequently found HAAs in fried or broiled meat products. These data suggest that consumption of meat cuts does not present a serious health risk from HAA-type contaminants.

Fatores de risco ambientais para o câncer gástrico: a visão do toxicologista

A carcinogênese é um processo altamente complexo do qual participam fatores de risco herdados e fatores de risco ambientais, tais como a alimentação, o hábito de fumar, a ocupação, e a exposição a radiação e a agentes químicos. A toxicologia experimental identifica as substâncias químicas potencialmente carcinogênicas e torna possível medidas regulatórias que objetivam reduzir a exposição humana a elas. A carcinogênese pode ser vista como consistindo de três seqüências distintas: a iniciação, a promoção e a progressão. A conversão neoplásica (iniciação) ocorre quando um evento genético (mutações, rearranjos cromossômicos, inserções ou deleções de genes e amplificação de genes) resulta em ativação de oncogenes e/ou em falta de expressão - ou inativação de produtos - de genes supressores de tumores. A promoção envolve a expansão clonal das células "iniciadas" e exige a proliferação celular. Estratégias efetivas para reduzir os riscos de câncer gástrico e os riscos de neoplasias de outras localizações devem incluir o controle de carcinógenos conhecidos, assim como a quimioprevenção, por meio de intervenções racionais no processo carcinogênico. Neste sentido, o desafio a ser enfrentado pelo toxicologista envolve o desenvolvimento de ensaios preditivos melhores e mais baratos e a elucidação dos mecanismos subjacentes à carcinogênese química.

The natural somatic mutation frequency and human carcinogenesis

Much recent attention has been paid to the important role of the DNA mismatch repair system in controlling the accumulation of somatic mutations in human tissues and the association of mismatch repair deficiency with carcinogenesis. In the absence of an intact mismatch repair system, cells accumulate mutations at a rate some 1000 times faster than normal cells, and this mutator phenotype is easily measured by the detection of the formation of new variant alleles at microsatellite loci. However, the mismatch repair system is not 100% efficient, even when intact, and the pattern of microsatellite alterations in a wide variety of tumors is consistent with these being due to clonal amplification from tissues that are genetically heterogeneous at microsatellite loci rather than mismatch repair deficiency in the tumor itself. On this basis, it can be estimated that the mutation frequency of microsatellites in normal human tissues is approximately 10(-2) per locus per cell. Similarly, a frequency of mutation at minisatellite loci in normal tissues of around 10(-1) per locus per cell can be estimated. Such elevated levels of mutation are consistent with a recent study of the frequency of HPRT mutation in human kidneys that demonstrated these to be frequent (average 2.5 x 10(-4) in individuals of 70 years or more) and exponentially related to age. Taken as a whole, the data suggest that somatic mutation in human epithelial cells may be some 10-fold higher than in peripheral blood lymphocytes and that the underlying rate of spontaneous mutation is sufficient to account for a large proportion of human carcinogenesis without the need to evoke either stepwise alteration to a mutator phenotype of clonal expansion at all the mutation steps in carcinogenesis. The exponential increase in mutation frequency with age is predictable on the basis that the mutation rate is controlled at the level of repair and that mutation in genes that affect the efficiency of these processes will gradually increase the underlying rate. In addition, the age relatedness of mutation frequency strongly supports the concept that mutation is cell division dependent and that cellular proliferation per se is an important risk factor for cancer. Comparison of somatic mutations with those in the human germline mutation suggests common mechanistic origins and that the high levels of somatic mutation that occur are a direct reflection of the germline mutation rate selected over evolutionary time. Thus, the somatic accumulation of mutations can be seen as a natural process within the human body and cancer a normal part of the human life cycle. This point of view may explain why it has been so difficult to significantly reduce cancer incidence and suggests that, for this to be achieved, the means of altering the natural somatic mutation rate needs to be identified.

Multiple synchronous primaries of the gastrointestinal tract: a molecular case report

Six synchronous gastrointestinal primaries were identified in a 70 year old male with no known cancer predisposition syndrome or recognized risk factors except alcohol abuse. These specimens appeared to be independent and unrelated by gross and histopathological examination. In order to further evaluate the six tumors, we analyzed selected DNA sequences for alterations in the K-ras oncogene and p53 tumor suppressor gene. In addition, three loci were analyzed to determine microsatellite instability. Using the polymerase chain reaction, single stranded conformational polymorphism, and DNA sequencing, we demonstrated that each primary manifests genetic characteristics typical of the tissue of origin. In addition, one primary, a moderately differentiated colon adenocarcinoma, exhibited mutations not detected in the other specimens. This study suggests that these synchronous primaries arose independently and progressed along different carcinogenic pathways.

Longitudinal distribution of arylamine N-acetyltransferases in the intestine of the hamster, mouse, and rat. Evidence for multiplicity of N-acetyltransferases in the intestine

Experimental and clinical evidence indicates that AcCoA:arylamine N-acetyltransferases (NATs; EC 2.3.1.5) are involved in the bioactivation and inactivation of a wide variety of arylamine, hydrazine, and carcinogenic arylamine xenobiotics. Longitudinal distribution of NATs in the intestine of the hamster, mouse, and two strains of rat was examined utilizing the model arylamine substrates procainamide(PA) and p-aminobenzoic acid (PABA) for the monomorphic (NAT1) and polymorphic (NAT2) enzymes in the rodent. NAT1 and NAT2 were distributed quite differently in each species examined. In particular, rat intestinal NATs were distributed equally throughout the intestinal tract. In contrast, hamster intestinal NATs decreased in activity from the proximal small intestine to the distal large intestine. Mouse NAT2 activity was highest in the cecum, whereas NAT1 was highest in the proximal small intestine. Although these model substrates have been shown to be selective for NATs, they are not specific. Therefore, a series of biochemical studies were undertaken to evaluate NAT multiplicity in the intestine of the F-344 rat. To assess multiplicity of NAT expression, selective inhibition, differential sensitivity to heat inactivation, and kinetic analysis were performed on intestinal cytosol. Eadie-Hofstee transformation of PA N-acetylation yielded a curvilinear plot indicative that a low affinity-high capacity enzyme aside from NAT1 (presumably NAT2) was contributing to PA N-acetylation activity. PA activity was found to exhibit approximately 4- to 5-fold greater thermostability than PABA activity. Furthermore, PA acetylation could be inhibited selectively with vinyl fluorenyl ketone (2.5 to 5 microM) but not with methotrexate (up to 2 mM). Taken together, these studies suggest the expression of both NAT1 and NAT2 in the intestine of the F-344 rat.

Analysis of synergism in hepatocarcinogenesis based on preneoplastic foci induction by 10 heterocyclic amines in the rat

The effects of simultaneous treatment with 5 or 10 heterocyclic amines at low dose levels on hepatocarcinogenesis in rats were investigated using a medium-term liver bioassay protocol based on the two-stage carcinogenesis hypothesis with diethylnitrosamine initiation (200 mg/kg, i.p.). Five carcinogenic heterocyclic amines in experiment 1 (Trp-P-1, Glu-P-2, IQ, MeIQ, MeIQx) and experiment 2 (Trp-P-2, Glu-P-1, MeAalphaC, AalphaC, PhIP) were administered together or individually in the diet at levels of 1/1, 1/5, or 1/25 carcinogenic doses, and all 10 chemicals were given at 1/10 or 1/100 levels in experiment 3. Induction of preneoplastic glutathione S-transferase placental form (GST-P)-positive foci in the liver was generally increased in the combination groups over the sums of the 5 or 10 individual effects. Thus, based on the heteroadditive concept, synergism was observed for each combination, being most obvious in the group given all 10 chemicals at the 1/10 dose levels. However, the values for the combined groups were generally close to the averages of the 5 or 10 data gained for the heterocyclic amines alone at the corresponding higher doses, indicating the possibility of isoadditivity. Based on these findings, we propose here a new statistical method for analysis of combined effects of multiple chemicals, and, using this, we demonstrated (true) synergism with some heterocyclic amine combinations. The importance of dose-response curves for evaluation of combination effects is discussed.

The mutagenic properties of DNA minor-groove binding ligands

This review summarises mutagenesis-related research on the major classes of DNA minor groove binding ligands. These compounds can bind to DNA covalently or non-covalently, and span a range of DNA sequence selectivities. Many of the non-covalent binders show effects on topoisomerase enzymes in mammalian cells, with the bisbenzimidazoles being the most active. Mutagenic effects consistent with topoisomerase inhibition are observed in vitro. Many of these compounds induce aneuploidy and polyploidy, properties which may also contribute to carcinogenic processes. Similarly, uvrA trapping by some minor groove binders may alter mutagenetic processes by inhibiting efficient repair. Distamycin has been shown to enhance the mutagenicity of ethidium bromide in bacteria by an undetermined mechanism. However, the inhibitory effects of minor groove binders on human DNA repair systems have not yet been reported. Hoechst 33258 and distamycin cause chromosome decondensation in both mouse and human cells particularly at heterochromatic regions which are rich in AT content. Various minor groove binders have been shown to induce fragile sites in cultured lymphocytes from susceptible individuals, which may have a propensity to develop particular cancers. Investigation of the relationship between fragile site inducing drugs and chromosomal rearrangements in fragile site carriers has not been investigated but may yield interesting results. Some DNA alkylating minor groove binders can generate lesions extremely toxic to mammalian cells (e.g., CC-1065 and analogues), and induce a range of DNA sequence changes in vivo, both at the site of covalent bonding as well as at surrounding sequences. This may be typical of alkylating minor groove binders which have a binding site size of several base pairs, and which stabilise helical structure. Minor groove binders have effects on gene expression in vitro by inhibiting the sequence selective binding of various transcription factors to DNA. These effects may result in expression or repression of downstream genes also. This class of ligand thus offers the possibility of mutations targeted to specific genes or genomic regions. It will be interesting to determine whether such examples of targeted mutagenesis, as has already been observed with CC-1065 and adozelesin, will result in an enhanced or in a lowered capacity to promote neoplastic disease. However it should be noted that pentamidine, a minor groove binder used in the treatment of AIDS-related PCP, has thus far shown no mutagenic effects in nuclear DNA and only a weak effect in mitochondrial DNA of yeast. These results suggest that minor groove binding does not necessarily lead to mutagenesis.

Genomic instability in 1p and human malignancies

Both cytogenetic and molecular genetic approaches have unveiled non-random genomic alterations in 1p associated with a number of human malignancies. These have been interpreted to suggest the existence of cancer-related genes in 1p. Earlier studies had employed chromosome analysis or used molecular probes mapped by in situ hybridization. Further, studies of the various tumor types often involved different molecular probes that had been mapped by different technical approaches, like linkage analysis, radioactive or fluorescence in situ hybridization, or by employing a panel of mouse x human radiation reduced somatic cell hybrids. The lack of maps fully integrating all loci has complicated the generation of a comparative and coherent picture of 1p damage in human malignancies even among different studies on the same tumor type. Only recently has the availability of genetically mapped, highly polymorphic loci at (CA)n repeats with sufficient linear density made it possible to scan genomic regions in different types of tumors readily by polymerase chain reaction (PCR) with a standard set of molecular probes. This paper aims at presenting an up-to-date picture of the association of 1p alterations with different human cancers and compiles the corresponding literature. From this it will emerge that the pattern of alterations in individual tumor types can be complex and that a stringent molecular and functional definition of the role that Ip alterations might have in tumorigenesis will require a more detailed analysis of the genomic regions involved.

DNA adduct formation by hormonal steroids in vitro

We examined the binding of various steroid hormones to DNA in vitro by means of 32P-postlabeling. Seventeen steroid hormones and cholesterol (CS) were incubated with human liver DNA at 37 degrees C for 1 h under aerobic conditions in the absence of catalysis. The reaction mixtures were analyzed by the nuclease P-1 version of 32P-postlabeling. The results showed that cortexolone (CX), prednisolone (PS), cortisone (CN), cortisol (CL), tetrahydrocortisol (TC), corticosterone (CC), 11-deoxycorticosterone (DC), dexamethasone (DX), dihydrocortisol (DL), and aldosterone (AL) covalently bound with DNA. However, progesterone (PG), 17 alpha-hydroxyprogesterone (HG), estrone (E1), estradiol (E2), estriol (E3), testosterone (TS), cortol (CR) and the original compound for biosynthesis, CS, did not form adducts. In absence of DNA, the steroids themselves did not give rise to any spot on TLC under the same conditions. The dose-responses of DNA binding by DC, DL, CC, CL and CN were linear. The relative adduct labeling of reactive steroids at a concentration of 2 mM were as follows: 68.8 (CX), 53.2 (PS), 39.6 (CN), 29.9 (CL), 20.9 (TC), 12.9 (CC), 12.3 (DC), 7.5 (DX), 4.7 (DL), 1.2 (AL) adducts per 10(8) nucleotides. Reactive and nonreactive steroids were distinguishable by the presence or absence of the carbonyl group (-CO-CH2OH) at carbon seventeen (C17) of the cholesterol skeleton. This implies that the electrophilic carbonyl or a neighboring group perhaps involved in the formation of covalent bond with DNA. To investigate the nature of target base(s) of these DNA reactive steroids, mononucleotides of all four bases of DNA were reacted with CN, CL, CC and cochromatographed with the obtained spots of DNA reactions. The results of which stated that these steroids and guanine reaction gave the same spots as observed in DNA reaction, indicating guanine is the main target of these DNA reactive steroids. Hep G2 human hepatocellular carcinoma cells were used as an alternative model. Although nine steroids (CL, DL, TC, PS, DX, PG, E2, TX, CR) did not react with intracellular DNA under our experimental conditions, our findings suggested that some hormonal steroids can form covalent DNA adducts in vivo.

Frequency of Ki-ras mutations and DNA alkylation in colorectal tissue from individuals living in Manchester

Most human colorectal cancers arise through the accumulation of a series of genetic alterations such as point mutations within the Ki-ras and p53 genes, but the chemical carcinogens that may be implicated in these events are still unidentified. In a previous study, we showed that DNA from human colorectal tissue contained O6-methyldeoxyguanosine (O6-MedG), a promutagenic lesion arising from exposure to as yet unidentified methylating agents. To address whether such exposure may result in oncogene activation in human colorectal tumors, we examined another series of paired normal and tumor DNA samples from the lower intestinal tract for the presence of O6-MedG in DNA (as a marker of exposure) and for mutations within the Ki-ras gene. After isolation by high pressure liquid chromatography, O6-MedG was quantified by a radioimmunoassay with a limit of detection of 0.01 mumol O6-MedG/mol dG. The frequencies of methylation were 33%, 52%, and 48% for normal DNA and 58%, 32%, and 63% for tumor DNA isolated from the cecum, sigmoid colon, and rectum, respectively. Overall, 35% of the individuals had no detectable O6-MedG in the DNA from both their tumor and normal tissue. Ki-ras mutations were initially identified by a restriction site mutation assay and then sequenced to ascertain the mutations thus detected. The frequencies of mutations in tumor DNA isolated from the cecum, sigmoid colon, and rectum were 28%, 29%, and 42%, respectively. DNA isolated from macroscopically normal tissue was found to contain Ki-ras mutations in 14% of sigmoid colon samples and 12% of rectal samples. Most base mutations were in codon 12 (72%), and 64% were GC-->AT transitions: 28% and 8% were GC-->TA and CG-->CG transversions, respectively. All mutations were at the second base of either codon 12 or codon 13 except for a single GC-->TA transversion at the first base of codon 13 in a rectal tumor sample. There was no association between the presence of O6-MedG in DNA from either normal or tumor tissue or both normal and tumor tissue and the incidence of Ki-ras mutations or GC-->AT transitions in mutated Ki-ras genes. It remains to be determined, however, whether there is a relationship between methylating-agent exposure and Ki-ras mutations, as (i) the presence of O6-MedG in colorectal DNA in these samples may not represent the exposure when Ki-ras mutational activation was occurring (i.e., at some unknown time in the past), (ii) interindividual differences in repair-enzyme activity may alter susceptibility to a mutational event after exposure, (iii) the predominant mutagen in the colon and rectum may not be a methylating agent (e.g., nitric oxide), and (iv) exposure to methylating agents need not result in oncogene activation in human tissues but may perhaps promote the emergence of the mutator phenotype.

Inhibitory potential of pregnancy and lactation on mammary carcinogenesis induced by a food-derived carcinogen, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine, in Sprague-Dawley rats

Effects of pregnancy and lactation on a food-derived carcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP)-induced mammary carcinogenesis were investigated in female Sprague-Dawley rats. When rats were administrated PhIP in the diet (100 ppm) for 15 weeks, palpable subcutaneous tumors were first detected at week 24 in three nulliparous rats, while tumor latency was apparently increased in animals undergoing delivery and weaning during the PhIP administration period. However, since the incidence of palpable tumors was not consistently increased in the nulliparous animals, possibly due to the short PhIP-treatment, and the fact that spontaneous tumors developed in parous animals after week 44, the final incidences and multiplicities of histopathologically confirmed adenocarcinomas at week 48 were not significantly different between the two groups. These findings art partly supported by the epidemiological evidence that breast cancer is more prevalent in nulliparous than in parous women.

Triple primary malignant neoplasms including a malignant brain tumor: report of two cases and review of the literature

BACKGROUND

Two rare cases of triple primary malignant neoplasms (PMN), including malignant brain tumors, which were glioblastoma multiformes, are described.

METHODS

The clinical characteristics and underlying genetic alterations in triple or more PMN, including malignant brain tumors are discussed with intensive review of the literature.

RESULTS

The first patient, a 77-year-old male, suffered metachronously from tubular adenocarcinoma of the stomach, transitional cell carcinoma of the bladder, and glioblastoma in the brain. This glioblastoma had loss of heterozygosity in exons 7-8 in p53 gene. The second patient, a 68-year-old male, developed papillary adenocarcinoma of the lung, adenocarcinoma of the rectum, and glioblastoma in the brain during a period of 7 years. In 42 such cases described in the literature, age distribution demonstrated two characteristic peaks, one in the third decade and the other over 50 years of age. The younger group consisted mainly of Turcot's syndrome, and of a case of Li-Fraumeni familial cancer syndrome. On the other hand, neither of these hereditary cancer syndromes were contained in the elder group. Regarding the site of PMN, colorectal cancers were associated most frequently with malignant brain tumors, followed by stomach cancers, and thyroid cancers. Malignant brain tumors, mostly glioblastoma multiforme, tend to occur as the last tumor of triple or more PMN.

CONCLUSIONS

These results suggest that genetic background might play an important role in tumorigenesis of PMN in the younger group, whereas epigenetic factors would be more important in the older group. Characteristic organ association and factors influencing carcinogenesis, such as aging, environmental carcinogens, and underlying genetic alterations in these tumors are further discussed.

Extrapolation of heterocyclic amine carcinogenesis data from rodents and nonhuman primates to humans

Twenty different heterocyclic amines have been isolated and identified from cooked foods especially beef, fish, pork and fowl. Other HCAs have also been isolated but their structure remains to be elucidated and new HCAs are likely to be identified in the future. The HCAs are highly mutagenic and all ten HCAs that have been tested for carcinogenic activity, produce tumors in mice and rats. For humans the average daily intake of HCAs is in quantities of 10-20 mg/person/day. The HCAs are procarcinogens and are activated by the cytochrome P450 system especially CYP 1A2. Rodents, monkeys and humans have the capacity to activate HCAs. Studies using hepatic microsomes demonstrated that humans have a greater capacity to activate the majority of HCAs tested than rodents or cynomolgus monkeys. Three HCAs are currently under evaluation in nonhuman primates for carcinogenic activity and one, IQ, is highly carcinogenic inducing primary hepatocellular carcinomas in the majority of cynomolgus monkeys treated. Epidemiological studies, although not definitive, are supportive of an association of HCAs intake to the etiology of human cancer. Risk assessments from animal data show a risk of HCAs to humans in the range of 10(-3) to 10(-4) which is an order of magnitude greater than compounds currently regulated by the U.S. Food and Drug Administration or the Environmental Protection Agency. Taken together evidence from mutagenicity data, activation by various species including humans, carcinogenicity in animals, human consumption data, epidemiological studies and risk assessment, supports the conclusion that HCAs are probable human carcinogens.

Some characteristics of neoplastic cell transformation in transgenic mice

The role of the expression of different cellular genes and viral oncogenes in malignant cell transformation is discussed. We pay special attention to the role of the genes for growth factors and their receptors and homeobox genes in oncogenesis. Based on both the literature and our own data, specific features of tumors developed in transgenic mice are discussed. All of these data are used to analyze current theories of multistep oncogenesis and the stochastic component in this process. We suggest that all known evidence about the mechanisms of oncogenesis be used in studying the problem at various structural and functional levels in an organism. The chapter shows that transgenic mice are a most suitable model for studying various aspects of malignant transformation from the molecular to the organismal and populational levels.

Evidence for the presence of two tumour-suppressor genes for hepatocellular carcinoma on chromosome 13q

The concept that genetic changes accumulate during development and progression of cancer is widely accepted. Frequent allelic losses at chromosome 13q have been found in hepatocellular carcinomas (HCCs), and a known tumour-suppressor at 13q14, the retinoblastoma (RB) gene, is thought to be the target of those events. However, no strong evidence has emerged to support a significant role of RB during hepatocarcinogenesis. To investigate the minimal area(s) of loss on chromosome 13q in HCCs, we analysed DNAs isolated from 92 tumours for loss of heterozygosity (LOH) at 13 loci on chromosome 13q, using polymorphic microsatellite markers. In 30 (32.6%) of 92 cases we detected LOH for at least one locus on chromosome 13q and 20 revealed a partial or interstitial deletion of chromosome 13q. Deletion mapping of these 20 tumours indicated two separate commonly deleted regions: one was located in the region including RB and the other was located in the region including the BRCA2 locus. These findings suggest that at least one putative tumour-suppressor gene for HCC other than RB, possibly BRCA2, exists on chromosome 13q.

Accumulation of genetic changes during development and progression of hepatocellular carcinoma: loss of heterozygosity of chromosome arm 1p occurs at an early stage of hepatocarcinogenesis

To investigate cumulative genetic changes during development and progression of hepatocellular carcinoma (HCC), we examined DNAs isolated from 104 tumors for loss of heterozygosity (LOH) at 13 loci on six chromosomal arms and for an increase of copy number ("multiplication") of alleles on 8q, using polymorphic microsatellite markers. A comparison of genetic features with clinicopathological stages of these tumors revealed that LOH on 1p had occurred in tumors at an early stage or with a well-differentiated histological phenotype (8/26; 31%) as well as in tumors at more advanced stages. Genetic alterations on chromosome arms 4q, 8p, 8q, 13q, 16q, and 17p were more often observed in tumors of more advanced stages and poorer differentiation grades. When size was the criterion for comparison, LOH on 1p was observed frequently even in tumors smaller than 2 cm (6/16; 38%), whereas allelic losses on 16q were detected frequently only in larger tumors. These results suggest that the putative tumor suppressor gene(s) assumed to be located on 1p may be involved in an early step of carcinogenesis in liver tissue and that the other genetic alterations examined here may play important roles in progression of HCC.

How much of the human disease burden is attributable to environmental chemicals?

According to the evaluations made by IARC, 66 agents or exposures have been recognised as human carcinogens. About 60% of all cancer cases occur in people over 65 years of age, which is indeed the most important risk factor for cancer, including both the duration of exposure to the variety of carcinogenic agents and allowing expression of genetically determined disorders. Diet as a source of mutagens as well as other environmental mutagens may affect blood pressure either directly or by favouring the development of arteriosclerosis: mutagens have been shown to accelerate arteriosclerotic plaque development. Some concern has also been expressed on the possible adverse effect that environmental chemicals may have on reproductive ability, on the basis of the decline in semen quality reported over the past decades. Better defining the role of aetiological agents would result in a more precise definition of attributable risks. Of particular interest to the latter goal are the studies of inter-individual variability in the susceptibility to carcinogens, which will hopefully contribute to define the role of low-level exposure to carcinogens.

Chemical properties of the ultimate metabolites of 2-amino-5-phenylpyridine (PHE-P-1) and its ortho-methyl derivative

The reactivity of the N-acetoxy metabolite of 2-amino-5-phenylpyridine (Phe-P-1), a pyrolysis product of phenylalanine, towards 2'-deoxyguanosine (dG), 2'-deoxyguanosine-3'-monophosphate (dGMP) and DNA was studied and compared with that of the ortho-methyl derivative. Reaction of 2-acetoxyamino-5-phenylpyridine (N-OAc-APP) with dG resulted in substitution at the 8-position of this nucleoside by the ortho carbon of the amine. The major reaction, however, was acetylation of dG. In contrast, 2-acetoxyamino-3-methyl-5-phenyl-pyridine (N-OAc-MeAPP) mainly attacked the 8-position of dG by the exocyclic nitrogen and hardly any acetylation of the nucleoside occurred. The adducts were chromatographically isolated and characterized by their mass and NMR spectra. Upon reaction of N-acetoxy compounds with DNA and dGMP, formation of the same adducts was observed, besides the formation of minor amounts of unidentified compounds, as was established by 32P-postlabelling analysis. The amount of DNA-bound amine, formed by the interaction of N-OAc-APP with DNA, was approximately 15 times smaller than that observed after the reaction with the corresponding ortho-methyl derivative under the same conditions.

Analysis of the potential carcinogenicity of coffee and its related compounds in a medium-term liver bioassay of rats

The potential carcinogenicity of coffee and related compounds was examined using a medium-term liver bioassay based on the induction of glutathione S-transferase placental form (GST-P)-positive foci in F344 rats. A total of 230 males were initially injected with diethylnitrosamine (200 mg/kg body weight, ip) or saline as controls and 2 wk later were fed on diet or drinking water supplemented as follows for 6 wk: 5% regular instant coffee; 5% decaffeinated instant coffee; freshly brewed coffee, 8 g in 140 ml water; 0.1% caffeine, 0.2% methylglyoxal, 0.2% glyoxal; or 0.3% theophylline in the drinking water (w/v); and 0.4% theobromine in the diet (w/w). All rats were subjected to two-thirds partial hepatectomy at wk 3 and killed at wk 8. The resultant values for GST-P-positive hepatic focus induction were slightly increased with methylglyoxal and decreased with glyoxal and theobromine compared with the corresponding controls. Although the increase in number of foci for methylglyoxal was statistically significant at P < 0.05, the value was within the historical control levels. Regular and decaffeinated instant coffee as well as fresh-brewed coffee, caffeine and theophylline exerted no effects on focus development. Thus, the coffee-related compounds examined demonstrated no obvious enhancing potential, and it is therefore concluded that coffee and its main constituents are not carcinogenic for the rat liver.

Changes of antigen expression on human hepatoma cell lines caused by sodium butyrate, a differentiation inducer

We have previously reported the effect of a differentiation inducer, sodium butyrate (SB), on human hepatocellular carcinoma (HCC) cell lines, demonstrating that it was a potent inducer of differentiation. In the present study, we investigated the alteration in expression of an antigen defined by a murine monoclonal antibody, H2, as well as alterations in the expression of other antigens, on the HCC cell lines HCC-T, HCC-M, and PLC/PRF/5, since it is known that specific antigenic changes occur during the differentiation of leukemic cells. The expression of the antigen defined by H2 increased immunocytochemically on HCC-T, HCC-M, and PLC/PRF/5 during treatment with SB. A flowcytometric study showed that almost all the HCC-T and HCC-M cells treated with SB highly expressed this antigen after 5 days' treatment. The antigen expression detected by H2 decreased after the removal of SB from the medium. On the other hand, antigen expression detected by another monoclonal antibody, 5C11, was not changed by this treatment. The expression of intracellular adhesion molecule (ICAM)-1 in HCC-T increased slightly, but that of beta 2-microglobulin and HLA-DR did not change. These results demonstrated that some antigen expression was changed by SB treatment and that the antigen defined by H2 seemed to be highly expressed on human HCC cells in the differentiated state.