Genotoxicity evaluation of sesamin and episesamin.

Sesamin is a major lignan that is present in sesame seeds and oil. Sesamin is partially converted to its stereoisomer, episesamin, during the refining process of non-roasted sesame seed oil. We evaluated the genotoxicity of these substances through the following tests: a bacterial reverse mutation assay (Ames test), a chromosomal aberration test in cultured Chinese hamster lung cells (CHL/IU), a bone marrow micronucleus (MN) test in Crlj:CD1 (ICR) mice, and a comet assay using the liver of Sprague-Dawley (SD) rats. Episesamin showed negative results in the Ames test with and without S9 mix, in the in vitro chromosomal aberration test with and without S9 mix, and in the in vivo comet assay. Sesamin showed negative results in the Ames test with and without S9 mix. In the in vitro chromosomal aberration test, sesamin did not induce chromosomal aberrations in the absence of S9 mix, but induced structural abnormalities at cytotoxic concentrations in the presence of S9 mix. Oral administration of sesamin at doses up to 2.0g/kg did not cause a significant increase in either the percentage of micronucleated polychromatic erythrocytes in the in vivo bone marrow MN test or in the % DNA in the comet tails in the in vivo comet assay of liver cells. These findings indicate that sesamin does not damage DNA in vivo and that sesamin and episesamin have no genotoxic activity.

Cancer in Japan: Prevalence, prevention and the role of heterocyclic amines in human carcinogenesis.

In this article, three topics are being studied in order to understand the present state of cancer in Japan. First, the statistics on cancer mortality indicates that the mortality from cancer in young individuals has been decreasing during the last 50 years, although the total mortality from cancer has been steadily and steeply increasing. Second, epidemiological analyses of cancer causes in Japan indicated that 50 % of cancer cases are preventable, and that prevention of infection and refraining from tobacco smoking will reduce cancer mortality by about 40 %. Third, mutagenic/carcinogenic heterocyclic amines present in cooked meat/fish have been suggested to be carcinogenic in humans in many epidemiological studies carried out in Japan and other countries.

Modification of the 32P-postlabeling method to detect a single adduct species as a single spot.

The original 32P-postlabeling method developed by Randerath and his colleagues has been modified to detect a single type of adduct as a single spot in thin-layer chromatography (TLC), because some types of adducts gave multiple adduct spots by the original method. In the remodified methods, DNA is first digested with micrococcal nuclease and phophodiesterase II and then labeled with [gamma-32P]ATP under standard or adduct-intensification conditions. Since the labeled digest includes adducted mono-, di-, and/or oligo-deoxynucleotides, it is further treated with phosphatase and phosphodiesterase prior to TLC. The labeled digest is treated with nuclease P1 (NP1) in method I, and with T4 polynucleotide kinase and NP1 in method II, and then with phosphodiesterase I in both cases, and subjected to TLC. The advantage of these methods is that the number of adduct species formed can be estimated by TLC.

Beta-catenin mutations in liver tumors induced by 2-amino-3,4-dimethylimidazo[4,5-f]quinoline in CDF1 mice.

Heterocyclic amines are potent mutagens and carcinogens formed in cooked protein rich foods. In this study, we screened liver tumors induced by 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ) in CDF1 mice for beta-catenin and APC mutations and other genetic alterations shown to occur in human hepatocellular carcinomas (HCC), including mutations in the p53 and H-ras genes, c-myc amplification and E-cadherin promoter methylation. SSCP followed by direct DNA sequencing revealed mutations in exon 2 of the beta-catenin gene in 2 of 16 liver tumors (12.5%). Promoter methylation of the E-cadherin gene was detected in one liver tumor induced by MeIQ. There were no mutations in the mutation cluster region of the APC gene, in exons 5-8 of the p53 gene, or in codons 12, 13 and 61 of the H-ras gene, nor c-myc amplification in any of liver tumors induced by MeIQ. These data indicate that except for the occasional disruption of the Wnt pathway through beta-catenin mutations, the genetic pathways involved in the development of HCC differ significantly between human liver cancer and tumors induced in mice by MeIQ, but do not rule out the possibility that heterocyclic amines constitute a carcinogenic risk factor in humans.

Studies on mammary carcinogenesis induced by a heterocyclic amine, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine, in mice and rats.

Ten heterocyclic amines (HCAs) that are produced by heating amino acids, proteins, or proteinaceous food such as fish and meat were examined for carcinogenicity in rats and mice. Three of them, 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ), and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), have been shown to induce mammary cancer in female F344 and/or SD rats, but none of the HCAs induced mammary cancer in CDF(1) mice. This report reviews our recent studies on mammary carcinogenesis of PhIP in various strains of mice and on the roles of genomic instability in the rat mammary carcinogenesis of PhIP. We demonstrated that the survival time from mammary adenocarcinomas was shorter in PhIP-treated BALB/c mice than that in the untreated control, and with a significantly higher incidence in the C.B-17 strain of mice compared with that of the control. To clarify mechanisms of mammary carcinogenesis, we examined genomic instability in rat mammary cancer induced by PhIP. Mammary cancers were induced in F344 x SD F(1) rats harboring the lacI transgene, and two cell lines were established from two adenocarcinomas. They showed a greater than 10-fold higher frequency of spontaneous mutations than that of the primary culture of normal mammary epithelial cells, in the lacI transgene and the hprt endogenous gene during cell replication. Nucleotide sequencing revealed that almost all types of mutations were increased, with a remarkable increase of A:T --> C:G mutation. This genomic instability was not attributed either to alterations of mismatch-repair enzymes or to p53. These mutational characteristics were also observed in the original tumors. Single-nucleotide instability (SNI) might be implicated in the mammary cancer induced by PhIP.

Transient forebrain ischemia induces expression of serine/threonine protein phosphatase 1 mRNA in the vulnerable regions of gerbil brain.

Apoptosis is thought to be implicated in delayed neuronal cell death following transient forebrain ischemia. Recently, apoptosis in neurons induced by an inhibitor of serine/threonine (ser/thr) protein phosphatases (PPs) has been reported. In this study, we investigated the effect of transient forebrain ischemia on the expression of ser/thr PPs in the brain of Mongolian gerbils. At 24 h after 5-min bilateral carotid artery occlusion, Northern blotting analysis revealed the increase of PP1 mRNA expression in the vulnerable CA1 region of the hippocampus and striatum, but not in the cortex and CA3 region. In contrast, the protein level of PP1 detected by Western blotting analysis decreased in all regions. We conclude that the inhibition in PPs expression in the vulnerable regions may affect cell death after transient forebrain ischemia.

A rat colon cancer model induced by 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine, PhIP.

2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) is one of the most abundant heterocyclic amines contained in cooked meat and fish, and induces aberrant crypt foci (ACF), putative preneoplastic lesions of the colon, and colon cancers in male rats when administered orally. As has been reported previously, F344 rats are susceptible to induction of ACF by PhIP, while ACI rats being relatively resistant. Approximately one-fourth of ACF induced by PhIP in F344 rats are dysplastic; exhibiting lesions with structural distortion of the crypt, decrease of goblet cells, nuclear stratification and enlargement of nuclei. Dysplastic ACF demonstrate beta-catenin accumulation, mainly in the cytoplasm, and increased cell proliferation in crypts. These dysplastic ACF are, therefore, strongly considered to be putative preneoplastic lesions of the colon.A genetic trait affecting the susceptibility to colon carcinogenesis in F344 rats was mapped to chromosome 16, between D16Rat17 and D16Wox3, using the number of ACF as a surrogate biomarker for colon carcinogenesis. Since the number of dysplastic lesions is well correlated with the total number of ACF, being approximately one-fourth of the total ACF as described above in F344 rats and will be described elsewhere in ACI rats, the gene involved in the susceptibility to ACF induction may possibly be partly responsible for the susceptibility to colon carcinogenesis by PhIP. We, thus, tentatively referred the name of the candidate susceptibility gene on rat chromosome 16 as susceptibility to colon tumor (Sct). In the present study, the colonic lesions induced by PhIP were well refined histologically and genetically, and the multi-step profiles of colon cancer development by PhIP were well characterized and revealed to be similar to the multi-step model of colon carcinogenesis in humans. The PhIP-induced colon cancer model in rats, thus contributes as a relevant tool to elucidate genetic factors responsible for susceptibility to colon carcinogenesis in human. Other unknown genetic or epigenetic alterations, which are essential for the development of early lesions of colon carcinogenesis, could also be clarified using this system.

Modification of the ³²P-postlabeling method to detect a single adduct species as a single spot.

The original (32)P-postlabeling method developed by Randerath and his colleagues has been modified to detect a single type of adduct as a single spot in thin-layer chromatography (TLC), because some types of adducts gave multiple adduct spots by the original method. In the remodified methods, DNA is first digested with micrococcal nuclease and phosphodiesterase II and then labeled with [γ-(32)P]ATP under standard or adduct intensification conditions. Since the labeled digest includes adducted mono-, di-, and/or oligo-deoxynucleotides, it is further treated with phosphatase and phosphodiesterase prior to TLC. The labeled digest is treated with nuclease P1 (NP1) in Method I, with T4 polynucleotide kinase and NP1 in Method II, and then with phosphodiesterase I in both cases and subjected to TLC. The advantage of these methods is that the number of adduct species formed can be estimated by TLC.

Induction of abnormal nuclear shapes in two distinct modes by overexpression of serine/threonine protein phosphatase 5 in Hela cells.

Okadaic acid-sensitve serine/threonine protein phosphatase 5 (PP5) is expressed ubiquitously in various tissues and is considered to participate in many cellular processes. PP5 has a catalytic domain in the C-terminal region and three tetratricopeptide repeat (TPR) motifs in the N-terminal region, which are suspected to function as a protein-protein interaction domain. Physiological roles of PP5 are still largely unknown, although several PP5-binding proteins were reported and a few in vivo functions of PP5 were suggested. In the present study, the effects of expression of the full-length wild-type PP5 fused with EGFP (EGFP-PP5(WT)) and its phosphatase-dead mutant EGFP-PP5(H304A) were investigated. Transient expression of either EGFP-PP5(WT) or EGFP-PP5(H304A) in HeLa cells induced deformed nuclei with a 10-fold frequency compared to that of EGFP. Abnormal-shaped nuclei were also substantially increased by induced moderate expression of PP5 in tet-on HeLa cells. Many HeLa cells expressing EGFP-PP5(WT) possessed multi-nuclei separated from each other by nuclear membrane, while expression of EGFP-PP5(H304A) induced deformed nuclei which were multiple-like in shape, but not separated completely and were surrounded by one nuclear membrane. These results suggest that PP5 plays important roles at the M-phase of the cell cycle, especially in separation of chromosomes and formation of nuclear membrane.

Effect of chlorophyllin-chitosan on excretion of dioxins in a healthy man.

We investigated the usefulness of chitosan and chlorophyllin-chitosan (chl-chitosan) administration for reduction of the body burden of environmental dioxins, including polychlorinated dibenzo-p-dioxins/dibenzofurans (PCDDs/ Fs) and coplanar polychlorinated biphenyls (Co-PCBs), by examining the excretion levels in the feces and sebum of a healthy man. The volunteer ate the same three meals every day during the 40-d experiment, which was composed of five phases (I-V) of 8 d each. In phase I (days 1-8), the volunteer was given only the basal diet. In phases II-V, 0.2 g of chitosan, 0.6 g of chitosan, 0.2 g of chl-chitosan, and 0.6 g of chl-chitosan, respectively, were administered immediately after each meal. We measured daily the amount of dioxins occurring in the feces and sebum during the last 5 d of each phase. The total toxicity equivalency (TEQ) of the dioxin in phases I-V were 27, 26, 38, 36, and 67 pg/d in the feces and 20, 19, 16, 16, and 14 pg/d in the sebum, compared with 74 pg/d in the food. The excretion of dioxins in the feces was significantly increased in phases III, IV, and V, being 140% (p < 0.05), 135% (p < 0.05), and 249% (p < 0.01) of the control level (phase I). Although the dioxin in the sebum was slightly decreased in phase V as compared with the control level, the total amount of excreted dioxin in feces and sebum was increased significantly in phase V, being 174% of the control level, which is almost the same level as that in the food. This indicates that chl-chitosan can prevent accumulation of dioxin, at least at the intake level of normal foods.

Unfolding of higher DNA structures formed by the d(CGG) triplet repeat by UP1 protein.

Fragile X syndrome is caused by expansion of a d(CGG) triplet repeat in the 5'-untranslated region of the first exon of the FMR1 gene resulting in silencing of the gene. The d(CGG) repeat has been reported to form hairpin and quadruplex structures in vitro, and formation of these higher structures could be responsible for its unstable expansion in the syndrome, although molecular mechanisms underlying the repeat expansion still remain elusive. We have previously proved that UP1, a proteolytic product of hnRNP A1, unfolds the intramolecular quadruplex structures of d(GGCAG)5 and d(TTAGGG)4 and abrogates the arrest of DNA synthesis at d(GGG)n sites. Here, we demonstrate that the d(CGG) repeat forms a peculiar DNA structure, which deviates from the canonical B-form structure. In addition, UP1 was demonstrated by CD spectrum analysis to unfold this characteristic higher structure of the d(CGG) repeat and to abrogate the arrest of DNA synthesis at the site. This ability of UP1 suggests that unfolding of unusual DNA structures of a triplet repeat is required for DNA synthesis processes.

Heterocyclic amines: Mutagens/carcinogens produced during cooking of meat and fish.

Research leading to the discovery of a series of mutagenic and carcinogenic heterocyclic amines (HCAs) was inspired by the idea that smoke produced during cooking of food, especially meat or fish, might be carcinogenic. More than ten kinds of HCAs, actually produced by cooking or heating of meat or fish, have now been isolated and their structures determined, most being previously unregistered compounds. They are highly mutagenic towards Salmonella typhimurium in the presence of S9 mix and are also mutagenic in vitro and in vivo toward mammalian cells. HCAs have now been chemically synthesized in quantity and subjected to long-term animal testing. When HCAs were fed in the diet, rodents developed cancers in many organs, including the colon, breast and prostate, and one HCA produced hepatomas in monkeys. The lesions exhibited alteration in genes including Apc, beta-catenin and Ha-ras, and these changes provide clues to the induction mechanisms. The HCAs are oxidized to hydroxyamino derivatives by cytochrome P450s, and further converted to ester forms by acetyltransferase and sulfotransferase. Eventually, they produce DNA adducts through the formation of N-C bonds at guanine bases. There are HCA-sensitive and resistant strains of rodents and a search for the responsible genes is now under way. While the content of HCAs in dishes consumed in ordinary life is low and not sufficient in itself to explain human cancer, the coexistence of many other mutagens/carcinogens of either autobiotic or xenobiotic type and the possibility that HCAs induce genomic instability and heightened sensitivity to tumor promoters suggest that avoidance of exposure to HCAs or reduction of HCAs' biological effects as far as possible are to be highly recommended. Usage of microwave ovens for cooking and supplementation of the diet, for example with soy-isoflavones, which have been found to suppress the occurrence of HCA-induced breast cancers, should be encouraged. Advice to the general public about how to reduce the carcinogenic load imposed by HCAs would be an important contribution to cancer prevention.

Unfolding of quadruplex structure in the G-rich strand of the minisatellite repeat by the binding protein UP1.

The mouse hypervariable minisatellite (MN) Pc-1 consists of tandem repeats of d(GGCAG) and flanked sequences. We have previously demonstrated that single-stranded d(GGCAG)(n) folds into the intramolecular folded-back quadruplex structure under physiological conditions. Because DNA polymerase progression in vitro is blocked at the repeat, the characteristic intramolecular quadruplex structure of the repeat, at least in part, could be responsible for the hypermutable feature of Pc-1 and other MNs with similar repetitive units. On the other hand, we have isolated six MN Pc-1 binding proteins (MNBPs) from nuclear extracts of NIH 3T3 cells. Here, we describe one of those MNBPs, MNBP-B, that binds to the single-stranded d(GGCAG)(n). Amino acid sequences of seven proteolytic peptide fragments of MNBP-B were determined, and the cDNA clones were isolated. MNBP-B was proven identical to the single-stranded DNA-binding protein, UP1. Recombinant UP1 bound to single-stranded d(GGCAG)(n) and other G-rich repetitive sequences, such as d(GTCAGG)(n) and d(GTTAGG)(n). In addition, UP1 was demonstrated by CD spectrum analysis to unfold the intramolecular quadruplex structure of d(GGCAG)(5) and d(TTAGGG)(4) and to abrogate the arrest of DNA synthesis at the d(GGG)(n) site. This ability of UP1 suggests that unfolding of quadruplex DNA is required for DNA synthesis processes.

Efficient induction of rat large intestinal tumors with a new spectrum of mutations by intermittent administration of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine in combination with a high fat diet.

In the present study we have established novel intermittent protocols featuring a high fat (HF) diet for efficient induction of large intestinal tumors with a relatively small amount of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP). In protocol 1, F344 male rats were first fed a diet containing 400 p.p.m. PhIP for 2 weeks, followed by continuous administration of a HF diet without PhIP for 108 weeks. In protocol 2, 2 week PhIP treatments were repeated three times with 4 week intervals on the HF diet alone, followed by continuous feeding of the HF diet for 42 weeks. At termination of the experiments, 16 (3 of 19) and 45% (9 of 20) of the rats had developed a total of three and 13 large intestinal tumors with protocols 1 and 2, respectively. The tumor incidence in protocol 2 was comparable with that observed with continuous feeding of 400 p.p.m. PhIP for 52 weeks, after exposure to only approximately 10% of the amount of carcinogen. Five of nine (55%) tumors harbored mutations in either the beta-catenin or Apc gene, while all demonstrated accumulation of beta-catenin protein in the cytoplasm and nucleus. This suggests that other unknown genetic alterations in the Wnt-Apc-beta-catenin signaling pathway could have been involved in the development of tumors. By further modifying this intermittent protocol with HF diet, one could expect more efficient induction of lesions with much smaller amounts of PhIP in a shorter period. In addition, this model could provide a means to elucidate genetic alterations in large intestinal tumors induced by relatively low levels of carcinogenic insult, mimicking the cases of human colon carcinogenesis induced by exposure to environmental carcinogens.

Induction of intestinal tumors and lymphomas in C57BL/6N mice by a food-borne carcinogen, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine.

2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) is the most abundant heterocyclic amine contained in cooked meat and fish. Although PhIP has been demonstrated to induce various types of tumors in rats, lymphomas predominated in mice using the CDF1 strain. To investigate the carcinogenic activity of PhIP on other organs in mice with a different genetic background, PhIP was administered to C57BL / 6N mice. After a 40-week administration of 300 ppm of PhIP in a high-fat diet followed by continuous feeding with a high fat diet, C57BL / 6N mice developed adenomas and adenocarcinomas in the small intestine, the incidences being 52% in males and 68% in females at weeks 95 and 70, respectively. Lymphomas of B-cell origin also developed in both sexes as frequently as in the CDF1 strain, incidences being 48% in males and 32% in females. Although the incidence in PhIP-treated female mice did not differ from that in the control mice, lymphomas developed significantly earlier in the PhIP-treated mice. The present study demonstrated that the intestinal tract is another potential target of PhIP-induced carcinogenesis in mice, and that the carcinogenic activity of PhIP could be affected by the genetic background of the animals.

LRP130, a protein containing nine pentatricopeptide repeat motifs, interacts with a single-stranded cytosine-rich sequence of mouse hypervariable minisatellite Pc-1.

Recently, we have identified and purified minisatellite DNA binding proteins (MNBPs) that bind to the mouse hypervariable minisatellite Pc-1, from NIH3T3 cells. This study describes the isolation and characterization of a mouse leucine-rich protein (mLRP130) as one of the MNBPs that binds to the C-rich strand of Pc-1. The mLRP130 cDNA was demonstrated to encode a polypeptide of 1306 amino-acid residues with a deduced molecular mass of 137 kDa, and the mLRP130 mRNA is detected in various organs, including heart, brain, liver, skeletal muscle, kidneys and testes. The mLRP130 protein has nine copies of pentatricopeptide repeat (PPR) motifs that are considered to serve as protein-protein interactions. Two forms of the mLRP130 protein were detected in NIH3T3 cells with an approximate molecular mass of 140 kDa (mLRP130) and 100 kDa (mLRP130der), and were detected mainly in nuclear and cytoplasmic fractions, respectively. Immunofluorescence microscopic analysis demonstrated dominant localization of mLRP130 at the perinuclear region, and also in the nucleus and cytoplasm with dot- or squiggle-like staining. The immunoprecipitated mLRP130 bound to the single-stranded d(CTGCC)8, but not to its complementary G-rich strand of d(GGCAG)8 or double-stranded form. Possible biological roles of mLRP130 are discussed in association with the stability of minisatellite DNA sequences.

LRP130, a single-stranded DNA/RNA-binding protein, localizes at the outer nuclear and endoplasmic reticulum membrane, and interacts with mRNA in vivo.

LRP130 (also known as a LRPPRC) is an RNA and single-stranded DNA-binding protein, and recently identified as a candidate gene responsible for the Leigh syndrome, a French-Canadian type cytochrome c oxidase deficiency. However, the biological function of LRP130 still remains largely unresolved. In the present study, we found that the C-terminal half of the mouse LRP130 located within a 120 amino acid sequence (a.a. 845-964) binds to synthetic RNA homopolymers, poly(G), poly(U), and poly(C), as well as r(CUGCC)(6). Assessment of the subcellular localization indicated both nuclear/endoplasmic reticulum (ER) and mitochondrial fractions to be positive. To further analyze the subcellular localization of LRP130, a nuclear/ER fraction was fractionated into the nucleoplasm (NP) and nuclear envelope (NE)/ER, and the latter was further separated into outer nuclear membrane (ONM)/ER and inner nuclear membrane (INM) by treatment with Triton X-100. LRP130 was detectable in all three fractions, and the distribution pattern was in good accordance with that known for ONM/ER proteins. Interestingly, immunostaining of HeLa cells demonstrated nuclear rim staining of LRP130, specifically at the outside of the NE and also at ER, and association of LRP130 with poly(A)(+) RNA was restricted only to the ONM/ER fraction. Overexpression of full-length mouse LRP130 fused with EGFP resulted in nuclear accumulation of poly(A)(+) RNA in HeLa cells. Taking all these results together, it is suggested that LRP130, a novel type of RNA-binding protein, associates with mRNA/mRNP complexes at the outside of NE and ER, and plays a role in control of mRNA metabolisms.

Effective pretreatment of human serum samples for dioxin analysis by solid phase extraction and blue-chitin column cleanup.

For the analysis of dioxins, polychlorinated dibenzo-p-dioxins/dibenzofurans (PCDDs/Fs) and coplanar polychlorinated biphenyls (Co-PCBs), devising simple and economical methods is important, especially for mass screening of human exposure. Pretreatment of samples, namely the extraction and cleanup methods that are widely used at present, needs to be improved for savings in time, manpower, and solvent consumption. In the present study, we applied solid phase extraction (SPE) using octadecyl (C18) and a blue-chitin column in place of liquid-liquid extraction (LE) and an active-carbon column with serum samples, frequently used for assessment of human exposure. Efficacy of the new pretreatment methods was demonstrated by successful high resolution gas chromatography-high resolution mass spectrometry (HRGC-HRMS) of the major 17 PCDDs/Fs and 12 Co-PCBs that are on the list of WHO/IPCS (1997) hazardous dioxins with toxic equivalent factor (TEF) values. SPE is timesaving and requires less manpower and organic solvent as compared with the LE that is presently widely used. Concerning cleanup with blue-chitin, the amount of toluene applied as eluent could be reduced to 1/3, as compared with the active-carbon case. The combination of SPE and blue-chitin for pretreatment of serum saves time and manpower, is accurate and uses less organic solvent than LE with active carbon cleanup.

Frequent and multiple mutations at minisatellite loci in sporadic human colorectal and gastric cancers--possible mechanistic differences from microsatellite instability in cancer cells.

Minisatellites (MNs), composed of 5 to 100 nucleotide repeat units, range from 0.5 to 30 kb in length, and have been reported to be mutated in various human malignancies. In this study, frequencies of MN mutations in sporadic human colorectal (34 cases) and gastric cancers (24 cases) at various clinicopathological stages were assessed by multilocus DNA fingerprint analysis with three MN probes, Pc-1, 33.6 and 33.15. MN mutations were observed in both colorectal and gastric cancers, but at a significantly higher frequency in the former (56%) than in the latter (25%). Multiplicities of MN mutations were 1.50 +/- 1.81 and 0.46 +/- 1.10 in colorectal and gastric cancers, respectively, and the difference was also significant. Neither the presence nor multiplicity of MN mutations in either colorectal or gastric cancer cases had any correlation with the pathological stage, histological grading or the presence of microsatellite instability (MSI). Although the biological relevance of MN mutations still remains to be clarified, a subset of colorectal and gastric cancers could feature a new type of genomic instability, distinct from MSI.

Destruction of quadruplex by proteins, and its biological implications in replication and telomere maintenance.

The minisatellite DNA Pc-1 consists of tandem repeats of d(GGCAG). We previously reported that a d(GGCAG)n strand folds into an intramolecular quadruplex under physiological conditions and that during replication the progression of DNA polymerase is blocked by the quadruplex in vitro. Therefore, the formation of the quadruplex was supposed to be responsible for the hypermutable features of Pc-1. Then, we have identified proteins that bind to Pc-1, one of which is hnRNP A1. Here, we have demonstrated that hnRNP A1 destroys the quadruplex of Pc-1 on binding and abrogates the arrest of DNA polymerase at the repeat. Thus, hnRNP A1 functions as if it is a chaperon to assist Pc-1 DNA to form the proper folding suitable for replication. We have also found that hnRNP A1 and a related protein, hnRNP D, destroy the quadruplex of telomere DNA, which suggests the involvement of these proteins in telomere maintenance as DNA chaperons.