Search

Proliferative diabetic retinopathy is an important cause of visual impairment. We investigated whether the polymorphism of the beta 3-adrenoreceptor (beta 3-AR) gene, which is associated with insulin resistance and an earlier onset of NIDDM, was associated with proliferative diabetic retinopathy (PDR) in 215 Japanese NIDDM patients with a duration of diabetes of > or = 10 years. The polymorphism of the beta 3-AR gene was determined by polymerase chain reaction-restriction fragment length polymorphism analysis. The Trp64Arg allele of the beta 3-AR gene was significantly more frequent in the NIDDM patients with PDR (P = 0.002), but not in those with non-PDR (P = 0.151), than in NIDDM patients without diabetic retinopathy. Those with the mutation had an earlier onset of diabetes, a longer duration of diabetes, and higher current and maximal BMI values, compared with those without the mutation. Moreover, this mutation was also associated with higher serum triglyceride and decreased HDL-cholesterol levels. When adjustment was made for age, age at diagnosis, duration of diabetes, current BMI, systolic blood pressure, HbA1e, and serum lipids in a multiple regression analysis, a significant association was found between the Trp64Arg allele and diabetic retinopathy (P = 0.039). The Arg/Arg or Arg/Trp genotype was significantly associated with PDR, compared with the Trp/Trp genotype, with an odds ratio of 2.55 (95% CI 1.25-5.16). We concluded that the beta 3-AR gene polymorphism is a newly identified risk factor for PDR.

Ninety-four esophageal squamous cell carcinomas were examined for loss of heterozygosity at several loci on the long arm of chromosome 17 (17q), using restriction fragment length polymorphism markers. Loss of heterozygosity was observed in 56 (62%) of 91 tumors that were informative with at least one marker. Comparison of these results with clinicopathological data indicated that the losses on chromosome 17q had occurred at an early stage of carcinogenesis. Detailed deletion mapping in these tumors revealed that the region commonly deleted was within the segment between loci defined by two markers at chromosomal band 17q21.3.

We have constructed a high-resolution cytogenetic map of human chromosome 17 with 342 cosmid markers, each newly isolated from a cosmid library constructed from a human-mouse hybrid cell line containing a single human chromosome 17. Direct mapping on R- and/or G-banded (pro)metaphase chromosomes by fluorescence in situ hybridization localized these markers throughout the chromosome, although density was highest in the R-band-dominant regions of 17p13, 17p11.2, 17q11.2-q12, 17q21.3, 17q23, and 17q25. By screening some of the cosmid clones, we identified 71 polymorphic systems with 43 markers; 11 of these are VNTRs. As our high-resolution cytogenetic map contains a large number of markers, it can provide useful landmarks for a contig map of chromosome 17. Furthermore, the map will contribute to positional cloning of aberrant genes responsible for inherited diseases such as Miller-Dieker syndrome (MDS), Smith-Magenis syndrome (SMS), and familial early-onset breast cancer, as well as putative tumor suppressor genes on this chromosome.

A genetic linkage map of markers for the short arm of human chromosome 8 has been constructed with 14 polymorphic DNA markers on the basis of genotypes obtained in 40 CEPH reference families. This unbroken map spans 45 cM in males and 79 cM in females. The 14 markers include three genes, MSR, LPL, and NEFL, and one anonymous DNA segment that were previously assigned to chromosome 8. The other 10 marker had been isolated from a chromosome 8-specific cosmid library and physically localized to chromosomal bands by fluorescence in situ hybridization. The order of loci determined by genetic linkage was consistent with their physical locations. This map will facilitate efficient linkage studies of human genetic diseases that may be segregating on chromosome 8p and will provide anchor points for development of high-resolution maps for this chromosomal region.

We have constructed a genetic linkage map of the distal long arm of human chromosome 6 with 16 polymorphic DNA loci, including six VNTRs, based on genotypic analysis in 40 CEPH pedigrees. The genetic map spans 31 cM and the distance is similar to that estimated by physical methods. The order of loci is basically consistent with physical localizations determined by fluorescence in situ hybridization. This map will be useful for isolation of putative tumor suppressor genes on chromosome 6q that are implicated in certain malignancies and will assist studies of genetic diseases.

To obtain new RFLP markers for construction of a high-resolution map of human chromosome 8, a cosmid library was constructed from a somatic hybrid cell that contained chromosome 8 as the only human component in mouse genomic background. Eighty-eight new RFLP markers were isolated and characterized, and 71 of them were sublocalized to chromosomal bands by fluorescent in situ hybridization (FISH). Of these, 36 were localized to the short arm, 34 to the long arm, and 1 to the centromeric region. Five markers defined VNTR loci. This work represents the first extensive isolation and physical mapping of RFLP markers on human chromosome 8. These new markers will serve as useful resources for linkage mapping of loci for inherited diseases and for efforts to identify a putative tumor suppressor gene(s) on chromosome 8.

Fifty-four clones containing human inserts were selected from a cosmid library constructed from a somatic cell hybrid containing chromosome 11p15.3-p15.5 as its only human complement. In 32 of these clones, 63 polymorphic systems were identified with a panel of restriction enzymes: 57 conventional RFLP systems and 6 highly polymorphic VNTR systems. Although we examined the cosmid with only seven enzymes, 18 clones (including 6 VNTRs) were polymorphic with three or more enzymes. The results suggested that DNA sequences on the peritelomeric region of chromosome 11p tend to be highly variable. Because these markers are highly informative, they will be excellent resources for investigations of hereditary diseases and tumor suppressor genes in this region of chromosome 11.

We have constructed a high-resolution cytogenetic map with 168 DNA markers, including 90 RFLP markers for human chromosome 11. The cosmid clones were mapped by fluorescence in situ suppression hybridization, in which discrete fluorescent signals can be detected directly on prometaphase R-banded chromosomes. Although these cosmid clones were distributed throughout the chromosome, they had some tendency to localize in the regions of R-positive band, such as 11p15, 11p11.2, 11q13, 11q23, and 11q25. Since these regions of chromosome 11 are considered to contain genes responsible for certain genetic diseases, cancer breakpoints involved in chromosome rearrangements, and tumor-suppressor genes, this high-resolution cytogenetic map will contribute to the molecular characterization of such genes. This map will also provide many landmarks essential for construction of the complete physical map with contigs of cosmid and YAC clones.

Thirty-one new RFLP systems corresponding to 24 loci have been identified from a chromosome 10-specific cosmid library. Twelve of the markers on the proximal long arm (cen-q11.2) of this chromosome, including four RFLP systems for the RET locus, will be especially useful in efforts to identify the gene responsible for multiple endocrine neoplasia type 2A (MEN2A). The new panel of markers also may contribute to fine-scale mapping of tumor suppressor genes associated with glioblastoma multiforme or renal cell carcinoma, because allelic deletions in these tumors have implied the presence of a tumor suppressor gene(s) on chromosome 10.

We have isolated 68 new RFLP markers on human chromosome 6. Of these, 64 were localized on chromosomal bands by the fluorescent in-situ hybridization (FISH) method, 25 on the short arm and 39 on the long arm. Their distribution was uneven; the markers were localized predominantly in regions of R-positive banding. Eleven markers defined VNTR loci. This expanded collection of DNA markers will contribute to high-resolution linkage mapping of genes causing inherited disorders and will provide useful reagents for isolation of putative tumor-suppressor genes on chromosome 6 that appear to be involved in malignancies. Furthermore, the new markers will be guideposts for detailed linkage and physical maps of this chromosome.

In order to detect common regions of deletion, 219 breast tumors were examined for loss of heterozygosity at several loci on chromosomes 3p, 16q, and 17 by restriction fragment length polymorphism analysis. Allelic deletions of loci on chromosomes 3p, 13q, 16q, and 17, and amplification of the erbB2 oncogene, were analyzed and compared with histopathological and clinical features. Common regions of deletion were detected within chromosomal bands 3p13-14.3, 16q22-23, 17p13 (two separated loci), and 17q21. Concordant losses of alleles on chromosomes 3p, 13q, 16q, 17p, and 17q were observed. A significant association was detected between loss of heterozygosity on chromosomes 17p and 17q and amplification of the erbB2 oncogene (17p, P = 0.000721, by Fisher's exact test; 17q, P less than 0.001, chi 2 = 12.135). Furthermore, tumors showing highly malignant phenotypes had accumulated more genetic changes at the loci studied than those having less malignant phenotypes on the basis of histopathological classification, lymph node metastasis, and tumor size. These results suggested that accumulation of genetic alterations, including loss of function of tumor suppressor genes on chromosomes 3p, 13q, 16q, and 17, and amplification of the erbB2 oncogene, may contribute to tumor development and/or progression in primary breast cancer.

A genetic linkage map for human chromosome 3 has been constructed using 41 polymorphic DNA markers genotyped in 40 CEPH reference families. The map spans a genetic distance of 261 cM in males and 413 cM in females; the ratio of these distances (approximately 1.6 in favor of female meioses) was fairly constant across the map. Frequency of recombination was relatively uniform throughout much of the chromosome, except that in both telomeric regions recombination was more frequent than the physical distances would predict. The genetic map was basically in agreement with physical localization of 24 loci that were mapped by fluorescent in situ hybridization. This map can be used for linkage studies for genetic diseases, and it will serve as a step toward a high-resolution map for human chromosome 3.

It has been shown that a nuclear factor, interferon regulatory factor-1 (IRF1), plays a key role in the regulation of the type I interferon (IFN) system by affecting the transcription of type I IFN and some IFN-inducible genes. In this study, the human IRF1 gene was mapped to chromosome 5 by analysis of mouse-human somatic cell hybrids. Furthermore, by linkage analysis based on RFLP (restriction fragment length polymorphism), the gene was localized to chromosome 5q23-q31, the region that has been termed "critical" for some myeloid disorders.

To obtain new RFLP markers on human chromosome 11 for a high-resolution map, we constructed a cosmid library from a Chinese hamster x human somatic hybrid cell line that retains only human chromosome 11 in a Chinese hamster genomic background. A total of 3,500 cosmids were isolated by colony hybridization with labeled human genomic DNA. DNA was prepared from 130 of these cosmid clones and examined for RFLP. In 62 of them, polymorphism was detected with one or more enzymes; four RFLPs were VNTR systems. All polymorphic clones were assigned to one of 22 intervals obtained by mapping on a deletion panel of 15 somatic hybrid cell lines containing parts of chromosome 11; 11 clones were finely mapped by in situ hybridization. Although RFLP markers were scattered on the whole chromosome, they were found predominantly in the regions of R-banding. These DNA markers will contribute to fine mapping of genes causing inherited disorders and tumor-suppressor genes that reside on chromosome 11. Furthermore, as one-third of the cosmid clones revealed a band or bands in Chinese hamster DNA, indicating sequence conservation, this subset of clones may be useful for isolating biologically important genes on chromosome 11.

Our laboratory has constructed linkage maps of the human chromosomes to use as a tool towards the goal of cloning by position the genes responsible for genetic disorders. Construction of the map required the development of polymorphic marker systems in the form of Restriction Fragment Length Polymorphisms (RFLPs). Work by Yusuke Nakamura in the laboratory led to the identification of more than 200 highly informative Variable Number Tandem Repeat (VNTR) markers. The hypervariable nature of these marker loci has allowed individualization at the DNA level. Techniques for individualization have subsequently been adopted by diverse fields including gene mapping, cancer genetics and forensic biology. These markers have also become a resource to test hypotheses as to how the VNTRs generate their intrinsic variability. We have demonstrated that the hypothesis that VNTRs generate their variability by unequal exchange between homologous chromosomes in incorrect (Wolff et al., 1988; Wolff et al., 1989). Our data are consistent with intrachromosomal models such as unequal sister chromatid exchange and replication slippage. Using DNA derived from nonhuman primate species, we have tested hypotheses that try to explain the sequence relationship at dispersed VNTR loci. Our data reveal that VNTR loci are most likely not related by transposition but rather arose independently at multiple loci.

A genetic linkage map of six polymorphic DNA markers close to the gene (APC) for familial adenomatous polyposis (FAP) on chromosome 5q is reported. One hundred fifty-five typed members of nine FAP kindred provided more than 90 meioses for linkage analysis. A number of crucial recombination events have been identified which are informative at three or more loci, allowing confident ordering of parts of the map. There was no evidence of genetic heterogeneity, with all families showing linkage of at least one chromosome 5 marker to the gene. Recombination data and two-point linkage analysis support a locus order of centromere-pi 227-C11P11-ECB27-L5.62-APC-EF5.44-YN5.48-telomer e, although EF5.44 could lie in the interval L5.62-APC or ECB27-L5.62. No recombinants were identified between APC and either EF5.44 or YN5.48, but published deletion mapping in colorectal carcinomas and linkage analysis in FAP suggest that YN5.48 is 1-3 cM from APC. The present study suggests that YN5.48 and L5.62 delineate a small region of chromosome 5 within which the EF5.44 locus lies very close to the APC gene. These data not only allow use of flanking markers for presymptomatic diagnosis of FAP but also provide a high-density map of the region for isolation of the APC gene itself and for further assessment of the role of chromosome 5 deletions in the biology of sporadic colorectal cancer.

We describe 14 new restriction fragment length polymorphisms, corresponding to 13 loci on the long arm of chromosome 11. A detailed genetic map of chromosome 11q has been constructed from these and other loci (a total of 31 loci) typed in 59 reference families. The 23 most informative markers were selected to establish a map with a strongly supported order; regional localizations are provided for eight other markers. The loci span 88 cM in males and 148 cM in females and form a dense continuum on 11q. These ordered polymorphic markers will be of help in studying the genes responsible for several diseases that have been localized to this region, including genes responsible for multiple endocrine neoplasia type I (MEN1), ataxia telangiectasia (AT), tuberous sclerosis (TSC), and some forms of asthma and rhinitis.

The human vimentin gene and a random DNA segment (D10S39) were mapped to the short arm of human chromosome 10 by linkage analysis. A panel of somatic cell hybrids and monosomy cell-lines, which divide chromosome 10 into seven regions, was used to localize 10 polymorphic markers on this chromosome. The physical map locations obtained correlate well with linkage maps of chromosome 10. Two markers which have been shown to be closely linked to the gene for multiple endocrine neoplasia type 2A map distal to a translocation breakpoint in band 10q11.2.

Elsewhere we have reported an efficient method for isolating VNTR (Variable Number of Tandem Repeats) markers. Several of the VNTR markers isolated in those experiments were sequenced, and a DNA sequence of 9 bp (GNNGTGGG) emerged as an apparent consensus sequence for VNTR markers. To confirm this result and to develop more VNTR markers, we synthesized nine different 18-base-long oligonucleotides whose sequences each included GNNGTGGG. When 102 cosmid clones selected by these oligonucleotides were tested for polymorphism, 34 (33%) of them showed multiallelic VNTR polymorphisms (average heterozygosity 68%). This procedure represents a new and efficient approach for isolating additional VNTR markers and supports the idea that the GNNGTGGG sequence may play an important role in the generation of the multiallelic systems within the human genome.

Human osteosarcomas frequently show loss of alleles on chromosome 17 as well as those on chromosome 13 that harbors the retinoblastoma gene, indicating concerted operation of another tumor-suppressing gene on chromosome 17. To assign the affected gene to a defined region of chromosome 17, we performed mitotic recombination/deletion mapping by the use of 10 polymorphic loci on chromosome 17. Of 37 tumors studied, 28 (75.7%) showed loss of heterozygosity on chromosome 17. The affected regions varied among tumors, ranging in extent from a whole chromosome to a distal segment of the short arm. However, allele loss in one region, notably in 17p13 between D17S1 and D17S30, was common to all 28 tumors, suggesting the presence of a tumor-suppressing gene in this defined region.