Search

Factors predisposing to the phenotypic features of high total cholesterol (T-Cho) in human plasma have not been clearly defined. Here we report an association between two variations in the apolipoprotein B48 receptor gene (APOB48R) and plasma T-Cho levels among 352 adult individuals in Japan. By analyzing phenotypic associations between age- and gender-adjusted levels of plasma T-Cho, low-density lipoprotein (LDL) cholesterol (LDL-C), and high-density lipoprotein (HDL) cholesterol (HDL-C), we detected a significant correlation between genotypes of the A419P variation and adjusted T-Cho levels. Among homozygous G-allele carriers (n=265), heterozygous carriers (n=78), and homozygous minor C-allele carriers (n=9), T-Cho levels were 2.43+/-0.21 mg/cm(3), 2.48+/-0.24 mg/cm(3), and 2.63+/-0.21 mg/cm(3), respectively, indicating a codominant T-Cho-elevating effect of the minor C-allele (r=0.15, P=0.007). A similar effect was detected for c.934-960/del (r=0.13, P=0.015). Linkage disequilibrium (LD) analysis detected significant LD among eight variant sites that included neighboring loci. Our results indicate that variations in APOB48R and nearby genes are among the many factors involved in hypercholesterolemia. The etiological studies should now include consideration of this novel aspect of the mechanism(s) leading to hypercholesterolemic disease.

Genes encoding the serine proteinase inhibitor B family (SERPINBs) are mainly clustered on human chromosome 18 (18q21). Several serpins are known to affect malignant phenotypes of tumor cells, so aberrant genetic variants in this molecular family are candidates for conferring susceptibility for risk of cancer. We investigated whether eight selected non-synonymous variations within SERPINB loci at 18q21 might be associated with risk of prostate cancer in Japanese men. A case-control study involving 292 prostate-cancer patients and 384 controls revealed significant differences in regard to distribution of four missense variations in genes encoding plasminogen activator inhibitor 2 (PAI2) and SERPINB10. The most significant association was detected for the N120D polymorphism in the PAI2 gene (P = 5.0 x 10(-5)); men carrying the 120-N allele (120-N/N and 120-N/D genotypes) carried a 2.4-fold increased risk of prostate cancer (95% confidence interval 1.45-4.07). Associations were also detected for three other missense polymorphisms in those two genes. Strong linkage disequilibrium in the region encompassing PAI2 and SERPINB10 extended to about 50 kbp. The results suggested that missense variations in one or both of these genes confer important risks for prostate cancer, and may be themselves tumorigenic. Although confirmative replication studies on larger cohorts are awaited, clinical examination of these variations may become useful for identifying individuals at high risk for prostate cancer.

Among multiple factors influencing osteoporosis, genetic variations involved in bone-mineral metabolism can affect risks predisposing to the disease onset. Here, we studied single-nucleotide polymorphisms (SNPs) in the pro-opiomelanocortin (POMC) gene for possible association with bone mineral density (BMD) among 384 adult Japanese women and observed significant correlation between adjusted BMD and three SNPs in the promoter region (r>0.14, p<0.01). The most significant correlation was observed for -2353G/A (r=-0.16, p=0.002); homozygous carriers of the major (G) allele had the highest BMD (0.405+/-0.054 g/cm2) while heterozygous carriers were intermediate (0.390+/-0.053 g/cm2) and homozygous A-allele carriers had the lowest BMDs (0.369+/-0.048 g/cm2). Although no association was detected between these SNPs and body weight or body mass index (BMI), significant association was detected between the -2313A/C genotype and plasma total cholesterol level (r=-0.12, p=0.019). We propose that POMC is among the likely susceptibility genes for osteoporosis and may also be involved in dyslipidemia.

To disclose genetic mechanisms involved in development or progression of hepatocellular carcinoma (HCC), we used a genome-wide cDNA microarray consisting of 8,448 genes to compare gene-expression profiles among 12 liver-cirrhosis nodules (LCNs) and five specimens of HCC excised from a single patient and carefully prepared by laser-capture microdissection (LCM). The expression patterns enabled us to identify 72 genes that were frequently upregulated and 57 that were downregulated specifically in the LCN specimens as compared to the HCCs. We also documented upregulation of 31 genes and downregulation of seven others in both HCC and LCN tissues. Several types of intracellular kinase, including receptor-type kinase, were upregulated in LCNs. Expression patterns of HCCs and LCNs generally represented two genetically distinct groups when subjected to a hierarchical clustering analysis, although expression profiles of two of the LCNs resembled the HCC pattern. Analysis of allelic losses at microsatellite loci revealed that LCNs showed frequent loss of heterozygosity (LOH) (33%) in chromosomal regions 6q and 22q; over half of the LCNs had lost an allele for at least one of the 28 loci examined. The presence of early genetic changes among LCNs, with additional genetic changes occurring during formation of HCCs, suggests that hepatocellular carcinogenesis follows the multistep model established for colon cancers and that some LCNs may be precancerous lesions.

Factors predisposing to the phenotypic features of higher total cholesterol (T-Cho) have not been clearly defined. Here we report an association between a C/T single nucleotide polymorphism at IVS17+8 in the inter-alpha-trypsin inhibitor heavy chain 4 gene (ITIH4) and plasma total cholesterol levels in 351 adult individuals from an east-central area of Japan. Age and gender-adjusted levels of plasma T-Cho, LDL-cholesterol, triglyceride, and HDL-cholesterol were analyzed. When we separate the subjects into two genotypic groups regarding this single nucleotide polymorphism (SNP), those who lack the T-allele had significantly higher plasma T-Cho levels than the others who bear T-allele (mean 252.3 mg/dl versus 241.7 mg/dl; p=0.009). Of the 309 individuals without the T-allele, approximately 90% presented with hypercholesterolemia, whereas only 10% were hypercholesterolemic among 42 individuals with the T-allele (p <0.0001). These data suggest that genetic variation at ITIH4 locus is one of the likely candidate determinants for plasma cholesterol metabolisms.

Plasma lipid and lipoprotein in general reflect the complex influences of multiple genetic loci, for instance, even familial hypercholesterolemia (FH), a representative example of monogenic hyperlipidemia, often presents with phenotypic heterogeneity. In the course of investigating familial coronary artery disease in Utah, we studied 160 members of an eight-generation extended family of FH in which 69 members were affected with type IIa hyperlipoproteinemia (HLPIIa; high plasma cholesterol) and ten with type IIb hyperlipoproteinemia (HLPIIb; high plasma cholesterol as well as plasma triglyceride). Soluble epoxide hydrolase ( EPHX2, sEH) plays a role in disposition of epoxides in plasma lipoprotein particles. Intrafamilial correlation analysis of the modifier effect of Glu287Arg substitution in the EPHX2 gene was carried out among 79 LDLR mutation carriers and 81 noncarriers. In the carriers, plasma cholesterol levels were elevated among carriers of the 287Arg allele (mean +/- SD=358 +/- 72 mg/dl) in comparison with 287Glu homozygotes (mean +/- SD=302 +/- 72 mg/dl) (p=0.0087). Similarly, in the LDLR mutation carriers, the plasma triglyceride levels were elevated among carriers of the 287Arg allele (mean +/- SD=260 +/- 100 mg/dl) in comparison with 287Glu homozygotes (mean +/- SD=169 +/- 83 mg/dl) (p=0.020). No such gene-interactive effect was observed among noncarriers of the LDLR mutation. Half of the patients who presented with HLPIIb had inherited a defective LDLR allele as well as an EPHX2-287Arg allele, whereas the majority who presented with HLPIIa had a defective LDLR allele but not an EPHX2-287Arg allele. These results indicate a significant modification of the phenotype of FH with defective LDLR allele by EPHX2-287Arg variation in our studied kindred.

Factors predisposing to the phenotypic features of hypertriglyceridemia have not been clearly defined. Here we report an association between a missense coding region polymorphism Asn985Tyr in the retinitis pigmentosa 1 gene ( RP1) and plasma triglyceride (TG) levels in 332 adult individuals from an east-central area of Japan. Age and gender-adjusted levels of LDL-cholesterol, TG, and HDL-cholesterol were analyzed. When we separate the subjects into two genotypic groups regarding this amino acid variation, those who lack the 985-Asn allele (asparagine at residue 985) had significantly higher plasma TG levels than the others who had at least one 985-Asn allele (mean: 175.8 mg/dl vs 123.3 mg/dl; P=0.0006, Mann-Whitney test). Similarly, the former subjects had significantly lower HDL-cholesterol levels than the latter (mean: 48.0 mg/dl vs 53.8 mg/dl; P=0.038). Of the 280 individuals without a 985-Asn allele, approximately half of the individuals presented with hypertriglyceridemia, whereas only a quarter were hypertriglyceridemic among 52 individuals with the 985-Asn allele ( P=0.04). Although this SNP marker may itself be in linkage disequilibrium with other unexamined functional variants within this locus, our data suggest that genetic variation at the RP1 locus is one of the likely candidate determinants for plasma triglyceride and HDL-cholesterol metabolisms.

Twin and family studies had shown that genetic factors are important determinants of bone mass. Multiple genes might be involved. One candidate gene, the reversion-induced LIM gene ( RIL), is a PDZ and LIM-domain-containing protein and has been localized within the cytokine cluster of chromosome 5 (5q31.1). In a genetic study of 370 adult Japanese women, we investigated the correlation between radial bone mineral density (BMD) and a genetic variation (-3333T-->C) of the 5'-flanking region of RIL gene. A significant association was identified between the RIL variation -3333T-->C and radial BMD ( r=0.15, P=0.003). The variation of the RIL locus may be an important determinant of osteoporosis.

Natural tooth loss represents a major medical issue within the elderly population, since it impairs masticatory function critical for oral intake of essential nutrition. Contribution of genetic factors has been implicated in the determination of natural tooth loss; degree of reduction in number of natural teeth remaining intact (NTI) varies among individuals; thus, heterogeneity in NTI might reflect genetic variation within the population. One candidate gene, the matrix Gla protein gene (MGP), has been implicated in the pathogenesis of bone loss through a repression of bone/tooth formation. We have investigated a possible association between the CA repeat polymorphism at the human MGP gene locus and the NTI in 458 elderly Japanese women. In 916 chromosomes tested, ten alleles of the polymorphic nucleotide repeat were observed (designated A1-A10), among which five alleles were regarded as major alleles to be tested for the association. Twenty-seven women who possessed an A6 allele (164 bp) had significantly higher NTI than the remaining participants (n=431), who did not carry an allele of that size (mean: 10.0 teeth vs 5.6 teeth; P=0.007, Mann-Whitney test). An eight-year longitudinal follow-up study of NTI suggested that the genetic variations at the MGP locus did not affect the rate of tooth loss in the elderly period. These results suggest that genetic variation at the MGP gene locus is associated with some determinants for tooth loss in elderly women.

Osteoporosis is believed to result from interplay among multiple environmental and genetic determinants, including factors that regulate bone-mineral density (BMD). Recent quantitative trait locus analysis in human suggested a possible involvement of chromosomal region 1p36.2-p36.3 for determination of BMD. The brain natriuretic peptide (BNP, also named NPPB) gene lies within this candidate region for BMD determination. Overexpression of the BNP resulted in skeletal overgrowth in transgenic mice. Association analysis between nucleotide variations of the BNP gene and radial BMD in 378 Japanese postmenopausal women revealed a significant association of the -381T/C variation of the BNP gene with radial BMD (r = 0.17, P = 0.01). Homozygous T-allele carriers had the lowest BMD values (0.395 +/- 0.056 g/cm(2)), homozygous C-allele carriers had the highest (0.429 +/- 0.051 g/cm(2)), and heterozygous individuals had intermediate radial BMD values (0.405 +/- 0.048 g/cm(2)), indicating a dosage effect. Accelerated bone loss also correlated with the -381 T allele in a 5-year follow-up study (r = 0.21, P = 0.017). These results suggest that variation of BNP may be an important determinant of postmenopausal osteoporosis, in part through the mechanism of accelerated postmenopausal bone loss.

Factors predisposing to the phenotypic features of higher total cholesterol (TC) have not been clearly defined. Here we report an association between a promoter SNP (-1323T>C) in G-substrate gene (GSBS) and TC levels in 368 adult individuals from an east-central area of Japan. Age and gender-adjusted levels of LDL-cholesterol, TG, TC, and HDL-cholesterol were analyzed. When we separate the subjects into two genotypic groups regarding T allele, those who bear the T allele had significantly higher plasma TC levels than the others who lack the T allele (mean; 239.6 mg/dl vs. 210.6 mg/dl; p=0.003; Mann-Whitney test). Of the 341 individuals with the T allele, approximately 80% individuals presented with hypercholesterolemia, whereas only 44% were hypercholesterolemic among the 27 individuals without the T allele (p=0.0001). These results indicate a significant elevating effect of plasma TC levels by a SNP in the putative regulatory region of the G-substrate gene in our studied population. These data suggest that genetic variation at the G-substrate gene may be one of the determinants for plasma lipoprotein levels.

Lipid and lipoprotein concentrations in plasma generally reflect complex influences of multiple genetic loci. Even an autosomal dominant disorder, familial hypercholesterolemia (FH), is characterized by phenotypic heterogeneity, as low-density lipoprotein (LDL) levels vary widely within the same pedigree. Molecular screening for LDL receptor ( LDLR) mutations among 75 patients with clinically apparent FH led to identification of a novel splice-site mutation (IVS14+1 G>A) shared by 14 patients. Genealogical research confirmed that all 14 carriers were part of the same 1135-member pedigree with a common ancestor. The mutation resulted in an abruptly truncated LDLR protein, reducing functional LDLR activity by half in heterozygous carriers of the mutant allele. Of the 208 members of the kindred who were screened for the presence of this LDLR mutation, we identified 94 carriers and 114 noncarriers. Nine principal apolipoprotein genes that might affect LDL cholesterol differentially according to LDL-receptor status were examined in this pedigree. Strikingly lower total cholesterol and LDL-cholesterol values were observed among the majority of the LDLR mutation carriers who were simultaneously homozygous for the -265C variant of apoA-II (total cholesterol: 324 +/- 8 vs 244 +/- 19 mg/dl, P = 0.0015; LDL-cholesterol: 237 +/- 8 vs 155 +/- 18 mg/dl, P = 0.0008). In vitro transfection assays showed that transcriptional activity of the apoA-II promoter was reduced by 30% in the -265C variant as compared with the -265T variant. We thus concluded that one variant of the apoA-II gene was associated with reduced plasma LDL cholesterol only in FH patients.

Leukemia inhibitory factor (LIF) is a pleiotropic cytokine implicated in various pathological conditions, such as rheumatoid arthritis and osteoporosis. Despite the possible importance of LIF as a therapeutic target, little is known about the bioregulation of the human LIF gene. We here sequenced the entire structure of the LIF gene of 48 alleles in the Japanese population. These experiments identified four single-nucleotide polymorphisms (SNPs) and determined their allelic frequencies from a 48-allele sequence in the Japanese population. All four SNPs found in the LIFgene were located within exon 3, that is, a C/T at nucleotide (nt) position 3951, a C/G at nt position 4376, an A/C at nt position 4442, and a G/A at nt position 5961 (nucleotide numbering starts from the ATG start codon). Based on the genotypic data, we constructed four major haplotypes in the tested population. Two-way comparisons of SNPs revealed complete linkage disequilibrium between SNPs at positions 3951, 4376, and 4442. These results may prove to be useful as genetic markers for population-based disease-association studies in osteoporosis.

Osteopontin (OPN) is one of the major noncollagenous bone matrix proteins produced by osteoblasts and osteoclasts. We systematically surveyed the entire structure of the OPN gene for single-nucleotide polymorphisms (SNPs) by directly sequencing 48 alleles derived from 24 unrelated Japanese individuals. We identified 13 SNPs in the OPN gene. Ten polymorphisms were identified in introns 1, 3, and 5; 2 in the coding region of exons 6 and 7; and 1 in the 3' untranslated region of exon 7. Allele frequencies for some of the polymorphisms were significantly different from those reported in the United States National Center for Biotechnology Information (NCBI) dbSNP database. These polymorphisms will be useful in genetic studies to evaluate the role of OPN proteins in bone metabolism.

Interleukin (IL11) is a member of the interleukin 6 (IL6)-related cytokine subfamily, which stimulates T cell-dependent development of immunoglobulin-producing B cells. IL11 is also an important paracrine regulator of bone metabolism that induces formation of osteoclasts. In the work reported here, we sequenced the entire IL11 structural gene of 48 alleles in a Japanese test population. These experiments identified ten single-nucleotide polymorphisms (SNPs) and determined their allelic frequencies. One polymorphism was identified upstream of exon 1, one in exon 3, four in intron 4 and four in the 3' untranslated region (3'UTR) of exon 5. Based on the genotype data, we constructed six haplotypes in the tested population. Two-way comparisons of SNPs revealed two combinations in complete linkage disequilibrium, one with SNPs at nucleotide positions 2753, 3644, 5154, and 5568, and another with SNPs at positions 3686, 5141, and 5734. These results will be useful in disease-association studies where a contribution of the human IL11 gene has been suspected, especially in disorders affecting immune response and bone metabolism.