| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of Biochemistry (E.H., H.H.B.) and Neuropsychiatry (J.W.K.) and Kohwang Medical Research Institute (B.-K.C., S.V.Y., J.-H.C.), College of Medicine, Kyung Hee University, Seoul 130-701, Republic of Korea; Department of Obesity Management (M.-J.K.), Graduate School of Obesity Science, Dongduk Women’s University, Seoul 136-714, Republic of Korea; Imagine Obesity Research Institute (J.-J.R., D.-J.O., T.-H.R., K.-H.K.), Seoul 138-170, Republic of Korea; Medical Science Institute (H.J.L.), Kangwon National University, Chunchon 200-170, Republic of Korea; and Department of Preventive Medicine (D.-H.S.), School of Medicine and Institute for Medical Science, Keimyung University, Daegu 704-701, Republic of Korea
Address all correspondence and requests for reprints to: Jong Woo Kim, M.D., Ph.D., Faculty of the Department of Neuropsychiatry, Kyung Hee Medical Center, Kyung Hee University, Hoigi-dong, Dongdaemun-gu, Seoul 130-701, Republic of Korea. E-mail: eyha{at}khu.ac.kr; psyjongwoo{at}freechal.com.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to test the hypothesis that single nucleotide polymorphisms (SNPs) in SDC3 are associated with obesity in the Korean population.
Design/Setting/Subjects: We conducted a population-based cohort study consisting of 229 control and 245 study subjects and a second independent study consisting of 192 control and 115 study subjects.
Main Outcome Measurement: Body mass index (BMI) was measured.
Results: First, Sdc3 mRNA expression in the brain of ob/ob mice was profoundly increased, compared with control mice. Next, all three nonsynonymous SNPs [T271I (rs2282440, C>T), D245N (rs4949184, C>T), and V150I (rs2491132, C>T)] in the SDC3 gene in control female subjects (BMI < 23, n = 229) and obese female subjects (BMI > 30, n = 245) were genotyped. We demonstrated the presence of clear ethnic differences in three nonsynonymous SDC3 SNPs among African-Americans, Chinese, Europeans, and Koreans. Of three SNPs in SDC3, rs4949184 was not associated with obesity and the other two SNPs (rs2282440 and rs2491132) were strongly associated with obesity (P < 0.0001), and the results were confirmed in the second independent study group. Haplotype analysis also revealed strong association with obesity (
2 = 76.92, P < 0.000001).
Conclusions: There are ethnic differences in the SDC3 polymorphisms, and the polymorphisms are strongly associated with obesity.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SDC is a family of four-transmembrane heparin sulfate proteoglycans (14). The functions of SDCs are diverse, ranging from involvement in cell adhesion to regulation of the signaling of heparan sulfate binding growth factors and even to organization of cell matrix adhesion and signaling (15, 16). Evidence that SDC3 is involved in the regulation of energy balance comes from the report that Sdc3 null mice respond to food deprivation with reduced reflex hyperphagia (12). Sdc3 null mice on a high-fat diet also exhibit resistance to high-fat diet-induced obesity, compared with wild-type mice, and accumulated less adipose mass presenting better glucose tolerance than wild-type controls (13). Increased plasma leptin, insulin, and glucose levels were shown in transgenic mice expressing Sdc1 in multiple somatic tissues and specific regions of the brain in which it is not normally expressed. Overexpression of Sdc1 has demonstrated hyperphagia and maturity-onset obesity. Sdc3 is expressed in hypothalamic feeding centers. The protein expression level of Sdc3 in the hypothalamus is increased by deprivation of food. It has been hypothesized that SDC3 may facilitate the antagonistic actions of agouti-related protein (12). However, the detailed mechanism of SCD3 in relation to energy balance remains to be elucidated.
Given the evidence that SDC3 is involved in the regulation of energy balance, we hypothesized that SDC3 polymorphism may be associated with obesity. Because few genetic studies concerning the syndecan 3 single nucleotide polymorphisms (SNPs), the SNPs were chosen based on the possibilities that the chosen SNP may influence the function of SDC3. Thus, we chose and investigated the nonsynonymous SNPs located in the coding region of extracellular domain and set out the study to test the hypothesis that polymorphisms in SDC3 are associated with obesity.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Two hundred forty-five obese female subjects with body mass index (BMI) greater than 30 (aged 32 ± 5 yr) and 229 control female subjects with BMI less than 23 (aged 31 ± 6 yr) were selected according to classification of Korean Society for the Study of Obesity (underweight, BMI 
18.5; normal, BMI 18.5 to 23; moderately obese, BMI 23 to 25; obesity I, BMI 25.0 to 30; obesity II, BMI > 30). Subjects with BMI between 23 and 30 were excluded in this study to eliminate ambiguous genetic effects of SDC3. Another independent study group that comprised of 115 obese female subjects with BMI greater than 30 (aged 35 ± 8 yr), and 192 control female subjects with BMI less than 23 (aged 32 ± 7 yr) were also recruited from Dongsan Medical Center, Keimyung University. All studies were carried out according to the Declaration of Helsinki guidelines. Written informed consent was also obtained from each subject. This study was approved by the Ethics Review Committee of the Medical Research Institute, College of Medicine, Kyung Hee University.
Animals and RT-PCR
Young (3–5 months) C57BL genetically obese mice (ob/ob; body weight 45.7 ± 2.1 g, n = 10) and their wild-type counterparts (body weight 21.0 ± 0.8 g, n = 10) were housed at room temperature with free access to standard food pellets and water. Also 2-month C57BL (body weight 15.5 ± 0.9, n = 20) mice were fed with either standard diet (n = 10) or 60% high-fat diet (n = 10) for 6 wk. Mice were killed by rapid neck disarticulation, and the brains were homogenized. All procedures were performed in accordance with institutional guidelines with approval from the Animal Care Committee of Kyung Hee University. Total RNA of brain was isolated by RNA isolation kit from Zymo Research (Orange, CA). RT-PCR for Sdc3 (sense, 5'-ggagttctacgcttagcagagccac-3', antisense, 5'-atctgggtgattttgaattggaggg-3') was performed. PCRs were electrophoresed through a 1.5% agarose gel and the 443-bp PCR product was visualized by ethidium bromide staining and UV irradiation. The mRNA quantification was performed by measuring the band density in triplicate using densitometer (Frog2000; Corebiosystem, Seoul, Korea).
Genotyping
Restriction fragment length polymorphism based method was used in this study. Peripheral blood samples for DNA extraction from all subjects were collected in an EDTA tube. Genomic DNA was extracted from the white blood cells with the use of a DNA isolation kit for mammalian blood (Macherey-Nagel GmbH & Co., Düren, Germany). Three SNPs in the coding region of the SDC3 gene were studied for association with obesity. The third nucleotide from 3' end of sense primer for rs2282440 was mutated from G to A to create the restriction enzyme HpyCH4III recognition site (5'-acnvgt-3') (sense, 5'-ccgcagcccctcccacagctaccacctcattgact-3', antisense, 5'-atgctctgagctgacttcctgctct-3', artificially mutated primer sequence is shown in bold). A 338-bp fragment surrounding the T271I (rs2282440) was amplified and the PCR fragment was digested with the restriction enzyme HpyCH4III (T: 264 + 74 and C: 229 + 74 + 35 bp). The D245N (rs4949184) SNP was genotyped using BcnI restriction enzyme (sense, 5'-cccaggtggagatgatgccttggcc-3', antisense, 5'-cttcctgctctgtaaactgggggca-3', T: 347 and C: 208 + 139 bp). C and T allele of the V150I (rs2491132) SNP was identified with the use of RsaI (sense, 5'-tctcaacaccctcacctgagccacc-3', antisense, 5'-ctccactaccatggctaccactgct-3', C: 205 and T: 142 + 63 bp). To validate the restriction fragment length polymorphism method, random duplication of the PCR procedures (about half of all PCR performed) were performed and PCR products of each polymorphism were validated by sequencing methods (ABI Prism 377, Applied Biosystems, Foster City, CA).
Anthropometric measurements
Body weight and height of all subjects were measured. BMI was defined as weight in kilograms divided by the square of height in meters.
Statistical analysis
All SNPs conformed to Hardy-Weinberg expectations before further analysis. Differences in allele frequencies of three SDC3 SNPs for patients and control subjects were compared using the 2 test. The odds ratio (OR) and 95% confidence interval were calculated to quantify the association between SDC3 and obesity at the 5% level of significance. The EH program was used for the investigation of the relative risks associated with haplotype and linkage disequilibrium (17). The SAS Statistical Software Package (release 8.02; SAS Institute Inc., Cary, NC) was used.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
, the expression level of Sdc3 in the brain of ob/ob mouse was increased 1.6 ± 0.08-fold (P < 0.05), compared with that in control mouse. Also, the expression level of Sdc3 in the brain of high-fat diet-induced obese mice was increased 1.5 ± 0.14-fold (P < 0.05). This result may confirm the previously reported finding that SDC3 is involved in the regulation of energy balance (18).
|
. Genotype distributions of three SNPs were all in Hardy-Weinberg equilibrium. T271I (rs2282440) and V150I (rs2491132) showed strong associations with the obesity. The frequency of the CC homozygote of T271I (rs2282440) polymorphism in the control was more than 2-fold of that in the obesity (21.0% in the control vs. 9.8% in the obesity), which suggests the possibility that CC genotype in T271I (rs2282440) polymorphism may be resistant to obesity in the Korean population. The frequency of C allele in T271I (rs2282440) in controls was 12.3% higher than that in obese subjects (45.6 vs. 33.3%, P < 0.0001). The frequency of the TT homozygote of V150I (rs2491132) polymorphism in the controls was more than 5 times higher than that in the obesity subjects (6.6 vs. 1.2%), and that of CT heterozygote in the controls was more 2 times higher than that in the obesity subjects (43.7 vs. 20.8%). The T allele of V150I (rs2491132) in the controls was about 2.5 times more frequent that in the obese subjects (28.4 vs. 11.6%, P < 0.0001). The ORs of T271I (rs2282440) and V150I (rs2491132) polymorphisms were 1.68 and 3.01, respectively. As shown in Table 1, of three polymorphisms studied, the D245N (rs4949184) polymorphism was not significantly associated with obesity (P = 0.65).
|
To confirm the association of SDC3 SNPs with obesity, we recruited a second independent study group and investigated three SDC3 SNPs (Table 2). As shown in Table 2, we found that the results from the second independent study replicated the results in Table 1
, showing that T271I (rs2282440) and V150I (rs2491132) are significantly associated with obesity. Genotype and allele frequencies of SDC3 polymorphisms in the second independent study were similar to those in Table 1, which may confirm the association of SDC3 SNPs with obesity.
|
). Of the eight haplotypes, the frequency of the most common haplotype (TCC) in obese subjects was 20.3% higher than that in control subjects (58.7 vs. 38.4%). The frequencies of the second most common haplotype CCC were not significantly different between obese and control subjects (25.1 vs. 29.7%). The haplotype frequency of TCT in obese was almost 4 times lower than that in control subjects (4.8 vs. 16.0%), and the frequency of CCT haplotype in obese subjects was almost 2 times lower than that in control subjects (5.5 vs. 10.7%). Whereas TTC and TTT haplotypes were not found in control subjects, 1.9 and 1.3% of TTC and TTT were observed in obese subjects. Statistical analysis revealed that significant association in haplotype frequency is present between control and obese subjects (2 = 76.92, P < 0.000001). In consideration of the absence of association between the D245N (rs4949184) and obesity, we evaluated haplotypes formed by T271I (rs2282440) and V150I (rs2491132) and observed stronger association (P < 0.0000000001 vs. P < 0.000001). We next evaluated the linkage disequilibrium, as quantified by the metrics D' and r2 (Table 4). SNPs T271I (rs2282440) and D245N (rs4949184) exhibited strong linkage disequilibrium (D' = 1.00). D' and r2 of T271I (rs2282440) and V150I (rs2491132) were 0.03 and 0.0003, respectively, and those of D245N (rs4949184) and V150I (rs2491132) were 0.06 and 0.0005, respectively. P values were all < 0.0001.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The C and T allele frequencies of T271I (rs2282440) SNP are reported to be 0.312 and 0.688 in Chinese, 0.432 and 0.568 in Japanese, and 0.977 and 0.023 in African-Americans, respectively (http://www.ensembl.org/). No variation is reported in Europeans (C allele frequency: 1.000). Clear ethnic difference is present. The C and T allele frequencies in the Korean population were 0.456 and 0.544, which are similar to those in the Japanese (Table 1). Ethnic difference is also found in D245N (rs4949184) SNP. The C and T allele frequencies are reported to be 1.000 and 0.000 in Chinese, 0.636 and 0.364 in African-Americans, and 0.771 and 0.229 in Europeans, respectively (http://www.ensembl.org/). The C and T allele frequencies in the Korean population (0.948 and 0.052) are close to those in the Chinese (Table 1). The ethnic difference in the C and T alleles of V150I (rs2491132) SNP is intriguing. The C and T allele frequencies of Koreans (0.716 and 0.284) are most closely related to those of Europeans (0.826 and 0.174), rather than to Chinese (1.000 and 0.000) (http://www.ensembl.org/). Moreover, the C and T allele frequencies of Koreans with BMI greater than 30 (0.884 and 0.116) are even closer to those of Europeans (Table 1).
SDC3 contains a large extracellular domain, followed by a glycine-rich transmembrane domain and a very short cytoplasmic tail (19). SDC3 consists of 384 amino acids and the molecular mass of SDC3 is 39,636 Da. Amino acids from 1 to 326 comprise extracellular, 327 to 351 transmembrane, 352 to 384 cytoplasmic domains, and 56 to 244 Ser/Thr-rich (mucin-like) region (UniProt, http://www.ebi.uniprot.org/; SwissProt, http://au.expasy.org/). Because SDCs differ in their extracellular domains and function as mediators of cell-cell and cell-matrix interactions (20), genetic variations in the extracellular domain in SDCs, very highly likely, may influence the functions of SDCs. As expected, all three nonsynonymous SNPs investigated herein are located in the extracellular domain of SDC3. Haplotype analysis of three SDC3 SNPs was found to be strongly associated with obesity in Koreans (Table 3). The V150I (rs2491132) SNP especially lies in both Ser/Thr-rich (mucin-like) and low-complexity regions (146–180). V150I (rs2491132) was found to be most strongly associated, among three SNPs investigated in this study, with obesity in Koreans (Table 1).
In summary, we provided evidence of ethnic differences in three nonsynonymous SDC3 polymorphisms [T271I (rs2282440), D245N (rs4949184), and V150I (rs2491132)] in the Korean population. Our data in the present study that suggest that variations in SDC3 are significantly associated with obesity also provide strong support that SDC3 is involved in the regulation of energy balance. To our knowledge, this is the first report to demonstrate the ethnic differences in three SNPs of SDC3 and the strong association among three SNPs of SDC3 and obesity in the Korean population. Because no other association studies concerning the SDC3 have yet been reported, more studies to elucidate the possible role of SDC3 in the development of obesity will be needed.
|
Footnotes |
|---|
Disclosure statement: The authors have nothing to disclose.
First Published Online October 3, 2006
1 E.H. and M.-J.K. contributed equally to this work. 
Abbreviations: BMI, Body mass index; OR, odds ratio; SDC, syndecan; SNP, single nucleotide polymorphism.
Received September 19, 2005.
Accepted September 25, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
