| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Internal Medicine I, Faculty of Medicine (K.H., Y.K., T.Y., H.Y.), and Department of Health Science Center (H.T.), Oita University, Oita 879-5593, Japan; Instituto Nacional de Diabetes, Endocrinología y Nutrición (J.A.H.B.), Universidad Technologica de Santiago (M.C.LL.), Universidad Autónoma de Santo Domingo (L.I.B.V.), and Centro de Gastroenterologia (B.J.W.), Santo Domingo, Dominican Republic; and Graduate School of Health and Nutritional Sciences, Nakamura Gakuen University (T.S.), Fukuoka 814-0198, Japan
Address all correspondence and requests for reprints to: Toshiie Sakata, M.D., Ph.D., Faculty of Nutritional Sciences, Nakamura Gakuen University, 5-7-1 Befu, Jounan-Ku, Fukuoka, 814-0198, Japan. E-mail: sakata{at}cc.nakamura-u.ac.jp.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Insulin resistance is a key factor in the etiology of type 2 diabetes and other lifestyle-related diseases (8, 9, 10, 11). Genes involved in insulin resistance are therefore likely to affect susceptibility to these diseases. One such gene, PC-1, encodes plasma cell glycoprotein-1 which may play an essential role in insulin resistance (12, 13, 14, 15). PC-1 is expressed widely in insulin-sensitive tissues and organs. Its activity was found to be elevated in dermal fibroblasts from type 2 diabetics (16, 17), which also displayed reduced autophosphorylation of the insulin receptor ß-subunit (16). PC-1 appears to inhibit insulin receptor signaling through interaction with the 
-subunit (18). When PC-1 cDNA was transfected into cultured cells, and its product overexpressed, a decrease in insulin receptor tyrosine kinase activity was observed in several studies (19, 20, 21, 22), but not in others (23).
An allelic polymorphism in exon 4 of PC-1 has been identified and designated K121Q. The Q (glutamine) variant had a stronger association with insulin resistance than the wild-type K (lysine) allele in nonobese, nondiabetic Caucasians from Sicily (24). In Finland and Sweden, it was associated with surrogate measures of insulin resistance (25). In contrast, there was no association with insulin resistance in Danish and Spanish Caucasians (26, 27). Furthermore, association studies examining the involvement of the PC-1 polymorphism in human obesity and type 2 diabetes were mostly negative (24, 25, 26, 27). In vitro studies, however, showed that cultured fibroblasts from KQ subjects were less active in insulin receptor autophosphorylation than those from KK controls (24). Moreover, when human MCF-7 cells were transfected with cDNAs for the Q or K allele, the Q allele was more effective in reducing insulin signaling than the K allele (28). How can one account for the discrepancies between ethnic groups or the in vitro vs. the in vivo findings? One possible explanation is that the low frequency of the Q allele and the extremely rare incidence of the QQ genotype in most human populations has made it difficult to obtain statistically significant associations. In the present study we aimed to test this idea by analyzing the PC-1 polymorphism in the population of the Dominican Republic, in whom obesity and type 2 diabetes are unusually prevalent.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We enrolled a total of 118 unrelated subjects for a 75-g oral glucose tolerance test (OGTT). They comprised 47 males and 71 females and ranged from 20–59 yr (mean, 38.4 ± 9.0 yr). The subjects were employees of the Centro de Gastroenterologia and the Instituto Nacional de Diabetes, Endocrinología y Nutrición (INDEN) in the city of Santo Domingo, Dominican Republic. All subjects enrolled were healthy volunteers. We excluded anyone diagnosed as being obese [body mass index (BMI), 
30 kg/m2] or having impaired glucose tolerance, type 2 diabetes, and/or diseases associated with insulin resistance.
Seven hundred and fifty-five unrelated subjects were enrolled in a case-control study. They comprised 252 males and 503 females, ranging from 17–85 yr of age (mean, 47.4 ± 13.3 yr). They were employees or out-patients of the Centro de Gastroenterologia and INDEN and consisted of healthy nonobese controls (n = 275), obese nondiabetics (n = 122), nonobese diabetics (n = 55), and obese diabetics (n = 303). Subjects with a BMI of 30 kg/m2 or greater were classified as obese. Type 2 diabetes was diagnosed according to the criteria of the American Diabetes Association (Report of the Expert Committee) (29). The type 2 diabetic patients were treated with diet therapy alone (40.5%), oral hypoglycemic drugs (35.2%), or insulin (24.3%).
The investigation was approved by the human genome committee of Oita University, Faculty of Medicine, and the ethical committees of INDEN and Centro de Gastroenterologia. All ethical assessments were made in accordance with the principles of the Declaration of Helsinki II. Written informed consent, as approved by the human genome committee, was obtained from all subjects.
Clinical data collection
Demographic data (gender and age) and BMI were recorded for all participants. The fasting plasma glucose (PG) levels of patients in the case-control study were measured with an enzyme assay kit (Kanto Kagaku, Tokyo, Japan), and hemoglobin A1c was determined using the microcolumn method (Bio-Rad Laboratories, Hercules, CA) for all the subjects.
The subjects were prohibited from smoking, eating, and drinking (except for water) after 2100 h on the night preceding the 75-g OGTT and during the test period. The 2-h test commenced between 0800–1000 h with the subjects seated throughout. Blood samples were taken immediately before (0 min) and 60 and 120 min after oral loading with 75-g glucose (Glutol Orange, Sigma-Aldrich Corp., St. Louis, MO). Systolic and diastolic blood pressures were measured before the blood samples were taken. The serum levels of high density lipoprotein, low density lipoprotein, and total cholesterol as well as triglyceride were determined using enzyme assay kits (Wako Pure Pharmaceuticals, Kanto Kagaku, Tokyo, Japan). Serum insulin was measured using a microparticle enzyme immunoassay kit (Eiken Kagaku, Tokyo, Japan). The areas under the curves associated with the PG (AUC-glucose), immunoreactive insulin (IRI; AUC-insulin), homeostasis model assessment-insulin resistance (HOMA-R), and homeostasis model assessment-ß cell function (HOMA-ß) were calculated as previously described (30, 31, 32). The insulin sensitivity index composite (ISI-comp) was calculated for the composite whole body insulin sensitivity index during OGTT using the following equation:
 |
Genotyping
Genotyping was performed using genomic DNA isolated from human leukocytes. The PCR conditions, specific primers, and experimental conditions for genotyping were described previously (24). Briefly, a DNA fragment of exon 4 was amplified using a forward (5'-ctgtgttcactttggacatgttg-3') and a reverse (5'-gacgttggaagataccaggttg-3') oligonucleotide primer pair. PCR products were digested with the restriction enzyme AvaII and analyzed by 12% native PAGE for 2 h at 500 V. Gels were stained with silver nitrate. The presence of K alleles was indicated by an intact band of 238 bp, and that of Q alleles by doublets of 148 and 90 bp. All genotyping tests were performed blind and in duplicate for each individual.
Statistical analysis
Data were expressed as the mean ± SD. The normal distribution and homogeneity of variance were tested before further statistical analyses. The relationship of age and BMI with genotype was assessed by one-way ANOVA. Gender distributions were compared using the 
2 test. The effect of genotype on clinical parameters was assessed by one-way analysis of covariance (ANCOVA) using genotypes as a factor. Age, BMI, and gender were used as covariates, followed by post hoc tests using the Bonferroni correction. The skewed parameters, i.e. IRI, AUC-IRI, HOMA-R, HOMA-ß, and ISI-comp, were transformed to natural logarithms to remove skewness and were analyzed by ANCOVA. Allele frequencies were calculated by the gene-counting method and compared by the 2 test. The Hardy-Weinberg equilibrium was also assessed by the 2 test, and glucose and insulin profiles during OGTT were compared by two-way ANCOVA. As mentioned, IRI profiles were transformed to natural logarithms to remove skewness in the two-way ANCOVA. P 
0.05 was considered statistically significant. Odds ratios (ORs) and 95% confidence intervals (CIs) adjusted for age, gender, and BMI (in the case of diabetes determinants) were calculated by logistic regression analysis. To clarify whether the Q allele is a critical determinant of diabetes, the allele was also analyzed in the logistic regression model by adjusting for age, gender, and BMI. All statistical analyses were performed using StatView version 5.0 software (SAS Institute, Cary, NC). The population-attributable risk percentage (PAR%) (33) of the K121Q polymorphism for type 2 diabetes was estimated as follows: PAR% = 100(X - 1)/X, and X = (1 - f)2 + 2f(1 - f)r + r2f2, where f is the frequency of the at-risk Q allele, and r is the OR of the Q allele, comparing control subjects and type 2 diabetics and adjusting for age, sex, and BMI in the logistic model.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
together with their PC-1 genotypes. Gender, age, BMI, and blood pressure were comparable among the three genotypes. The subjects’ fasting parameters of cholesterol, triglyceride, glucose, and AUC-PG showed no correlation with genotype. In contrast, both KQ (P = 0.002) and QQ subjects (P = 0.01) had higher AUC-IRI than KK subjects. QQ subjects (P = 0.038), but not KQ subjects (P = 0.24), had higher fasting insulin levels (IRI) than KK subjects. HOMA-ß was higher and ISI-comp was lower among QQ, but not KQ, subjects compared with the KK participants (P = 0.028 in each), whereas HOMA-R did not vary significantly among QQ, KQ, and KK subjects.
|
, both KQ and QQ subjects showed an upward shift of their insulin profile during the OGTT compared with the KK subjects (KQ, P = 0.027; QQ, P = 0.031, by two-way ANCOVA adjusted for gender as a covariate). The profiles of KQ and QQ subjects were comparable. There was, in contrast, no difference in glucose profiles during the OGTT among the three genotypes (Fig. 1), indicating that the Q allele is the most effective in producing insulin resistance.
|
The overall frequencies of the PC-1 gene codon 121 genotypes were: KK, 21.6% (163 of 755); KQ, 48.3% (365 of 755); and QQ, 30.1% (227 of 755). The overall Q allele frequency was 54.2% (819 of 1510), indicating an exceptionally high incidence of the QQ genotype and the Q allele in the population. The observed genotypes were in Hardy-Weinberg equilibrium. The frequency of the PC-1 genotypes in the controls was: KK, 26.9%; KQ, 46.2%; and QQ, 26.9% (Table 2). There was no statistically significant difference among these genotypes with respect to any phenotypic property, although the KK and QQ genotypes in the obese and/or diabetics seemed to differ slightly from those in the controls. When the incidence of K121Q genotypes was compared in the dominant and recessive models, the Q allele carriers (the KQ and QQ subjects) in the former model were higher with respect to all four phenotypes (P = 0.050) than those in the controls, particularly in the case of obese diabetics (P = 0.026; OR = 1.56). However, this was not the case in the recessive model (Table 2). The obese diabetics (P = 0.043; OR = 1.27) and the nonobese diabetics (P = 0.037; OR = 1.56) had a higher Q allele incidence than the controls (Table 2).
|
), in contrast to age (P = 0.001) and female gender (P < 0.0001). However, in the nonobese subjects (Table 4) logistic regression analysis showed that the QQ genotype as well as age (P < 0.0001) and BMI (P < 0.0001) acted as independent determinants of diabetes. The coefficient of determination (r2= 0.213) indicated that the fitness of this model is 21.3%. According to the logistic regression model (adjusting for age, gender, and BMI), the Q allele was a significant determinant of diabetes (OR = 1.633; 95% CI = 1.023–2.607; P = 0.0399). The PAR% of the Q allele was 43.3% in nonobese type 2 diabetes when calculated from the Q allele frequency of 0.518 and an adjusted OR of 1.633.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). The contribution of environmental factors to the pathogenesis of obesity in the Dominican Republic may be greater than that of genetic factors. Obesity and the PC-1 gene Q allele may act synergistically to cause insulin resistance, as suggested previously (34). Members of the nonobese and nondiabetic control groups who had KQ or QQ genotypes had a higher frequency of insulin resistance (as confirmed by insulin and glucose profiles in the 75-g OGTT) than their KK controls, a finding described previously for KQ subjects (24, 25, 26). A single Q allele was therefore sufficient to increase insulin resistance. However, judging from the assessments of insulin sensitivity, i.e. fasting plasma insulin, AUC-insulin, and ISI-comp, the homozygous QQ genotype was more potent in inducing insulin resistance than the heterozygous KQ genotype. Fasting insulin and HOMA-R are known markers for insulin sensitivity, mainly in the liver (35). On the other hand, both ISI-comp and insulin profiles after glucose loading reflect whole body insulin sensitivity, principally in muscle. The present findings thus indicate that the Q allele is involved in insulin resistance not only in liver, but also in muscle and at the whole body level. A major locus relating to insulin resistance in nondiabetic Mexican Americans has been mapped to chromosome 6q, where PC-1 is located (36). However, the researchers did not conclude that the gene responsible was PC-1. Also, a cluster of single nucleotide polymorphisms in the 3'-untranslated region was reported to be associated with increased PC-1 expression and abnormalities related to insulin resistance (37). Preliminary findings in our laboratory indicate that the single nucleotide polymorphism in the 3'-untranslated region is not associated with insulin resistance in the Dominican Republic (Hamaguchi, K., unpublished observations). Therefore, we suggest that the K121Q polymorphism per se may be responsible for the PC-1-related insulin resistance.
An unusual genetic characteristic of the Dominican Republic population is the remarkably high incidence of the Q allele, at 54.2% compared with approximately 12% in most other populations studied to date. Furthermore, 30.1% of Dominicans were QQ homozygotes compared with less than 2% of other ethnic populations. The OR of the Q allele for the determination of diabetes was 1.633 in the nonobese subjects (after adjusting for age, gender, and BMI). When the same OR of 1.633 was applied to other populations with a lower Q allele prevalence (e.g. 12%), the power of detecting a significant difference (1 - ß) (38) for the determination of diabetes was 42%, and the type II error (ß) was 58%. This indicates the high likelihood of failure to detect a significant difference. Hence, the high prevalence of the Q allele was advantageous for implicating the PC-1 Q allele in type 2 diabetes among the Dominican Republic population.
The heterozygous KQ genotype in Caucasians has been reported to be accompanied by insulin resistance (24, 25), but not by type 2 diabetes (24, 25, 26). The present study confirmed this observation. The extremely low incidence of QQ homozygosity in previous study populations precluded assessments of the impact this genotype has on insulin resistance and type 2 diabetes. In the present study insulin resistance was similar in both the KQ heterozygotes and the QQ homozygotes, whereas only homozygosity was associated with a significant risk of type 2 diabetes. One possible explanation for this disagreement is that the progression of diabetes is not totally consistent with insulin resistance, as type 2 diabetes is not necessarily accompanied by insulin resistance. Alternatively, differences in the genetic implications between allele and genotype must be borne in mind. To assess whether the Q allele is a critical determinant of diabetes, an analysis was carried out according to possession of the Q allele. The result revealed that the Q allele was a significant determinant of diabetes (OR = 1.633; 95% CI = 1.023–2.607; P = 0.0399). In contrast, logistic regression analysis on diabetic determinants demonstrated that the OR of the KQ heterozygotes did not reach statistical significance. Considering the gene dosage effect, these findings indicate that the QQ homozygotes may be more effective than the KQ heterozygotes, although the Q allele per se plays a significant role as a determinant of diabetes.
The exceptionally high incidence of the Q allele, at 51.8% in nonobese subjects, may raise concern for this population in terms of the potential for high rates of type 2 diabetes. The PAR% was thus calculated as 43.3% in the Dominican Republic, but it declined to 13.6% if the OR was fixed to the same level and the frequency of the Q allele was lowered to 12%, as observed in most other populations. It is unlikely that the high incidence of the Q allele in the Dominican Republic population is due to their Hispanic background, because Spanish Caucasians were recently found to have an overall Q allele frequency of 0.14, and the QQ genotype was absent (27). An alternative, and perhaps more likely, possibility is that an African genetic background has provided the high Q allele incidence. This is supported by archeological evidence of human evolution. Linkage disequilibrium analysis of the human genome suggests that north Europeans have a Nigerian ancestry (39). In other words, Africans are genetically more divergent than Europeans and could therefore have a higher frequency of the variant allele. Also, Africans have contributed considerably to the genetic makeup of the present Dominican Republic population (4). However, the possibility that Caribbean aborigines contributed to the high Q allele incidence cannot be excluded, because the genetic contributions from Africans and Caribbean aborigines have undergone extensive mixing. A genotyping study is now underway to determine whether the incidence of the Q allele is also high among African Americans, Nigerian Africans, and Native Americans.
Another intriguing explanation for the extraordinary high incidence of the Q variant may be positive selection, as proposed in the thrifty genotype hypothesis (40). As PC-1 functions in energy metabolism, possession of the Q allele may have been historically advantageous in the environment of the Dominican Republic. Peripheral insulin resistance may preserve energy reserves during famine and also protect from hypoglycemia (41). Between the late 15th and 19th centuries, 11–13 million African slaves were transported from west Africa to Caribbean countries, including the Dominican Republic. Due to severe shortages of food and water, 10–20% or more did not survive the journey or their first year in the Caribbean (42, 43). Most of the slaves on Hispaniola were forced to work on sugar plantations or in gold mines. These harsh conditions are assumed to have produced a genetic bottleneck effect and may have contributed to the predominance of a thrifty genotype in this population.
In conclusion, the present study demonstrated the high PC-1 Q allele incidence in the Dominican Republic and provided a novel insight into the PC-1 gene as a genetic background for type 2 diabetes. This study should allow further clarification of the role played by PC-1 in other lifestyle-related diseases associated with insulin resistance.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: ANCOVA, Analysis of covariance; AUC, area under the curve; BMI, body mass index; CI, confidence interval; HOMA-ß, homeostasis model assessment-ß cell function; HOMA-R, homeostasis model assessment-insulin resistance; IRI, immunoreactive insulin; ISI-comp, insulin sensitivity index composite; K, lysine; OGTT, oral glucose tolerance test; OR, odds ratio; PAR%, population-attributable risk percentage; PG, plasma glucose; Q, glutamine.
Received August 11, 2003.
Accepted December 4, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
gene expression, is associated with a reduction of both whole body insulin sensitivity and insulin receptor tyrosine-kinase activity. Diabetologia 40:282–289[CrossRef][Medline]
- subunit. Diabetes 49:13–19[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  L. S.-H. Wu, C.-H. Hsieh, D. Pei, Y.-J. Hung, S.-W. Kuo, and E. Lin Association and interaction analyses of genetic variants in ADIPOQ, ENPP1, GHSR, PPAR{gamma} and TCF7L2 genes for diabetic nephropathy in a Taiwanese population with type 2 diabetes Nephrol. Dial. Transplant., November 1, 2009; 24(11): 3360 - 3366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. F. Moore, K. A. Jablonski, C. C. Mason, J. B. McAteer, R. F. Arakaki, B. J. Goldstein, S. E. Kahn, A. E. Kitabchi, R. L. Hanson, W. C. Knowler, et al. The Association of ENPP1 K121Q with Diabetes Incidence Is Abolished by Lifestyle Modification in the Diabetes Prevention Program J. Clin. Endocrinol. Metab., February 1, 2009; 94(2): 449 - 455. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T Dwyer, L Blizzard, B Patterson, A-L Ponsonby, K Martin, S Quinn, M M Sale, S M Richards, R Morley, S Rich, et al. Association between birth weight and adolescent systolic blood pressure in a caucasian birth cohort differs according to skin type, CRH promoter or 11{beta}-HSD2 genotype Arch. Dis. Child., September 1, 2008; 93(9): 760 - 767. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. B. McAteer, S. Prudente, S. Bacci, H. N. Lyon, J. N. Hirschhorn, V. Trischitta, J. C. Florez, and for the ENPP1 Consortium The ENPP1 K121Q Polymorphism Is Associated With Type 2 Diabetes in European Populations: Evidence From an Updated Meta-Analysis in 42,042 Subjects Diabetes, April 1, 2008; 57(4): 1125 - 1130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Keene, J. C. Mychaleckyj, S. G. Smith, T. S. Leak, P. S. Perlegas, C. D. Langefeld, B. I. Freedman, S. S. Rich, D. W. Bowden, and M. M. Sale Association of the Distal Region of the Ectonucleotide Pyrophosphatase/Phosphodiesterase 1 Gene With Type 2 Diabetes in an African-American Population Enriched for Nephropathy Diabetes, April 1, 2008; 57(4): 1057 - 1062. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. D. Goldfine, B. A. Maddux, J. F. Youngren, G. Reaven, D. Accili, V. Trischitta, R. Vigneri, and L. Frittitta The Role of Membrane Glycoprotein Plasma Cell Antigen 1/Ectonucleotide Pyrophosphatase Phosphodiesterase 1 in the Pathogenesis of Insulin Resistance and Related Abnormalities Endocr. Rev., February 1, 2008; 29(1): 62 - 75. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Willer, L. L. Bonnycastle, K. N. Conneely, W. L. Duren, A. U. Jackson, L. J. Scott, N. Narisu, P. S. Chines, A. Skol, H. M. Stringham, et al. Screening of 134 Single Nucleotide Polymorphisms (SNPs) Previously Associated With Type 2 Diabetes Replicates Association With 12 SNPs in Nine Genes Diabetes, January 1, 2007; 56(1): 256 - 264. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bochenski, G. Placha, K. Wanic, M. Malecki, J. Sieradzki, J. H. Warram, and A. S. Krolewski New Polymorphism of ENPP1 (PC-1) Is Associated With Increased Risk of Type 2 Diabetes Among Obese Individuals Diabetes, September 1, 2006; 55(9): 2626 - 2630. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bacci, O. Ludovico, S. Prudente, Y.-Y. Zhang, R. Di Paola, D. Mangiacotti, A. Rauseo, D. Nolan, J. Duffy, G. Fini, et al. The K121Q Polymorphism of the ENPP1/PC-1 Gene Is Associated With Insulin Resistance/Atherogenic Phenotypes, Including Earlier Onset of Type 2 Diabetes and Myocardial Infarction Diabetes, October 1, 2005; 54(10): 3021 - 3025. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Abate, M. Chandalia, P. Satija, B. Adams-Huet, S. M. Grundy, S. Sandeep, V. Radha, R. Deepa, and V. Mohan ENPP1/PC-1 K121Q Polymorphism and Genetic Susceptibility to Type 2 Diabetes Diabetes, April 1, 2005; 54(4): 1207 - 1213. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
–