A Polymorphism in the Retinol Binding Protein 4 Gene is Not Associated with Gestational Diabetes Mellitus in Several Different Ethnic Groups

Abstract

Objective

Various Asian and Pacifific Islander groups have higher prevalence rates of type 2 diabetes and gestational diabetes. This increased incidence is likely to include genetic factors. Single nucleotide polymorphisms in the retinol binding protein 4 gene have been linked to the occurrence of type 2 diabetes. Hypothesizing a link between retinol binding protein 4 and gestational diabetes, we performed a candidate gene study to look for an association between an important retinol binding protein gene polymorphism (rs3758539) and gestational diabetes.

Study Design

Blood was collected from Caucasian, Asian, and Pacific Islander women diagnosed with gestational diabetes and from ethnically matched non-diabetic controls. DNA was extracted and real time PCR technology (TaqMan, Applied Biosystems) used to screen for the rs3758539 single nucleotide polymorphism located 5′ of exon 1 of the retinol binding protein 4 gene.

Results

Genotype and allele frequencies in the controls and gestational diabetes cases were tested using chi-square contingency tests. Genotype frequencies were in Hardy-Weinberg equilibrium. There was no association between the rs3758539 retinol binding protein 4 single nucleotide polymorphism and gestational diabetes in the Caucasian, Filipino, or Pacific Islander groups.

Conclusion

Interestingly, the rs3758539 retinol binding protein 4 single nucleotide polymorphism was not found to be associated with gestational diabetes. The absence of association suggests that gestational and type 2 diabetes may have more divergent molecular pathophysiology than previously suspected.

Introduction

Gestational diabetes mellitus (GDM) remains one of the most common clinical issues that obstetricians face. It has an overall prevalence in the United States ranging from 1–14%.¹ Mothers with GDM have an increased risk for hypertensive disorders of pregnancy and Cesarean delivery. Potential fetal complications include fetal macrosomia which may result in neonatal hyperbilirubinemia and hypoglycemia. Infants of diabetic mothers are also at increased risk of operative delivery, shoulder dystocia, and birth trauma.¹

The prevalence of both GDM and type 2 diabetes mellitus continue to increase in many racial/ethnic populations. The local prevalence of type 2 diabetes in Hawai‘i is increasing especially among Asian and Pacific Islander groups which include Native Hawaiians. These groups have demonstrated significantly higher rates of impaired glucose tolerance (15.5%) and type 2 diabetes (20.4%) compared to the overall US population.² In light of the fact that GDM prevalence rates tend to vary in direct proportion to the prevalence of type 2 diabetes in other populations, this increase in type 2 diabetes in Asians and Pacific Islanders may correlate with the increasing number of pregnancies complicated by GDM.^2,3 A recent report of the prevalence of GDM in Hawai‘i found that GDM rates were highest in Filipino women, followed by Chinese, Japanese, and Native Hawaiian/Pacific Islander women. Hawai‘i Caucasians were found to have the lowest GDM rate of all ethnic groups.⁴ These differences in GDM incidences are likely multi-factorial and are likely to include genetic factors.

Retinol binding protein 4 (RBP4) is the only specific transport protein for retinol (vitamin A) in the circulation. Circulating RBP4 is bound to transthyretin and binds to retinol to prevent its loss through the kidneys.⁵ Until recently this has been its only known function, however RBP4 has been since shown to be one of many proteins produced in adipose tissue. These adipokines signal changes of adipose tissue energy status to other organs and consequently either enhance (eg, leptin) or impair (eg, tumor necrosis factor alpha and resistin) insulin secretion and action.^6–8 Increased RBP4 plasma levels have recently been shown to be associated with both pregestational diabetes^8,9 and gestational diabetes.¹⁰ As illustrated in Figure 1, studies involving transgenic rodent models have suggested that RBP4 alters insulin sensitivity by both decreasing glucose uptake by muscle and increasing liver gluconeogenesis.⁸

Figure 1.

Proposed Role of Retinol Binding Protein in the Pathogenesis of Gestational Diabetes

The gene responsible for RBP4 is located on chromosome 10q and has been linked to an increased risk for type 2 diabetes in Caucasians¹¹ and Mexican Americans.¹² As displayed in Figure 2, specific single nucleotide polymorphisms (SNPs) on the RBP4 gene have since been shown to be both linked to serum RBP4 levels and type 2 diabetes.¹³ This link between RBP4 and type 2 diabetes is intriguing since gestational diabetes has been strongly associated with the latter development of type 2 diabetes. The pregnant state, due to increased insulin resistance, is thought to unmask pre-clinical type 2 diabetes. In fact, more than half of the gestational diabetics will go on to develop type 2 diabetes mellitus.^14,15

Figure 2.

The Retional Binding Protein Gene

Consequently, in light of this potential link between RBP4 and GDM the team sought to determine whether a specific genetic polymorp,hism of the RBP4 gene was associated with GDM in Filipino and Pacific Islander groups in Hawai‘i in addition to a Caucasian group from Utah.

Materials and Methods

Study Design

A case-control gene association study was performed comparing genotype frequencies of a single nucleotide polymorphism (SNP) mapping to the retinol binding protein 4gene (rs3758539, -803 C>T) in healthy pregnant women and pregnant women with GDM.

Subjects

Ethical approval for the study was obtained from the Institutional Review Board (IRB) of the University of Hawai‘i and Intermountain Healthcare in Utah. All participants provided written informed consent prior to providing any information or samples.

All pregnant women presenting to Labor and Delivery at Kapi‘olani Medical Center for Women and Children in Hawai‘i and Intermountain Healthcare in Utah were routinely screened for GDM at 24–28 weeks of gestation. Gestational age was determined by early ultrasound examination or mid trimester ultrasound consistent with the last menstrual period.

Blood was collected from women who met the Coustan-Carpenter criteria for GDM and ethnically matched non-diabetic controls. More specifically, screening tests were performed at 24 to 28 weeks of gestation with a 1 hour 50g oral glucose challenge test. Individuals with venous glucose levels greater than 140 mg/dl underwent a 3-hour 100g oral glucose tolerance test in the morning after a 12 hour fast. Women who met or exceeded at least 2 of the following venous blood glucose levels were diagnosed as having GDM: fasting 90 mg/dl, 1-hour 180 mg/dl, 2-hour 155 mg/dl, and 3-hour 140 mg/dl.¹⁴

The GDM cohort consisted of 88 Utah Caucasian, 82 Hawai‘i Filipino, and 19 Hawai‘i Pacific Islander women. The control cohort consisted of 315 Utah Caucasian, 286 Hawai‘i Filipino, and 32 Hawai‘i Pacific Islander women with normal glucose tolerance. Information regarding phenotypic characteristics such as age, BMI and glucose levels were not accessible.

Genotyping

Genomic DNA was extracted from whole blood using the Autopure DNA isolation system (Gentra Systems, Minneapolis, MN) following the manufacturers protocol. The SNP was genotyped using the TaqMan 5′-exonuclease SNP allelic discrimination assay by means of an ABI 7900 HT thermocycler (Applied Biosystems, Foster City, CA). Negative controls were included across the plates to ensure accuracy of genotyping. Genotyping errors were excluded by duplicate genotyping. Call rates for the SNP exceeded 97%.

The polymerase chain reaction (PCR) was carried out in a total reaction volume of 5 µl containing 10 ng of DNA and using the following amplification protocol: denaturation at 95°C for 10 minutes, followed by 50 cycles of denaturation at 92°C for 15 seconds, and annealing and extension at 60°C for 60 seconds. Post-PCR, the genotype of each sample was automatically attributed by measuring the allele-specific fluorescence in the ABI Prism 7900 HT Sequence Detection Systems, using the SDS 2.3 software for allele discrimination (Applied Biosystems).

Statistical Analysis

Allele and genotype frequencies were counted manually and the distributions were assessed for deviations from the Hardy-Weinberg equilibrium with the χ² test for goodness of fit. Genotype and allele frequencies were compared by contingency table analysis using χ² tests (http://statpages.org/ctab2x2.html). P<0.05 via the Pearson uncorrected method were considered significant. The Fisher's Exact test was utilized to calculate significance for the Pacific Islander group due to the small sample size. Descriptive data are expressed as mean value +/− SD.

A power analysis was performed to determine the number of cases and controls required utilizing a genetics power calculator.¹⁵ As stated before, GDM rates vary from 1–14% depending on the population studied.¹ Previous studies have noted the rs3758539 minor allele frequency to be 0.18 in Caucasians¹⁶ and 0.12 in Asians (Mongolians).¹³ Assuming a 10% GDM prevalence rate and a control : case ratio of 3.5, 267 controls/ 76 cases in the Caucasian cohort and 282 controls/ 80 cases in the Filipino cohort were necessary to achieve 80% power (B) at the P=0.05 significance level of detecting a genotypic relative risk of 2.0.

Results

The team genotyped a target SNP rs3758539 located 5′ of exon 1 of the RBP4 gene. Genotype and allele frequencies in the controls and GDM cases were tested using chi-square contingency tests. Genotype frequencies were in Hardy-Weinberg equilibrium.

In light of the power calculations, the study achieved sufficient sample size to adequately power our Caucasian and Filipino cohorts. Tables 1, 2, and 3 display the genotypic and minor allele frequencies for all three study groups. These frequencies were consistent with previously reported data in the Caucasian cohort (17.4% cases, 15.3% controls). Minor allele frequencies in the Filipino cohort were 13.4% in the case group and 11.4% in the control group. This appears to be the first reported data regarding rs3758539 in Filipinos.

Table 1.

SNP rs3758539 and Caucasian Allele/Genotypic Frequencies

		Caucasian
		GDM (n=88)	Controls (n=315)	P value
Genotype	CC	63.6%	72.4%	0.697
	TT	1.1%	3.2%	0.115
	CT	35.2%	24.4%	0.469
Allele	T	17.4%	15.3%	0.489

DNA nucleic acids: C=cytosine, T=thymine, A=adenine, G=guanine, CC=cytosine-cytosine, TT=thymine-thymine, CT=cytosine-thymine

Table 2.

SNP rs3758539 and Filipino Allele/Genotypic Frequencies

		Filipino
		GDM (n=82)	Controls (n=286)	P value
Genotype	CC	76.8%	79.0%	0.292
	TT	3.7%	1.7%	0.670
	CT	19.5%	19.2%	0.296
Allele	T	13.4%	11.4%	0.473

DNA nucleic acids: C=cytosine, T=thymine, A=adenine, G=guanine, CC=cytosine-cytosine, TT=thymine-thymine, CT=cytosine-thymine

Table 3.

SNP rs3758539 and Pacific Islander Allele/Genotypic Frequencies

		Pacific Islander
		GDM (n=19)	Controls (n=32)	P value
Genotype	CC	94.7%	71.9%	-
	TT	0%	0%	-
	CT	5.3%	28.1%	0.061
Allele	T	3%	14%	0.165

DNA nucleic acids: C=cytosine, T=thymine, A=adenine, G=guanine, CC=cytosine-cytosine, TT=thymine-thymine, CT=cytosine-thymine

There was no significant allelic or genotypic association between rs3758539 and GDM in the Caucasian, Filipino, or Pacific Islander cohorts. Although the Filipino GDM cohort displayed an increased minor allele homozygous genotype frequency (3.7% vs 1.7%), the association was not found to be significant (P values >0.05).

Discussion

Recently data has supported the role of various adipokines such as visfatin, adiponectin, resistin, and RBP4 in type 2 diabetes.⁶ RBP4 has specifically been associated with insulin resistance, obesity, type 2 diabetes, and GDM.^8,10,17–20

Since type 2 diabetes and GDM are believed to share similar pathophysiology, this study sought to determine whether a correlation existed between a specific SNP of the RBP4 gene and GDM. As mentioned previously, the team chose to examine rs3758539 in the study population. This particular SNP has already been linked to type 2 diabetes in the Mongolian population. The minor allele was significantly associated with type 2 diabetes and the minor allele homozygotes were reported to be associated with higher levels of serum RBP4 in non-pregnant diabetics. The authors of that study determined that the SNP could modify gene transcription efficiency by affecting the binding of transcription factors in vitro.¹³ In addition, a Chinese study found other non-coding RBP4 SNPs (+2333 G>A, +5388 C>T, +8201 T>A, and +8204 T>A) in linkage disequilibrium with rs3758539 to be significantly associated with circulating RBP4 levels in their population. These SNPs are displayed in Figure 2. The authors concluded that these SNPs may be genetic markers and that SNP rs3758539 played a functional role.²¹

Despite these findings, results did not show a significant association between our target SNP and GDM in any of the current study groups. There are possible methodological explanations for these results. First, due to insufficient numbers, the team pooled all Pacific Islanders (Native Hawaiian, Samoan, Micronesian, Tongan, etc.) into a single group. However, the pooled Pacific Islander group still did not reach a sufficient sample size to detect a significant association. Ideally, with a larger sample size, each Pacific Islander ethnicity could be evaluated separately. Second, since this database was not specifically designed to study diabetes, this study had incomplete data regarding other possible covariates such as body mass index and body fat distribution.

However, despite these limitations, this study appears to be the first published report of rs3758539 allele frequencies in a Filipino population. This population is at increased risk for GDM, as is evident by their increased prevalence rates relative to other ethnic populations.⁴ Establishing this baseline allele frequency rate may be helpful in designing future genetic research involving RBP4 in this population.

Interestingly, despite the apparent link between the RBP4 SNP rs3758539 and circulating RBP4 levels and insulin sensitivity, it was not found to be associated with GDM in the current study population. This study was adequately powered to detect an association in the Caucasian and Filipino groups. The absence of an association may suggest that GDM and type 2 diabetes may have more divergent molecular pathophysiology than previously suspected. However, larger studies are required to examine the association of RBP4 genetic variants in the other Asian and Pacific Islander populations.

Disclosure Statement

This project was supported by the National Center for Research Resources NIH grants R25 RR019321 “Clinical Research Education and Career Development (CRECD) in Minority Institutions,” and U54 RR014607 “Pacific Research Center for Early Human Development.” None of the authors identify any conflicts of interest.