The correlation between the rs2241767 variant and the plasma AdipoQ concentration in type 2 diabetes mellitus

Abstract

Some single-nucleotide polymorphisms (SNPs) of the ADIPOQ gene have been linked to variations in adiponectin (AdipoQ) concentrations and the risk of Type 2 Diabetes Mellitus (T2DM). This study investigated the relationship between the rs2241767 ADIPOQ variant and the plasma AdipoQ concentration in a Vietnamese cohort. A case-control study was conducted on 100 T2DM patients and 100 healthy people. Genotyping of rs2241767 was performed using the polymerase chain reaction – restriction fragment length polymorphism (PCR-RFLP), validated via Sanger sequencing. The AdipoQ concentration was quantified using an enzyme-linked immunosorbent assay (ELISA). Overall, no statistical association was observed between individual rs2241767 genotypes and plasma AdipoQ concentration, age, gender, and lipid disorder (P > .05). However, in an exploratory subgroup analysis of the combined homozygous genotypes (AA + GG), the AdipoQ level was significantly lower in the T2DM group compared to the control group (P = .03). The AdipoQ level was positively correlated with weight, BMI, LDL-Cholesterol concentration, and Metrnl concentration. No correlations were found with age, height, systolic blood pressure, diastolic blood pressure, Total Cholesterol, HDL-cholesterol, triglycerides, Glucose, HbA1c, or Metabolic Score of Insulin Resistance (MET-IR). The ADIPOQ rs2241767 variant does not demonstrate a primary correlation with plasma AdipoQ levels in type 2 Diabetes Mellitus patients, though combined homozygous states show exploratory differences between cases and controls.

1. Introduction

Type 2 Diabetes Mellitus (T2DM) has emerged as one of the fastest-growing noncommunicable diseases worldwide. In 2024, an estimated 589 million adults aged 20–79 years (11.1% of the global population) were affected, with T2DM accounting for more than 90% of all diabetes cases.^[1] Insulin resistance, a silent and progressive process, underlies the development of T2DM. Adipose tissue, now recognized as an endocrine organ, secretes adipokines that regulate metabolic homeostasis.Among these, adiponectin is the most abundant plasma adipokine and is strongly associated with insulin sensitivity. Reduced adiponectin levels have been consistently observed in T2DM, contributing to a vicious cycle in which impaired adipose tissue function exacerbates insulin resistance, further lowering adiponectin concentrations.^[2] Genetic variants within the ADIPOQ gene locus significantly influence baseline adiponectin levels and systemic susceptibility to T2DM While several variants across the promoter and coding regions (such asrs266729, rs2241766, and rs1501299) have been widely investigated,^[3] the single-nucleotide polymorphism rs2241767, located at the boundary of exon 2 and intron 2, remains highly compelling.Previous studies examining the association between the rs2241767 polymorphism and T2DM have yielded conflicting results, with some reporting significant correlations^[4,5] while others found no clear relationship,^[6] highlighting the need for further investigation across diverse populations, especially unexamined populations, such as the Vietnamese cohort. This study investigates the association between rs2241767 genotypes and plasma adiponectin concentrations in individuals with T2DM.

2. Methods

A case-control study was conducted with 100 T2DM patients and 100 healthy controls. All subjects participated in the study when enrolling in the examination at 103 Military Hospital, Hanoi, Vietnam, from April 2022 to August 2024.

The inclusion criteria for T2DM patients (case group) were based on the guidelines for type 2 Diabetes Mellitus diagnosis by the American Diabetes Association 2022; diagnosed for the first time. The exclusion criteria were: type 1 DM, endocrinological disorder diseases such as thyroid disease; chronic kidney disease; chronic liver diseases; cancer; and type 2 DM with complications, anemia, hemoglobinopathies such as thalassemia, and B12 deficiency. The healthy group was chosen from healthy people of similar ages.

The AdipoQ plasma concentration was determined based on the sandwich ELISA (My Biosource Reagent Kit, USA, code MBS8244715, lot CK7E45B). The family history of Diabetes, history of dysregulated lipids, biochemistry test, height, weight, blood systolic, and diastolic data were collected at hospital admission. The biochemistry test, including fasting blood glucose, urea, creatinine, total cholesterol, HDL-cholesterol, LDL-cholesterol, triglycerides, and uric acid, was analyzed in the AU5800 analyzer (Beckman Coulter, USA). The HbA1c was determined by the high-performance liquid chromatography method on Premier 9200 (Trinitybiotech, USA). The BMI was calculated using the formula:

$$\begin{array}{c}{\displaystyle }\mathrm{B}\mathrm{M}\mathrm{I}=\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\left(\mathrm{k}\mathrm{i}\mathrm{l}\mathrm{o}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{m}\mathrm{e}\right)/\mathrm{H}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}{\mathrm{t}}^2\left(\mathrm{M}\mathrm{e}\mathrm{t}\mathrm{r}\mathrm{e}\right)\end{array}$$

The insulin resistance was estimated through the Metabolic Score for Insulin Resistance (METS-IR) index^[7]:

$${\displaystyle \mathrm{M}\mathrm{E}\mathrm{T}\mathrm{S}\text{-}\mathrm{I}\mathrm{R}=\frac{\begin{array}{c}{\displaystyle }\mathrm{l}\mathrm{n}\{\left(2\times \mathrm{f}\mathrm{a}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{g}\mathrm{l}\mathrm{u}\mathrm{c}\mathrm{o}\mathrm{s}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{l}(\mathrm{m}\mathrm{g}/\mathrm{d}\mathrm{l})\right)\\ & {\displaystyle }+\mathrm{T}\mathrm{r}\mathrm{i}\mathrm{g}\mathrm{l}\mathrm{y}\mathrm{c}\mathrm{e}\mathrm{r}\mathrm{i}\mathrm{d}(\mathrm{m}\mathrm{g}/\mathrm{d}\mathrm{l})\}\times \mathrm{B}\mathrm{M}\mathrm{I})\end{array}}{\mathrm{l}\mathrm{n}\left(\mathrm{H}\mathrm{D}\mathrm{L}\text{-}\mathrm{c}(\mathrm{m}\mathrm{g}/\mathrm{d}\mathrm{l})\right)}}$$

2.1. SNP genotyping

Each participant was collected 2 ml EDTA blood sample was collected to extract DNA from leukocytes using the QIAamp DNA Blood Kits – Genomic DNA Extraction (250) Cat no./ID.51106. The DNA purity was checked by determining the OD. The data about the SNP was based on NCBI data.

The single nucleotide polymorphism of ADIPOQ rs2241767 was genotyped by polymerase chain reaction – restriction fragment length polymorphism (PCR-RFLP) method. The PCR for determination of rs2241767 was pre-degenerated at 95°C for 3 mins, followed by 35 cycles of denaturation at 95°C for 35s, annealing at 58°C for 40s, elongation at 72°C for 45s, and final elongation at 72°C for 5mins. The PCR forward primer was TTATGTTAACCATAAGCAACCTTA. The PCR reverse primer was TAGTAACCACCAACAGAGCC. The PCR products were run on 1.5% agarose gel electrophoresis. The cleavage reaction was carried out at 37°C for 3 hours with enzyme BspTI (AflII) (Thermo Fisher Scientific, USA). The products of cleavage reactions were run on 3% agarose gel electrophoresis. In the electrophoresis images, 2 fragments of 316 and 23 bp products were shown with the AA genotype, three fragments of 339, 316 and 23 bp products were shown with the AG genotype, and one fragment of 339 bp product was shown with the GG genotype. 10% of total samples were randomly chosen for Sanger Sequencing to confirm the PCR-RFLP results (Fig. 1).

Figure 1.

The electrophoresis of PCR products after incubating with the restriction enzyme (A) and the Sanger sequencing results of some samples (B). (A) M: DNA ladder 100 basepair (bp); Some electrophoresis images of PCR products appeared with 2 bands approximately more than 300 bp; and some appeared with only one band. (B) A sample with homozygous genotype GG and a sample with heterozygous genotype of AG.

2.2. Statistics

Data were collected and entered into Microsoft Excel 2013 software. Statistical analysis was processed using JASP software 0.19.3 (University of Amsterdam, Neitherland). The parametric variables were shown as mean and standard deviation. The nonparametric variables were expressed as median and interquartile range. When comparing the mean values of 2 parametric variable groups, use a Student t test; for three or more groups, use an ANOVA test. When comparing two nonparametric variable groups, use the Mann-Whitney test; for three or more groups, use the Kruskal-Wallis test. When comparing proportions between 2 groups, use the Chi-square test. Use the Spearman test to evaluate the correlation between nonparametric variables and other continuous variables. Hypothesis testing is considered statistically significant with a P-value < .05.

3. Results

The characteristics of the subjects were shown in Table 1. There was no difference in age, height, plasma uric acid concentration, red blood cell count, and hemoglobin concentration between the 2 groups. In the case group, the proportion of men was higher than that of women. The weight index, BMI, systolic blood pressure, diastolic blood pressure, glucose concentration, urea, creatinine, total cholesterol, LDL-cholesterol, triglyceride, and HbA1c of the case group were higher than those of the control group with statistical significance (P < .05). The concentration of HDL-cholesterol in the case group was lower than that in the control group (P < .001).

Table 1.

The characteristics of the subjects.

	T2DM group (n = 100)	Healthy group (n = 100)	P
Age (year, Mean ± SD)	54.14 ± 11.28	53.45 ± 8.50	.60
Gender (n, %)
Female	40 (40)	63 (63)	.01
Male	60 (60)	37 (37)
Height (cm, Mean ± SD)	159.57 ± 7.52	159.02 ± 6.17	.56
Weight (kg, Mean ± SD)	60.62 ± 9.34	58.17 ± 8.18	.04
BMI (Mean ± SD)	23.76 ± 2.80	22.99 ± 2.93	.05
Systolic blood pressure (mm Hg, Mean ± SD)	132.62 ± 20.42	116.88 ± 6.08	<.001
Diastolic blood pressure (mm Hg, Mean ± SD)	77.79 ± 9.88	72.05 ± 8.07	<.001
Glucose (mmol/l, Mean ± SD)	11.72 ± 4.81	5.06 ± 0.39	<.001
Urea (mmol/l, Mean ± SD)	5.50 ± 1.39	4.98 ± 1.08	.003
Creatinine (µmol/l, Mean ± SD)	80.87 ± 16.27	76.22 ± 14.41	.02
Total cholesterol (mmol/l, Mean ± SD)	5.84 ± 1.54	5.02 ± 0.84	<.001
HDL-Cholesterol (mmol/l, Mean ± SD)	1.20 ± 0.49	1.31 ± 0.26	<.001
LDL-Cholesterol (mmol/l, Mean ± SD)	3.54 ± 0.90	3.11 ± 0.75	<.001
Triglycerid (mmol/l, Mean ± SD)	3.55 ± 3.34	1.71 ± 0.93	<.001
Acid uric (µmol/l, Mean ± SD)	323.85 ± 87.79	309.60 ± 82.67	.22
GOT (U/l, Mean ± SD)	28.38 ± 18.94	24.29 ± 7.62	.04
GPT (U/l, Mean ± SD)	38.37 ± 3.87	21.39 ± 8.78	<.001
Red blood cell (T/l, Mean ± SD)	4.66 ± 0.36	4.80 ± 0.52	.36
Hemoglobin (g/l, Mean ± SD)	139.92 ± 7.10	140.28 ± 12.96	.93
HbA1c (%,Mean ± SD)	9.37 ± 2.25	5.62 ± 0.29	<.001
Adipoq (ng/l, Median, IQR)	1.64 (2.05)	2.03 (3.00)	.08

Note: P were significant at < .05.

BMI = body mass index, GOT = Glutamate oxaloacetate Transaminase, GPT = Glutamate Pyruvate Transaminase, HDL = high density lipoprotein, IQR = interquartile range, LDL = low density lipoprotein, SD = standard deviation, T2DM = Type 2 diabetes mellitus.

When analyzing the rs2241767 gene polymorphism, the proportions of AA, AG, and GG genotypes were 45%, 49%, and 6% in the case group and 53%, 39%, and 8% in the control group, respectively. Genotype distributions in both groups were consistent with Hardy–Weinberg Equilibrium. There was no difference in AdipoQ concentration between the case and control groups in each genotype subgroup (containing AA, AG, GG, and AG + GG). With the AA + GG genotype, the Adipoq concentration in the control group was higher than in the disease group, with statistical significance at P < .05 (Table 2).

Table 2.

The genotype of rs2241767 and the AdipoQ plasma concentration.

Genotype	T2DM group (n = 100)		Healthy group (n = 100)		HWE	p
	n (%)	Median (IQR)	n (%)	Median (IQR)
AA	45 (45)	1.26 (1.53)	53 (53)	2.38 (2.88)	+	0.34
GG	6 (6)	1.23 (0.25)	8 (8)	3.91 (3.55)	+	0.10
AG	49 (49)	2.27 (0.60)	39 (39)	1.20 (2.77)	+	0.99
AA + GG	51 (51)	1.26 (1.45)	61 (61)	2.76 (3.17)		0.03
AG + GG	55 (55)	1.40 (3.14)	47 (47)	2.15 (2.39)		0.99

P values were significant at < 0.05; HWE = Hardy Weinberg Equilibrium law, IQR = interquartile range; +: conform to HWE expectation.

No association was found between the rs2241767 genotype and the characteristics of age group (< 50 and ≥ 50 years old), gender (male and female), BMI group (< 25 and ≥ 25 kg/m2), and lipid disorders, including HDL-Cholesterol group, LDL-Cholesterol group, and Triglyceride group (Table 3).

Table 3.

Stratified analysis of ADIPOQ gene rs2241767 genotypes with T2DM susceptibility.

Chracteristics		Total		AA		AG		GG		P
		n	%	n	%	n	%	n	%
Age group	< 50 years	64	100	35	54.69	27	42.19	2	3.12	.26
	≥ 50 years	136	100	63	46.23	61	44.85	12	8.82
Gender	Male	98	100	50	51.02	42	42.86	6	6.12	.81
	Female	102	100	48	47.06	46	45.01	8	7.84
BMI group	< 25 kg/m2	148	100	72	48.65	66	44.60	10	6.75	.95
	≥ 25 kg/m2	52	100	26	50.00	22	42.31	4	7.69
HDL-C group*	Group 1	116	100	58	50.00	52	44.83	6	5.17	.49
	Group 2	84	100	40	47.62	36	42.86	8	9.52
LDL-C group	≤ 2.6 mmol/l	39	100	21	53.85	16	41.03	2	5.12	.75
	> 2.6 mmol/l	161	100	77	47.83	72	47.72	12	7.45
Triglycerid group	≤ 2.3 mmol/l	89	100	38	42.60	44	49.44	7	7.86	.28
	> 2.3 mmol/l	111	100	60	54.05	44	39.64	7	6.31

BMI = body mass index, HDL-C = HDL-Cholesterol, LDL-C = LDL-Cholesterol.

^*Group 1 with HDL-C ≥ 1.29 mmol/l (male) and ≥ 1.0 mmol/l (female); Group 2 with HDL-C < 1.29 mmol/l (male) and < 1.0 mmol/l (female). P value was statistically significant at < .05.

To evaluate whether circulating plasma AdipoQ could function as an independent diagnostic biomarker for T2DM status within this clinical setting, a post hoc receiver operating characteristic (ROC) curve analysis was executed. The Area Under the Curve (AUC) for plasma AdipoQ levels was 0.58 (95% CI: 0.49–0.66, P = .08), indicating poor independent predictive accuracy, consistent with the nonsignificant overall difference observed between cases and controls (Fig. 2).

Figure 2.

The receiver operating characteristic curve for plasma AdipoQ levels in type 2 diabetes mellitus diagnostic.

Spearman’s rank corelation analysis revealed that plasma Adipoq concentration was significantly and positively correlated with weight, BMI, and LDL-Cholesterol concentration. Conversely, AdipoQ levels did not correlate with age, height, systolic blood pressure, diastolic blood pressure, Total Cholesterol, HDL-Cholesterol, triglycerides, Glucose, HbA1c, or MET-IR (Table 4).

Table 4.

The correlation between the AdipoQ concentration and some variables.

Variables	Age	Height	Weight	BMI
rho	0,17	0,07	0,23	0,20
95% CI	−0.01–0.26	−0.08–0.21	0.10–0.35	0.06–0.34
P	> .05	> .05	.001	.004

Variables	Glucose	HbA1c	Urea	Creatinine
rho	−0.13	−0.08	0.05	−0.01
95% CI	−0.27–0.01	−0.21–0.06	−0.18–0.10	−0.13–0.12
P	> .05	> .05	> .05	> .05

Variables	TC	HDL-C	LDL-C	Triglycerid
rho	0.10	−0.02	0.16	−0.04
95% CI	−0.04–0.23	−0.03–0.26	0.02–0.29	−0.18–0.10
P	> .05	> .05	.03	> .05

Variables	AST/GOT	ALT/GPT	GGT	Acid uric
rho	0.16	0.07	−0.07	< 0.001
95% CI	−0.02–0.27	−0.08–0.21	−0.16–0.14	−0.15–0.14
P	> .05	.05	> .05	> .05

Variables	MET-IR		SBP	DBP
rho	0.09		−0.11	−0.05
95% CI	−0.43–0.24		−0.24–0.03	−0.19–0.09
P	> .05		> .05	> .05

P value was statistically significant at < .05.

BMI = body mass index, CI = confidence of interval, DBP = diastolic blood pressure, HDL-C = HDL-cholesterol, LDL-C = LDL-cholesterol, MET-IR = metabolic score of insulin resistance, rho = Spearman coefficient, SBP = systolic blood pressure, TC = total cholesterol.

4. Discussion

Adiponectin has been demonstrated for its physiological effects and mechanism of action. Many studies have shown that Adiponectin deficiency is closely related to weight, insulin resistance, and an increased risk of type 2 diabetes.^[2,8–12] In animal models, Adiponectin has been shown to cause weight loss and independently increase insulin sensitivity.^[13–15]Adiponectin activates the PPARγ pathway, thereby increasing β-oxidation (the main lipid degradation pathway), causing weight loss and increasing insulin sensitivity.^[16] Through the binding of AdipoQ and its receptor (ADIPOR2), Adaptor protein 1 is activated. It stimulated the Akt pathway, causing increased translocation of GLUT-4 in the membrane, increasing glucose uptake into cells.^[17,18]AdipoQ also acts through the activity of AMPK (Adenosine monophosphate activated protein kinase), contributing to reducing free radicals and stress caused by oxidative radicals (which are one of the causes of decreased insulin sensitivity).^[19,20] Decreasing Adiponectin concentration is one of the risks of type 2 diabetes. Our study aimed to evaluate the influence of different genotypes at the rs2241767 locus on Adiponectin concentration in people with T2DM. Based on data from genome-wide scan studies, the ADIPOQ gene is a susceptibility locus associated with type 2 diabetes, obesity, and coronary heart disease.^[21–23] There are two notable disease-related blocks: the promoter region and the region located on the boundary between exon 2 and intron 2.^[24] rs2241767 is in the intron 2 region of the ADIPOQ gene. Data on noncoding gene regions have not been fully reported to date. However, polymorphisms at intron areas could affect protein binding, cause mRNA translational slippage, stability, DNA methylation, gene regulation, and anticodon pairing. These include effects through affecting splicing, transcription, translation, RNA processing and stability, and chromosomal interactions.

The role of rs2241767 is interesting in research across many different disease areas. The diseases of interest are more or less related to, or belong to, the group of metabolic diseases, such as nonalcoholic liver disease,^[25] polycystic ovary syndrome,^[26] diabetes,^[5,27] recurrent miscarriage,^[28] and cardiovascular disease.^[29,30] Most studies only investigated the relationship between genetic characteristics and the risk of diseases. The assessment of the association between genotype and Adiponectin concentration was still limited, while changes in Adiponectin concentration were one of the main factors causing the disease. The results from studies were still inconsistent. A study by Du et al included 1105 people with T2DM and 1107 healthy people, showing that the AG genotype increased the risk of type 2 diabetes in the overweight or obese group; the concentration of Adiponectin in the AG + GG group was lower than that of the AA group in the case group.^[5] Ramya et al examined 8 SNPs in 1100 normal glucose tolerant (NGT) individuals and 1100 T2DM. The results showed that rs2241767 was not related to the risk of T2DM. There was a statistically significant difference between the 3 genotype subgroups in the NGT group, the T2DM group, the obese group, and the non-obese group. However, the author did not mention the concentration difference between the NGT and T2DM. The GG genotype has the highest average concentration, followed by AG and AA.^[31] Zusi et al studied 2 SNPs with Adiponectin concentration on 794 subjects, showing that the group carrying the GG gene had a higher Adiponectin concentration than those carrying the AA + AG gene. And the group carrying at least one G allele had a lower Leptin concentration.^[32] In our Vietnamese cohort, individual rs2241767 genotypes showed no primary association with either diabetes status or plasma adiponectin concentrations. However, pooling the opposite homozygous states (AA + GG) revealed significantly lower adiponectin levels in T2DM cases versus controls (P = .03). This finding must be interpreted strictly as exploratory, given that testing multiple inheritance models without rigorous alpha-adjustments significantly elevates Type I error risks. Biologically, analyzing a combined homozygous model (AA + GG) focuses on assessing the presence of any homozygosity effects versus heterozygous states, but its post hoc nature limits definitive mechanistic claims.

A key limitation of this study is its sample size (100 cases, 100 controls), which limits our statistical power to identify subtle genetic effects or complex multi-locus gene-environment interactions. Circulating adiponectin concentrations are polygenic, modulated by multiple variants acting across high linkage disequilibrium (LD) blocks within the ADIPOQ locus.^[33,34] Focusing entirely on a single isolated SNP restricts the scope of our findings. Future investigations utilizing broad polygenic risk scores (PRS) and larger cohorts are required to fully define these relationships.

5. Conclusion

The ADIPOQ rs2241767 variant does not demonstrate a primary correlation with plasma AdipoQ levels in T2DM patients, though combined homozygous states show exploratory differences between cases and controls.

Author contributions

Formal analysis: Lan Thi-Phuong Dam, Thinh Huy Tran.

Funding acquisition: Lan Thi-Phuong Dam.

Investigation: Lan Thi-Phuong Dam, Thinh Huy Tran.

Resources: Lan Thi-Phuong Dam.

Software: Lan Thi-Phuong Dam.

Writing – original draft: Lan Thi-Phuong Dam.

Data curation: Thinh Huy Tran.

Project administration: Thinh Huy Tran.

Visualization: Thinh Huy Tran.

Conceptualization: Thuan Quang Huynh.

Methodology: Thuan Quang Huynh.

Supervision: Thuan Quang Huynh.

Writing—review & editing: Thuan Quang Huynh.