Evaluation of the Effect of FOXO3 rs13217795 Genotype and Minor Allele (C) on Clinical Chemistry and Genetic Risk of Diabetes Among the Elderly Individuals from Northern India

Abstract

The forkhead box O family (FOXO) is expressed ubiquitously in a spatio-temporal manner and plays a key role in cellular metabolism, senescence, and aging. Genetic mutations in FOXO lead to metabolic diseases and cancer, and affect the longevity of individuals. Our study investigated how the genetic risk of type 2 diabetes mellitus (T2DM) altered due to an intronic variant rs13217795 of the longevity-associated FOXO3 gene in the geriatric population of North India. Genotypic characteristics of rs13217795 were determined among 347 age sex-matched (177 diabetic cases, 170 healthy controls) elderly individuals by TaqMan SNP assays after clinical assessment. Clinical chemistry and circulating cytokines level were assessed by biochemical and immunoassays. Genotype frequencies were not significantly (p = 0.526) different between cases and controls. The minor allele (C) frequency in diabetic cases and controls was 0.47 and 0.49, respectively (OR = 0.94, 95% CI = 0.69–1.26, p > 0.05). The minor allele was associated with lower fasting plasma glucose (FPG), fasting insulin, HOMA-IR, CRP, TNF-α, and IL-6 (p < 0.05). The homozygous minor allele carriers showed significantly lower levels of FPG, HOMA-IR, and TNF-α in T2DM patients. The minor allele (C) of intronic polymorphism in FOXO3 (rs13217795: T/C) confers the protective role characterized by its association with a decrease in glycemic and insulin resistance and proinflammatory markers.

Introduction

In developing countries, the term “healthy aging” might not be as effective as in developed countries because of limited health resources and relatively large needs. The elderly individuals are at higher risk of various illnesses, but sometimes they are neglected from their counterparts due to socioeconomic issues, environmental factors, and limited health resources. Type 2 diabetes mellitus (T2DM) is one such disease that most adversely affects the quality of life in the later span of life and will be prevalent among society with time. T2DM has a complex pathophysiology, and its onset and progression depend on the combined effect of genes, environmental factors, and aging. The incidence of diabetes increases with age in genetically susceptible individuals [Geesaman, 2006].

FOXO3 has been considered to play a significant role in the molecular basis of longevity by mediating key functions in various cellular and metabolic pathways including cell survival, senescence, and aging [Accili and Arden, 2004]. In recent findings, Lundell et al., [2019] showed that overexpression of FOXO3 dominant negative constructs decreased intramuscular glycogen content thus concluding that glucose storage is under the regulation of FOXO3 transcriptional activity which is involved in skeletal muscle metabolism [Lundell et al., 2019]. FOXO3 located in the small arm of chromosome 6 (6q21), and 41 single nucleotide polymorphisms (SNPs) associated with longevity were in the promoter region and many in intronic or near intron regions [Morris et al., 2015; Donlon et al., 2017]. According to the SNP database, variant rs13217795 is located in intron 2 (Chr 6:108652895) of the FOXO3 gene. The haploview plot shows the boundaries of the block of linkage disequilibrium represented by FOXO3 rs13217795 (Fig. 1). The intronic variants in FOXO3 (rs2802292, rs13217795) have been associated with longevity and healthy aging in American, German, Danish, and Chinese populations; however, the biological mechanism of these variants is still unclear [Willcox et al., 2008; Däumer et al., 2014; He et al., 2014]. FOXO3 plays a pivotal role in DNA damage, oxidative stress response, starvation, and caloric restriction which contributes to a healthy and increased lifespan [Sanese et al., 2019]. The variations in FOXO3 affects various pathways associated with energy metabolism, inflammation, immunity, cell cycle regulation, apoptosis and therefore affects longevity. The previous study showed that rs13217795 was associated with inflammatory diseases such as asthma and allergic rhinitis [Amarin et al., 2017]. However, the role of rs13217795 of the FOXO3 gene is not well established in T2DM. The earlier studies in the Chinese and south Indian Dravidian population reported variable outcomes of genetic analysis of this variant in T2DM [Nair et al., 2012; Mao et al., 2019].

Fig. 1.

a Haploview linkage disequilibrium (LD) plot of FOXO3 formed by using Guajarati people living in Houston (GIH) population data from 1000 genomes project showing selected SNP. LD blocks of FOXO3 (Chr 6: 108559835–108684774): Block 1 − Chr 6:108561930–108607472, Block 2 - Chr 6:108612454–108670483, Block 3 - Chr 6:108670603–108671443, Block 4 - Chr 6:108672958–108684385. Color scheme of LD plot: White (D' < 1, LOD <2). Blue (D' = 1, LOD <2). Shades of pink/red (D' < 1, LOD ≥2). Bright red (D' = 1, LOD ≥2). FOXO3 has 4 exons and 3 introns. Exon 1 - Chr 6:108559835–108560015, Intron 1 - Chr 6:108560016–108561085, Exon 2 - Chr 6:108561086–108561829, Intron 2 - Chr 6:108561830–108663454, Exon 3 - Chr 6:108663455–108664889, Intron 3 - Chr 6:108664890–108679826, Exon 4 - Chr 6:108679827–108684774. rs 13217795 (Chr 6:108652895) is located in LD block 2 and intron 2 of the FOXO3 gene. b Relative representation of LD blocks in relation to exons and introns of the FOXO3 gene. LD block 2 consists of the part of intron 2, whole exon 3, and part of intron 3. Chr, chromosome; LOD, logarithm of the odds.

Therefore, it seems to be the uncertain role of FOXO3 variants in determining the genetic risk of diabetes, particularly in the elderly population. We chose to establish the role of 1 of the intronic variants of FOXO3 (rs13217795) in the risk of T2DM in the elderly population. We have also investigated how the genotypes affect the various clinical determinants of diabetes and aging in determining the outcomes.

Materials and Methods

Study Design, Research Subjects and Ethical Aspects

A total of 347 individuals including 170 healthy controls and 177 T2DM cases coming from the northern part of India were enrolled for the study. The T2DM patients were screened for inclusion and exclusion criteria. The inclusion criteria for T2DM patients were age ≥60 years, diagnosis of T2DM according to International Diabetes Federation guidelines, 2017 [International Diabetes Federation, 2017], while patients having T1DM, hepatic, respiratory, autoimmune, and inflammatory disease, genetic disorder, cancer, on insulin therapy, and bacterial and viral infections were excluded from the study. The healthy controls were age ≥60 years and had no chronic physical, congenital, metabolic, cardiac, respiratory, inflammatory, or psychological illness and had fasting plasma glucose (FPG) <100 mg/dL and glycated hemoglobin (HbA1c) <6.5%.

Clinical Examinations and Laboratory Experiments

Five mL of the venous blood sample was withdrawn from each individual by venepuncture from the antecubital vein after overnight fasting. The blood was distributed into 3 parts: 1 mL in fluoride vial for FPG estimation, 2 mL in a plain vial for serum separation, and 2 mL in ethylene diamine tetra acetic acid (EDTA) vial for glycated hemoglobin (HbA1c) estimation and DNA isolation. The clinical chemistry was performed by using an auto-analyzer (Selectra pro-XL Clinical Chemistry System, The Netherlands) and related kits or as per standard protocols. The HbA1c was measured by “BIO-RAD” D-10 TM high-performance liquid chromatography (HPLC) and related kit. HOMA-IR (Homeostatic Model Assessment of Insulin Resistance) and HOMA-B (Homeostatic Model Assessment of beta-cell function) were calculated from the values of FPG and fasting insulin by the standard formulae. The serum level of insulin, C-reactive protein (CRP), tumour necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), cortisol, insulin-like growth factor-1 (IGF-1), interleukin-10 (IL-10), and adiponectin were measured by sandwich enzyme-linked immunosorbent assay (ELISA) by using a commercially available kit (insulin, CRP, IL-6 and 10, IGF-1, adiponectin (RayBiotech, USA), TNF-α (Invitrogen, USA), and cortisol (Abcam, USA), according to the manufacturer's protocol.

Genotype Analysis and Quality Control

1.0 mL of blood from the individuals was collected in EDTA vials and stored at −80°C until DNA extraction. The DNA was extracted from peripheral leukocytes using the modified salt precipitation method standardized in our laboratory [Miller et al., 1988; Yadav et al., 2016]. Each DNA sample was stored at −80°C until genotyping. We used real-time PCR-based TaqMan® SNP genotyping assays (TaqMan SNP assays C_9174543_10, Applied Biosystems, USA) for genotyping rs13217795. For quality control, positive and negative controls were used in each genotyping assay, and 5% of the samples were randomly selected and run in duplicates. The SNP had a 99.3% call rate. The cases and controls were blindly labeled during the genotyping analysis.

Statistical Analysis

The results for continuous variables are expressed as mean value ± standard deviation (SD); categorical variables are presented as number (%). The difference between the mean of groups is compared by unpaired Student's t-test or one-way ANOVA test and the χ² test for categorical variables. The Hardy-Weinberg equilibrium (HWE) for the SNP was evaluated by the χ² test goodness of fit test. The odds ratios and 95% confidence intervals were used to measure the relative degree of risk. The binary logistic regression analysis was used to calculate the odds ratio and confidence interval in genetic models. The interaction of alleles with various parameters was evaluated by linear regression analysis. All the statistical tests except HWE were performed using SPSS v. 25 (IBM SPSS Inc., USA). The HWE was calculated in excel. A 2-tailed p value < 0.05 was considered statistically significant for all tests.

Results

Demographic and Biochemical Parameters of Study Samples

The study comprised 177 individuals with T2DM and 170 non-diabetic individuals (control group). The results of the demographic and biochemical parameters are summarized in Table 1. The mean age and gender distribution were comparable between the groups. The T2DM group had significantly higher levels of BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), FPG, HbA1c, total cholesterol, triglyceride, low-density lipoprotein cholesterol (LDL-C), fasting insulin, and HOMA- IR compared to the control group. The serum levels of CRP, TNF-α, IL-6, and cortisol were significantly higher, whereas the IL-10, and adiponectin levels were significantly lower in the T2DM group than in the control. The mean IGF-1 level was lower in the T2DM group as compared to the control, but the difference was statistically nonsignificant.

Table 1.

Demographic and clinical characteristics of control and type 2 diabetes mellitus patients

Parameters	Control (n = 170)	Case (n = 177)	p value
Age, years	70.25±5.25	71.23±6.45	0.125
Gender, male/female, %	55.29/44.71	58.19/41.81	0.586
Smoking, yes/no, %	19.41/80.59	26.00/74.00	0.144
Drinking, yes/no, %	12.94/87.06	14.12/85.88	0.747
Age at diagnosis, years	−	55.99±6.97	−
Duration of DM, years	−	15.24±7.67	−
BMI, kg/m2	23.66±2.87	26.96±4.41	<0.001
Systolic BP, mm Hg	132.24±15.36	144.9±22.99	<0.001
Diastolic BP, mm Hg	82.78±8.07	88.12±13.74	<0.001
FPG, mg/dL	94.80±5.64	174.62±42.85	<0.001
HbAlc, %	5.24±0.57	8.43±1.94	<0.001
Total cholestrol, mg/dL	184.45±30.26	193.48±39.17	0.017
Triglyceride, mg/dL	137.58±56.47	183.52±65.34	<0.001
HDL-C, mg/dL	52.52±8.75	43.36±10.79	<0.001
LDL-C, mg/dL	104.41±30.18	113.41±41.72	0.022
Fasting insulin, µIU/mL	14.74±8.59	25.52±11.57	<0.001
HOMA-IR	3.48±2.04	11.33±7.04	<0.001
HOMA-B, %	171.19±107.84	88.42±43.43	<0.001
CRP, µg/mL	2.63±1.16	7.00±3.59	<0.001
TNF-α, pg/mL	15.38±7.22	30.84±12.50	<0.001
IL-6, pg/mL	4.22±2.20	9.77±5.60	<0.001
Cortisol, ng/mL	105.78±59.02	147.68±62.89	<0.001
IGF-1, ng/mL	173.38±63.01	162.08±67.31	0.108
IL-10, pg/mL	8.07±4.15	5.53±3.02	<0.001
Adiponectin, µg/mL	9.28±4.69	5.85±3.4	<0.001

The results are represented as mean ± SD (compared by unpaired Student's t-test) or percentage (compared by x² test). p < 0.05 considered significant in bold. BMI, body mass index; BP, blood pressure; CRP, C-reactive protein; FPG, fasting plasma glucose; HbA1c, hemoglobin A1C; HDL-C, high-density lipoprotein cholesterol; HOMA-IR, homeostasis model assessment of insulin resistance; HOMA-B, homeostasis model assessment of β-cell function; IGF-1, insulin-like growth factor 1; IL-6, interleukin 6; IL-10, interleukin 10; LDL-C, low-density lipoprotein cholesterol; TNF-α, tumor necrosis factor-alpha.

Association of Genotype, Allele Frequency to the Susceptibility of T2DM

The genotype frequencies were in agreement with the Hardy-Weinberg equilibrium in both the T2DM patients and the controls (p > 0.05). There were no significant differences in the genotype and allele frequencies for FOXO3 rs13217795 between the T2DM cases and controls (p > 0.05). The rs13217795 had a minor allele frequency of 0.49 and 0.47 in controls and T2DM patients, respectively. T and C were the major and minor alleles, respectively, in both the cases and controls. The association of FOXO3 SNP with T2DM risk revealed that rs13217795 was not associated with T2DM among the geriatrics of North India under any genetic model. After adjusting for age, gender, BMI, smoking, and drinking, the association remained statistically nonsignificant (Tables 2, 3).

Table 2.

Comparison of genotypic and allelic frequencies of rs13217795 between control and type 2 diabetes mellitus patients

	TT (%)	TC (%)	CC (%)	T (freq)	C (freq)	p valueb
Control	47 (27.65)	S0 (47.06)	43 (25.29)	174 (0.51)	166 (0.49)	0.447
Case	47 (26.55)	93(52.54)	37 (20.90)	1S7 (0.53)	167 (0.47)	0.471
OR (95%CI)	−			0.94 (0.69–1.26)
p valuea	0.526			0.664

^aChi-square test for genotype and allele distributions between cases and controls.

^bChi-square test for Hardy-Weinberg equilibrium in cases and controls. freq, frequency.

Table 3.

FOXO3 rs13217795 and susceptibility to type 2 diabetes mellitus under different genetic models

Genetic model		OR (95%CI)a	p valuea	OR (95%CI)b	p valueb
Codominant 1	TT versus TC	1.16 (0.70–1.92)	0.557	1.02 (0.57–1.80)	0.959
Codominant 2	TT versus CC	0.86(0.47–1.56)	0.622	1.03 (0.53–2.00)	0.941
Dominant	TT versus TC + CC	1.057 (0.66–1.70)	0.819	1.05 (0.62–1.79)	0.844
Recessive	TT + TC VS CC	0.78 (0.47–1.28)	0.332	1.00 (0.58–1.73)	0.997
Additive	—	0.94 (0.70–1.26)	0.664	1.02 (0.74–1.42)	0.904

Data were calculated by logistic regression analysis. Coding in genetic models: codominant 1 (TT = 0; TC = 1), codominant 2 (TT = 0; CC = 1), dominant 1 (TT = 0; TC = 1; CC = 1) recessive (TT = 0; TC = 0; CC = 1), additive model (TT = 0; TC = 1; CC = 2).

^aUnadjusted values.

^bAdjusted for age, gender, BMI, smoking, and drinking.

Relationships between Genotypes and Tested Parameters in T2DM Patients

A one-way ANOVA was conducted to establish their associations with the demographic and biochemical parameters between genotypes in T2DM patients (Table 4). There were no statistically significant differences between age, age at diagnosis of DM, duration of DM, BMI, SBP, DBP, and lipid profile in genotype groups (p > 0.05). Among the glycemic and insulin resistance parameters, the FPG and HOMA-IR were significantly different between genotypes (p = 0.032 and p = 0.011, respectively) while there was no significant difference for HbA1c, fasting insulin level, and HOMA-B (p > 0.05). Post hoc analyses using the Tukey post hoc criterion for significance indicated that the mean FPG and HOMA-IR were significantly lower in the CC genotype than in the TT genotype(p = 0.033 and p = 0.008, respectively); however, there was no significant difference for CC versus TC and TC versus TT (p > 0.05). On comparing proinflammatory (IGF-1, IL-10 and adiponectin) and anti-inflammatory markers (CRP, TNF-α, IL-6 and cortisol), there was only a significant difference for TNF-α (p = 0.012). On post hoc analysis, TNF-α was significantly lower in the CC genotype as compared to TC and TT genotype (p = 0.044 and p = 0.011, respectively), while there was no significant difference for TC versus TT (p > 0.05).

Table 4.

Association of clinical and laboratorial characteristics with genotypes in type 2 diabetes mellitus patients

Parameters	TT (n = 47)	TC (n = 93)	CC (n = 37)	p value
Age, years	71.47±7.65	71.54±5.75	70.14±6.50	0.513
Gender, male/female, %	59.57/40.43	56.99/43.01	59.46/40.54	0.943
Smoking, yes/no, %	31.91/68.09	25.81/74.19	18.92/81.08	0.402
Drinking, yes/no, %	12.77/87.23	15.05/84.95	13.51/86.49	0.928
Age at diagnosis, years	55.60±7.26	55.75±6.39	57.08±8.00	0.561
Duration of DM, years	15.87±7.80	15.78±7.22	13.05±8.38	0.150
BMI, kg/m2	27.12±4.10	27.41±4.57	25.64±4.21	0.114
Systolic BP, mm Hg	147.57±23.44	144.47±24.10	142.59±19.58	0.597
Diastolic BP, mm Hg	88.13±14.09	88.69±14.82	86.70±10.24	0.761
FPG, mg/dL	182.30±51.27	176.99±43.12	158.86±22.74	0.032
HbA1c, %	8.25±1.74	8.68±2.09	8.04±1.73	0.184
Total cholestrol, mg/dL	185.80±36.46	195.98±37.86	196.96±45.08	0.290
Triglyceride, mg/dL	199.47±71.21	178.05±59.74	177.03±69.41	0.149
HDL-C, mg/dL	41.60±11.62	43.84±10.41	44.40±10.61	0.414
LDL-C, mg/dL	104.29±37.48	116.52±41.39	117.16±46.79	0.217
Fasting insulin, µIU/mL	28.62±12.09	24.91±11.44	23.14±10.68	0.074
HOMA-IR	13.63±9.23	11.08±6.31	9.06±4.35	0.011
HOMA-B, %	92.51±38.91	85.36±45.21	90.93±44.85	0.609
CRP, µg/mL	7.82±3.7	6.86±3.62	6.29±3.26	0.133
TNF-α, pg/mL	33.60±13.01	31.48±11.70	25.72±12.63	0.012
IL-6, pg/mL	11.41±7.35	9.29±4.75	8.87±4.69	0.058
Cortisol, ng/mL	142.05±67.54	153.12±65.60	141.20±48.53	0.483
IGF-1, ng/mL	164.26±66.10	165.24±71.13	151.38±59.04	0.554
IL-10, pg/mL	5.13±3.60	5.77±3.08	5.42±1.88	0.482
Adiponectin, µg/mL	6.20±3.76	5.63±3.11	5.96±3.66	0.626

The results are represented as mean ± SD (compared by ANOVA) or percentage (compared by x² test). p < 0.05 considered significant in bold.

Effect of Minor Allele on Tested Parameters

We analyzed the association between alleles and quantitative parameters in this study (Table 5). The minor allele (C) of rs13217795 was associated with the decreases in FPG, fasting insulin, HOMA-IR, CRP, TNF-α, and IL-6 (p < 0.05). The FPG, fasting insulin, HOMA-IR, CRP, TNF-α, and IL-6 were decreased per minor allele by 0.29 × 10⁻² mg/dL, 0.01 μIU/mL, 0.02, 0.03 μg/mL, 0.01 pg/mL and 0.02 pg/mL, respectively. The minor allele did not show any association between age at diagnosis of T2DM, duration of T2DM, BMI, SBP, and DBP. Also for cortisol, IGF-1, IL-10, and adiponectin, no significant association was found.

Table 5.

Associations between the minor allele and clinical and laboratorial characteristics in type 2 diabetes mellitus patients

Parameters	B (95% CI)	β	t	p value
Age at diagnosis, years	0.7 × 10−2 (−0.8 × 10−2, 0.02)	0.07	0.93	0.356
Duration of DM, years	−0.01(−0.02, 0.3 × 10−2)	−0.12	−1.59	0.114
BMI, kg/m2	−0.02 (−0.04, 0.7 × 10−2)	−0.11	−1.41	0.161
Systolic BP, mm Hg	−0.20 × 10−2 (−0.70 × 10−2, 0.20 × 10−2)	−0.08	−1.00	0.317
Diastolic BP, mm Hg	−0.16 × 10−2 (−0.91 × 10−2, 0.59 × 10−2)	−0.03	−0.42	0.675
FPG, mg/dL	−0.29 × 10−2 (−0.52 × 10−2, −0.10 × 10−2)	−0.18	−2.45	0.015
HbA1c, %	−0.01 (−0.06, 0.04)	−0.03	−0.33	0.744
Total cholestrol, mg/dL	0.18 × 10−3 (−0.07 × 10−2, 0.40 × 10−2)	0.10	1.37	0.171
Triglyceride, mg/dL	−0.13 × 10−2 (−0.28 × 10−2, −0.02 × 10−2)	−0.13	−1.67	0.097
HDL-C, mg/dL	0.58 × 10−2 (0.35 × 10−2, 0.02)	0.09	1.23	0.221
LDL-C, mg/dL	0.19 × 10−2 (−0.05 × 10−2, 0.43 × 10−2)	0.11	1.49	0.137
Fasting insulin, µIU/mL	−0.01 (−0.02, −0.12 × 10−2)	−0.17	−2.24	0.027
HOMA-IR	−0.02 (−0.04, −0.01)	−0.23	−3.06	0.003
HOMA-B, %	−0.03 × 10−2 (−0.27 × 10−2, 0.21 × 10−2)	−0.02	−0.25	0.803
CRP, µg/mL	−0.03 (−0.06, −0.03 × 10−2)	−0.15	−1.99	0.048
TNF-α, pg/mL	−0.01 (−0.02, −0.36 × 10−2)	−0.21	−2.85	0.005
IL-6, pg/mL	−0.02 (−0.04, −0.19 × 10−2)	−0.16	−2.18	0.030
Cortisol, ng/mL	0.36 × 10−4 (−0.16 × 10−2, 0.17 × 10−2)	−0.33 × 10−2	−0.04	0.966
IGF-1, ng/mL	0.06 × 10−2 (−0.21 × 10−2, 0.90 × 10−2)	−0.06	0.81	0.419
IL-10, pg/mL	0.01 (−0.03, 0.04)	0.04	0.53	0.596
Adiponectin, µg/mL	0.61 × 10−2 (−0.04, 0.02)	0.03	0.40	0.692

Data were calculated by linear regression analysis. Additive model (TT = 0; TC = 1; CC = 2) was used to see the effect of the minor allele on various parameters. B, unstandardized regression coefficient; β (beta coefficient), standardized regression coefficients. p < 0.05 considered significant in bold.

Discussion

Aging is associated with progressive changes in carbohydrate metabolism, insulin resistance, and inflammation. Chronic inflammation is considered to be one of the potential contributors of age-associated illness and functional decline [Franceschi and Campisi, 2014]. There are various interlinked mechanisms including inflammation, oxidative stress, etc. that play a pivotal role in the pathophysiology of diabetes and aging [Mendoza-Núñez et al., 2011]. Diabetes and aging are linked at the molecular level; even diabetes is considered as an accelerated model of aging [Morley, 2008]. FOXO3 transcriptional factor is the downward target of insulin, IGF signaling, adiponectin, IL-6 and 10, and modulates age-associated diseases such as cancer, osteoporosis, and T2DM [van der Horst and Burgering, 2007].

We hypothesized that the longevity associated with the FOXO3 variant rs13217795 may modify the predisposition to T2DM in our target geriatric population. A study by Mao et al. [2019] reported that rs13217795 was associated with the risk of T2DM in the Chinese elderly [Mao et al., 2019], whereas another study reported no association of rs13217795 with T2DM in the middle-aged south Indian population [Nair et al., 2012]. Our results showed that rs13217795 was not associated with the risk of T2DM in the north Indian elderly individuals. However, variable data are available on the role of FOXO3 in the pathogenesis of T2DM. We analyzed the association of genotypes with clinical parameters of patients and major biomarkers of inflammation and aging. In our study, a significant association was observed between rs13217795 genotypes with FPG, HOMA IR, and TNF-α. Following previous studies on south Indian and Chinese patients with T2DM, we did not observe any association of rs13217795 with BMI and blood pressure [Nair et al., 2012; Mao et al., 2019]. A study by Kim et al. [2006] on FOXO3 promoter region SNP reported the association of genotypes with BMI in a Korean population.

The disruption of glucose metabolism is characteristic of aging [Geesaman, 2006]. FOXO3 transcription factor is the downward target of insulin and IGF signalling and is negatively regulated by it [Tissenbaum and Ruvkun, 1998]. The insulin through FOXOs regulates glycogenolysis and gluconeogenesis via modulating the expression of phosphoenolpyruvate carboxykinase, glucose 6-phosphatase, and pyruvate dehydrogenase kinase-4 genes [Nakae et al., 2001; Daitoku and Fukamizu, 2007]. In our study, the CC genotype was associated with low FPG and HOMA-IR. The minor allele (C) was associated with a decrease in FPG, fasting insulin, and HOMA-IR. However, we did not find any significant association between FOXO3 genotypes and HbA1c, fasting insulin level, and HOMA-B. The rs13217795 variant was found to be associated with low blood glucose in Chinese elderly diabetic women [Mao et al., 2019]; however, there was no significant difference found in the south Indian middle-aged population [Nair et al., 2012]. In a study by Sun et al. [2015], the longevity-associated FOXO3 variants were associated with lower levels of FPG, HbA1c, and HOMA-IR in long-lived individuals [Sun et al., 2015]. In the Danish twin cohort, minor allele of longevity-associated FOXO3 rs2802292 was found to be associated with low FPG, fasting insulin, and HOMA-IR [Banasik et al., 2011]. We did not find any difference in total cholesterol, triglyceride, HDL, and LDL levels between the genotype of selected SNP in T2DM patients.

Inflammation is a major risk factor in diabetes while the development and progression of diabetic complications are further enhanced with aging. FOXO3 is reported to be involved in the regulation of cytokine production, inhibition of NF-κB activation subdues T cell activation and proliferation, and lowering of inflammation [Lin et al., 2004]. FOXO3 regulates inflammatory cytokine production by dendritic cells and inhibits IL-6 and TNF-α production [Dejean et al., 2009]. Insulin, IGF-1, and adiponectin regulate FOXO3 by IGF-1/PI3K/AKT pathways through their respective receptors [van Heemst, 2010]. FOXO3 is the downward target of the insulin/IGF-1/phosphatidylinositol-3 kinase (PI3K)/AKT (Protein Kinase B) signal transduction pathway. TNF-α, IL-6, and CRP are involved in aging-associated inflammation and mortality [Giovannini et al., 2011]. In our study, while comparing the proinflammatory and anti-inflammatory markers between genotypes of T2DM patients, there was no significant difference for CRP, IL-6, cortisol, IGF-1, IL-10, and adiponectin, except for TNF-α. The TNF-α level was significantly lower in the CC genotype. In addition, minor allele was associated with a lower level of CRP, TNF-α, and IL-6. Previous reports suggested that FOXO3 modulates the TGFβ1-dependent pathway in monocytes that decreases the production of proinflammatory cytokines, TNF-α and IL-6, and enhances the production of the anti-inflammatory cytokine, IL-10 [Lee et al., 2013]. In a study by Willcox et al. [2017], the longevity-associated minor allele carrier of rs2802292 was associated with low plasma TNF-α levels in Japanese but not in Whites and Blacks [Willcox et al., 2017]. Insulin and IGF-1 signal via tyrosine kinase receptors regulate cellular growth, differentiation, glucose, and protein metabolism [Cai et al., 2017]. Depending on the metabolic state of the cell, FOXO3 regulates the anabolic-catabolic metabolism via catabolic hormones and insulin signaling. FOXO3 promoter region has 2 functional glucocorticoid responsive elements; thus, catabolic glucocorticoid hormone and intracellular AMP levels together are responsible for upregulation of FOXO3 transcriptional activity [Lützner et al., 2012]. In our study, we observed increased cortisol levels in the T2DM cases when compared to controls while its level was not associated with the genotype.

Due to the combined effect of genes, environment, and aging, the elders are at an increased risk of T2DM. The time of onset of age-related diseases is a major determinant of lifespan. In our study, we did not observe any association of rs13217795 with T2DM. The function of FOXO proteins on metabolism and inflammation is to coordinate the gene expression response to environmental stimuli [Morris et al., 2015]. In our study, genetic variation in FOXO3 may modulate inflammatory cytokine production; therefore, the presence of minor allele is associated with lower TNF-α and IL-6 levels. The selected intronic variants of FOXO3 exert a lesser effect which is not enough to cross the threshold of disease, or FOXO3 acts through a different mechanism in T2DM.

Conclusion

The FOXO3 variant rs13217795 was not related to susceptibility to T2DM in the geriatric population of northern India, but minor allele confers a significant reduction in the FPG, fasting insulin, HOMA-IR, CRP, TNF-α, and IL-6. Genetic architecture is one of the important factors in determining the functional output of a gene; therefore, the effect of intronic variant can further be understood by the inclusion of other SNPs in linkage disequilibrium with rs13217795. For a clinician, the approach to the treatment of the elderly patients is different from younger adults due to various comorbid conditions. Therefore, specific knowledge of such variants may be useful in selecting personalized treatment. Our target subjects belong to the northern part of India only; however, similar studies in other regions and ethnic groups might be useful to further establish the role of intronic variants of FOXO3 in T2DM.

Statement of Ethics

The protocols used in this study were in compliance with the Declaration of Helsinki and were approved by the Ethics Committee of the King George's Medical University Lucknow, India with reference No. 90th ECM II B-M.D.Ph.D./P2. All of the participants were informed about the study, and their consent was obtained prior to the enrolment.

Conflict of Interest Statement

The authors declare no conflicts of interest.

Funding Sources

To conduct the study, Indian Council of Medical Research (ICMR), New Delhi, provided financial support and a fellowship to S.H.

Author Contributions

S.H.: study design, experiments, and analysis; S.S.Y.: experiments, analysis and manuscript writing; M.B.: interpretation of data, and manuscript editing; K.U.: study concept, acquisition of data, and clinical supervision; S.K.: supervision of the study, and approval of the final version of the manuscript.

Data Availability Statement

All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.

S.H. and S.S.Y. share first authorship.