Association study between BGLAP RS1800247-polymorphic variant and type 2 diabetes mellitus development among hypertensive and non-hypertensive Ukrainians

Матеріали та методи. У дослідження залучили 153 пацієнтів з діагностованим ЦД2 і 311 осіб без будь-яких порушень вуглеводного обміну. Генотипування осіб за rs1800247-поліморфізмом гена BGLAP виконали за допомогою полімеразної ланцюгової реакції з аналізом довжини рестрикційних фрагментів (PCR-RFLP). Для оцінювання зв’язку між генотипами за rs1800247-поліморфізмом гена BGLAP і розвитком ЦД2 застосували логістичну регресію (враховуючи незалежну змінну «генотип × АГ») у межах домінантної, рецесивної, наддомінантної та адитивної моделей спадкування.

Nowadays it is known about at least two proteins of bone tissue that could influence on systemic energy metabolism. The first one is undercarboxylated osteocalcin (unOCN), which directly stimulates insulin production in pancreatic β-cells and increases peripheral tissue sensitivity to this hormone. The second one, osteotesticular tyrosine phosphatase (OST-PTP), regulates OCN gene expression in accordance with metabolic requirements of the bone tissue [1]. Bone reparation was the priority process of energy substrate distribution in animals, as the integrity of bones was vitally needed. Thus, active OST-PTP dephosphorylates the β-subunits of insulin receptor that leads to inhibition of metabolic and proliferative effects of this hormone. The consequence is the decreased rate of Runx 2, which is the crucial transcription factor for OCN gene expression. Temporary reducing of unOCN level in systemic circulation leads to hypoinsulinaemia, as well as decreased adipose, muscle and liver tissue sensitivity to insulin. As the result, more glucose reaches a bone cell that is important for synthesis and resorption of bone tissue [1][2][3].
The involvement of unOCN in the energy metabolism was confirmed in experimental and clinical studies. It is known that OCN gene-knockout mouse (Bglap -/-) have increased bone tissue formation without the influence on bone resorption. In the same time their phenotype was opposite to OST-PTP gene-knockout mouse (Ptprv -/-), which showed an impaired glucose tolerance and overweight [4,5]. unOCN binds to specific GPRC6A receptors and enhances insulin and adiponectin expression, as well as β-cell proliferation [6].
Meta-analysis indicates the presence of reverse correlation between OCN plasma concentration and insulin, fasting glucose and glycated hemoglobin in patients with type 2 diabetes mellitus (T2DM) [7]. Moreover, rs1800247-single nucleotide polymorphism (SNP) of BGLAP gene was associated with decreased risk of arterial hypertension (AH) and lower diastolic blood pressure [8].
Выводы. Установлено, что носители СС-генотипа имеют более благоприятные показатели липидного метаболизма среди украинцев с СД2 и без АГ. Однако отсутствует связь между rs1800247-однонуклеотидным полиморфизмом и возникновением СД2, а также показателями артериального давления. Relatively healthy subjects without any carbohydrate metabolism disorders (which was confirmed by a fasting plasma glucose level less than 5.6 mmol/L and a 75 g oral glucose tolerance test result less than 7.8 mmol/L) and negative family history of diabetes were enrolled in the control group. All the examined individuals were selected from hospital records in the 5th Sumy Clinical Hospital and Sumy Regional Diagnostic Center between 2011-2019. AH was diagnosed in 107 T2DM patients and 156 control subjects with systolic blood pressure higher than 140 mmHg and/or diastolic blood pressure higher than 90 mmHg and no antihypertensive therapy (according to the WHO, 1999). Polymerase chain reaction-restriction fragment length polymorphism analysis (PCR-RFLP) was used for BGLAP rs1800247-genotyping. Full information about genotyping protocol was presented in our previous research [9].

Materials and methods
The study design complies with the Declaration of Helsinki and was approved by the Ethic Committee of Medical Institute of Sumy State University and the Ethic Committee of Medical Institute of Sumy State University (number 4/02.18.09). A written informed consent was obtained from all participants.
Statistical analysis. Continuous variables were presented as the mean ± SD, categorical -as absolute and percentage values. Chi square (χ 2 ) test was used for comparing the categorical variables. Two-tailed Student's t-test was used to compare the mean values of two groups (with preliminary verification of the data distribution for normality through the Shapiro-Wilk test). The mean values of three groups were compared using ANOVA with further Bonferroni post hoc test. Logistic regression with interaction term "genotype × AH" was used for the association analysis between four models of inheritance: dominant, recessive, over-dominant and additive.
The adjustment for age, sex, smoking and body mass index (BMI) was used to exclude the influence of other T2DM risk factors. Bonferroni correction was applied for accurate results. The impact of rs1800247-C minor allele on systolic, diastolic, pulse and mean arterial blood pressure among diabetic patients was estimated via linear regression. All calculations were performed using Statistical Package for the Social Sciences software (SPSS, version 22.0, Chicago, IL, USA). A value of P ˂ 0.05 was considered as significant.

Results
The clinical characteristics of compared groups are shown in Table 1. Statistically significant differences in age, sex, BMI, fasting glucose, lipid profile and blood pressure parameters (P ˂ 0.05), but not among smokers (P = 0.66), was found in groups with AH. In contrast, T2DM patients and controls without AH were comparable in age, BMI, smoking status, triglyceride concentration and blood pressure parameters (P > 0.05), but not in sex, fasting glucose levels, total cholesterol, high density lipoprotein (HDL) cholesterol and low density lipoprotein (LDL) cholesterol concentrations (P ˂ 0.05).
Logistic regression with interaction term was used to study the influence of rs1800247 genotypes on the T2DM development There was no statistically significant associations neither in AH patients, nor in non-AH individuals in all models of inheritance ( Table 2).
Then we performed the linear regression models to compare the rs1800247 genotype impact on the arterial blood pressure values. No significant differences were found for systolic, diastolic, pulse and mean arterial blood pressure among T2DM patients ( Table 3). Table 4 indicates the parameters of lipid profile in T2DM patients with and without AH stratified by rs1800247 genotypes. Statistically significant differences were found between TT and CC carriers in total cholesterol (P = 0.012), HDL cholesterol (P = 0.015) and LDL cholesterol (P = 0.04) concentrations among T2DM individuals without AH.

Discussion
It is known that T2DM patients have lower carboxylated osteocalcin (cOCN) and unOCN concentrations than relatively healthy subjects [10,11]. The meta-analysis showed significantly decreased baseline serum total OCN in T2DM compared with non-T2DM subjects. Moreover, a unit elevation in serum total OCN was correlated with a mean increase in HOMA-B, as well as mean reduction in HbA1c, fasting plasma glucose, HOMA-IR and BMI [12].
In this study, we continued to study the association between BGLAP rs1800247 SNP and T2DM development among Ukrainians. The lack of studied correlation matches both our previous research [9] and Das et al. study, which excluded BGLAP rs1800247 SNP as a T2DM potential risk factor [13].
Cardiovascular diseases are widespread chronic complications in patients with T2DM. Animal and in vitro studies showed the protective effect of unOCN on vessels. This was explained by the enhanced expression of endothelial nitric oxide synthase (eNOS) and nitric oxide (NO) production [14]. Lower serum OCN concentration was found among hypertensive men, but not women. Moreover, serum OCN level was inversely associated with systolic blood pressure in Chinese men, but not women [15].
Another study showed that BGLAP rs1800247 was associated with lower risk of AH and diastolic blood pressure in Chinese population [8]. In contrast, our study indicates no relation between BGLAP rs1800247 and blood pressure level among T2DM Ukrainians that can be explained by ethnic differences. Despite this, in present study, we showed that T2DM non-hypertensive CC-carriers had significantly lower levels of total cholesterol and LDL cholesterol, but higher concentration of HDL cholesterol compared to those in the TT-genotype. The results obtained may indicate more favorable conditions for the lipid metabolism in CC-homozyguous of the examined groups among Ukrainians.

Conclusions
1. Non-hypertensive T2DM CC-carriers had significantly lower levels of total cholesterol (P = 0.012) and LDL cholesterol (P = 0.04), but higher concentration of HDL cholesterol (P = 0.015) compared to the TT-genotype in Ukrainian population.   2. No association was found between rs1800247 SNP and T2DM development among hypertensive Ukrainians (P a int b > 0.05). 3. No association was found between rs1800247 SNP and T2DM development among non-hypertensive Ukrainians (P a int b > 0.05). 4. There was no relation between rs1800247 SNP and blood pressure parameters (systolic, diastolic, pulse and mean blood pressure) among T2DM Ukrainians (P > 0.05).
Perspectives for further research. Further studies with extended groups of comparison are needed for the confirmation of results. Moreover, it will be useful to study the association between other BGLAP SNPs and T2DM and AH development.

Funding
This research was a part of the scientific project "Moleculargenetic and morphological features of lower limb tissues regeneration under conditions of chronic hyperglycemia", state registration No. 0117U003926.