The ENPP1 K121Q polymorphism modulates developing of bone disorders in type 2 diabetes: A cross sectional study

Gene 637 (2017) 100–107

Contents lists available at ScienceDirect

Gene journal homepage: www.elsevier.com/locate/gene

Research paper

The ENPP1 K121Q polymorphism modulates developing of bone disorders in type 2 diabetes: A cross sectional study

MARK

Nahid Neamatia, Seyed Reza Hosseinib, Mahmood Hajiahmadic, Sohrab Halalkhora, Hajighorban Nooreddinid, Haleh Akhavan Niakie, Bahare Korania, Hadi Parsianb,e,⁎ a

Department of Clinical Biochemistry, Faculty of Medicine, Babol University of Medical Sciences, Babol, Iran Social Determinant of Health Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran c Department of Biostatistics, Faculty of Medicine, Babol University of Medical Sciences, Babol, Iran d Department of Radiology, Faculty of Medicine, Babol University of Medical Sciences, Babol, Iran e Cellular and Molecular Biology Research Centre, Babol University of Medical Sciences, Babol, Iran b

A R T I C L E I N F O

A B S T R A C T

Keywords: ENPP1 gene K121Q polymorphism rs1044498 Osteopenia Osteoporosis Type 2 diabetes

Background: Osteoporosis and osteopenia are common diseases in every population. Type 2 diabetes mellitus (T2DM) can lead to the development of various complications, such as bone disorders especially among elderly individuals. Studies suggested that ectonucleotide pyrophosphatase/phosphodiesterase1 (ENPP1) is contributed in insulin resistance and also the inhibition of bone mineralization. In this study, association of K121Q (rs1044498) polymorphism of the ENPP1 gene with T2DM and bone disorders is evaluated. Methods: Four-hundred-and-ninety females who were classified based on bone mineral density (BMD) at lumbar spine and femur were included in this study. In addition, participants were classified according to their diabetes status. K121Q polymorphism was evaluated by the PCR-PFLF technique. One-way ANOVA was used for comparison of various analyzed factors in diseases subgroups and K121Q genotypes. Association of K121Q polymorphism with diabetes and bone disorders was evaluated by logistic regression. Results: Significant association was observed between K121Q polymorphism with osteoporosis and osteopenia (p = 0.041, p = 0.029, respectively), but a similar pattern was not observed in T2DM status (p = 0.723). Moreover, in diabetic patients, K121Q polymorphism showed a better prediction potential for the development of bone disorders in comparison to non-diabetic subjects (p = 0.018; OR = 4.63, p = 0.540; OR = 1.31). There were no significant differences between K121Q genotypes with FBS, Ca, P, vitamin D, PTH and BMD status. Conclusions: The present study implies that K121Q polymorphism of ENPP1 gene is able to modulate the development of bone disorders in T2DM. Therefore in diabetic patients screening of this polymorphism is suggested for the monitoring of these persons.

1. Introduction Osteoporotic fractures usually occur due to decrease in bone mineral density (BMD) and destruction of bone structure (Genant et al., 1999). This disease is an important problem for societies because its prevalence is dramatically increasing in recent years (Cummings and Melton, 2002). Around the world, there are more than two hundred million people with this disease (Kannus, 2003). In Iran 50% of men and 70% of women, over 50 years, suffer from bone problems (osteoporosis or osteopenia) (Rahnavard et al., 2009).

Bone strength is affected by many risk factors such as age, gender, body mass index (BMI), menopausal and nutritional status, hormonal disorders, use of alcohol and some drugs, smoking, physical activity, genetic factors and so on. Studies have shown that > 50% (50–80%) of BMD variations are affected by genetic factors. Nowadays, evaluation of BMD is one of the main methods for estimation of bone strength and is used for identifying and predicting the future fractures risks (Hui et al., 1988, Lau et al., 2001, Dontas and Yiannakopoulos, 2007). In addition to osteoporosis, diabetes mellitus is another serious disease. Unfortunately, the number of people that suffer from diabetes

Abbreviations: AHAP, Amirkola health and aging project; BMD, Bone mineral density; BMI, Body mass index; Ca, Calcium; DEXA, Dual energy X-ray absorptiometry; ENPP1, Ectonucleotide pyrophosphatase/phosphodiesterase1; FBS, Fasting blood sugar; HWE, Hardy-Weinberg Equilibrium; P, Phosphorus; PCR, Polymerase chain reaction; PTH, Parathyroid hormone; RFLP, Restriction fragment length polymorphism; T2DM, Type 2 diabetes mellitus; vit D, Vitamin D ⁎ Corresponding author at: Clinical Biochemistry, Social Determinant of Health Research Center, Health Research Institute, Babol University of Medical Sciences, Ganjafrooz Ave, Babol, Iran. E-mail address: [email protected] (H. Parsian). http://dx.doi.org/10.1016/j.gene.2017.09.042 Received 10 June 2017; Received in revised form 17 September 2017; Accepted 19 September 2017 Available online 21 September 2017 0378-1119/ © 2017 Published by Elsevier B.V.

Gene 637 (2017) 100–107

N. Neamati et al.

2. Materials and methods

is increasing in all countries. Currently, 382 million people worldwide have diabetes and this number is expected to rise up to 592 million until 2035 (Guariguata et al., 2014). Diabetes as a metabolic disease can lead to complications such as microvascular diseases, retinopathy, neuropathy, nephropathy, and even functional disorders (Association, A. D, 2010). In addition, some researchers have shown that diabetic individuals are more susceptible to osteoporosis than non-diabetic ones. Diabetes could be a predisposing factor to various kinds of fractures. It has been proposed that the incidence rate of fractures in wrist, leg, and hip in elderly diabetics is much more than non-diabetics (Karimifar et al., 2012, Jiajue et al., 2014, Jiao et al., 2015). Despite the impact of recognized risk factors that are involved in developing the diabetes and osteoporosis, substantiated evidence suggests that certain genetic factors could be effective in occurrence and severity of their complications. It is suggested that ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) is one of the candidate genes that is involved and contributed to the pathogenesis of diabetes and also osteoporosis (Gaulton et al., 2008, Cheung et al., 2010, Ermakov et al., 2010). It has been reported that extracellular matrix mineralization is an important process in bone formation, resorption and also its maintenance (Harmey et al., 2004; Forriol and Shapiro, 2005). Any defect in this process could be associated with changes in bone strength and structure. Inorganic phosphorus (Pi) is an essential element for bone mineralization and is tightly regulated by three enzymes: ENPP1, alkaline phosphatase, liver/bone/kidney (ALPL) and ANKH inorganic pyrophosphate transport regulator (ANKH) (Ermakov et al., 2010). ENPP1 that is one of the subtypes of ENPP family, serves as a mineralization modulator by hydrolyzing tri-nucleotide phosphate and generating pyrophosphate (PPi) (Harmey et al., 2004). Another enzyme, called ALPL, is able to hydrolyze PPi to Pi and provide required inorganic phosphate for bone mineralization (Hessle et al., 2002). ANKH serves as a transporter of PPi from intracellular to extracellular space and vice versa (Ho et al., 2000). PPi by itself is an inhibitor of bone mineralization. Interestingly, it serves as ALPL substrate for production of Pi and can contribute in bone mineralization. In addition to the above mentioned function, another property that is attributed to the ENPP1 gene is that encoding a transmembrane glycoprotein that interacts with insulin receptor and inhibits its tyrosine kinase activity (Maddux et al., 1993). Therefore, it can involve in insulin signaling. Thus ENPP1 may be a suitable candidate for research in the diabetes field due to insulin resistance. A linkage between hip, spine and wrist BMD with ENPP1 gene polymorphisms has been reported and some studies suggested ENPP1 as a susceptible gene for BMD variations (Cheung et al., 2010; Ermakov et al., 2010). To our best knowledge, there is no available data on the association between diabetes and osteoporosis (or osteopenia) regarding to ENPP1, rs1044498 (K121Q) polymorphism. K121Q (rs1044498) polymorphism of ENPP1 gene is a functional missense mutation where adenine is replaced by cytosine and results in transcoding of glutamine amino acid instead of lysine at codon 121 of mRNA transcript. There are some controversial data that suggested an association between rs1044498 polymorphism with insulin resistance in type 2 diabetes mellitus (T2DM) patients (Keshavarz et al., 2006, Saberi et al., 2011, Jing et al., 2012, Prakash et al., 2013, Badaruddoza et al., 2014). Summary of some investigations about the association between ENPP1 polymorphisms and its related phenotypes are presented in Table 1. The present study aimed to determine probable association between rs1044498 polymorphism of ENPP1 gene with osteoporosis in T2DM patients.

2.1. Study design Among the 1616 individuals that participated in Amirkola health and aging project (AHAP) there were 733 old women (> 60 years old) (Hosseini et al., 2013). After excluding the person who have taken medications related to BMD (such as alendronate or Ibandronate, …), 490 females were recruited and were evaluated for BMD at lumbar spine (L1-L4) and femur using a dual energy X-ray absorptiometry (DEXA) densitometer (DMS Lexxos DR, French). Based on WHO criteria, participants were stratified in three separate groups according to their BMD: osteoporosis, osteopenia and normal groups. If a person had BMD 2.5 standard deviations (or more) below than average value for young healthy adults, i.e. T-score ≤ − 2.5 SD, considered as osteoporosis patient; T-score higher than −1.0 SD as normal and BMD measures between − 2.5 < T-score ≤ − 1 considered as osteopenia. In our study, 279 females suffered from osteoporosis, 154 females had osteopenia and 57 subjects had no sign of any bone defect. The status of body weight, height and serum levels of biochemical and hormonal biomarkers such as fasting blood sugar (FBS), calcium (Ca), phosphorus (P), parathyroid hormone (PTH), and vitamin D (vit. D) were extracted from the AHAP database. Body mass index (BMI) was calculated by dividing weight (Kg) to height (m2). In addition to the classifications of participants according to their BMD results, we also classified our included persons according to their diabetes status. Diabetes status was determined based on physician report sheet and measurement of fasting blood glucose level in two separate sampling (Association, A. D, 2014). Our study was approved by our university's ethics committee. 2.2. DNA extraction, PCR and RFLP Genomic DNA has been extracted from leukocytes in whole blood using QIAamp DNA blood mini kit (Qiagen, Korea) according to manufacture protocol. Quality of the extracted DNA was assessed by running on 1% agarose gel and its concentration measured by nanodrop (Thermoscientific, USA). DNA extraction has been done in base AHAP study and resulted DNAs have been stored at − 80 °C. To amplify the rs1044498 region from the ENPP1 gene, primers designed by Gene Runner software (version 5.1.06 beta) and confirmed via NCBI (primer blast) and online oligo-analyzer. The forward primer was 5′-GTAGTGGCAGATTCTGTGAGTGAC-3′ and the reverse one was 5′-CCGCTAAGACGCTGGAAGATACC-3′. Amplification was performed in 25 μl total reaction mixture volume contained forward and reverse primer (10 pmol), tris-HCL buffer (pH 8.8) (1 mM), dNTPmix (0.2 mM), MgCl2 (1.5 mM), taq DNA polymerase (1 unit) and DNA template (20 ng). The amplification was performed according to the following program: initial denaturation at 94 °C for 5 minutes, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 30 s, extension at 72 °C for 1 min and final extension at 72 °C for 10 min. In order to ensure the amplification of the ENPP1 rs1044498 region, its electrophoresis was performed by running on 2% agarose gel for 30 min and ethidium bromide staining. The amplified DNA was visualized on a trans-illuminator under ultraviolet light by gel documentation and analysis system (UVdoc, England). In order to assess the K121Q polymorphism, 10 μl of PCR product was digested by 3.3 Unit of AvaII restriction enzyme (Thermoscientific, USA) at 37 °C for 7 h. After digestion, the presence of K121Q polymorphism was determined by running on polyacrylamide gel electrophoresis for 30 min and ethidium bromide staining. Samples with KK genotype had one fragment (i.e. one band), samples with QQ genotype had two separate fragments and samples with KQ genotype had three separate fragments. The lengths of the PCR product and resulting fragments after digestion were assessed according to the 50 bp DNA size markers. PCR 101

Gene 637 (2017) 100–107

N. Neamati et al.

Table 1 Summary of some investigations about the association between ENPP1 polymorphisms and its related phenotypes. Name

Phenotype

Population

Odd's Ratio (95% CI)

P-value

rs1044498 A/C K [Lys] → Q [Gln] MAF: 0.34

cardiovascular disease

Italian

5.94 (1.88–18.78)

0.002

Hypertension

Malaysian

–

Type 2 Diabetes mellitus (T2DM)

central Indian

1.199 (0.81–1.77) 0.96 (0.90–2.00) –

Iranian Danish Korean

Obesity

Mexican children

0.911 0.630–1.319 1.02 (0.81–1.29) 1.583 (0.78–3.19) 1.44 (1.06–1.94) 1.41 (1.13–1.76) 0.84 [0.72;0.97]

Bone size

Chuvashia

–

Japanese north Indian North Indian Punjabi Chinese rs7754561 A/G MAF: 0.45 rs1974201 G/A/C/T MAF: 0.41

Hip geometry variation

United States

–

Finding

Reference

(Bacci et al., 2011)

0.774

Among obese individuals, Q121 allele introduced as independent predictor of cardiovascular diseases. No significant association observed

0.3689

No significant association observed

(Vasudevan et al., 2009) (Tripathi et al., 2013)

0.70

No significant association observed

(Saberi et al., 2011)

0.6

No significant association observed

0.89

No significant association observed

(Rasmussen et al., 2000) (Seo et al., 2008)

0.83

No significant association observed

(Keshavarz et al., 2006)

Higher risk (1.58 fold) of T2DM has been observed

(Prakash et al., 2013)

significant association of Q121 observed

(Badaruddoza et al., 2014) (Jing et al., 2012)

⁎

0.199 0.015 0.003 0.02 0.01

⁎

⁎

The association of Q121 was confirmed using a meta-analysis Significant association observed

⁎

⁎

−5

3.78 × 10

Functional role of rs7754561 on bone fragility and osteoporosis Significant association observed

(Mejía-Benítez et al., 2013) (Ermakov et al., 2010) (Cheung et al., 2010)

Studies in which a significant association has been reported between genes and polymorphism, has been marked by asterisk above the Probability value level.

(HWE) were appraised using Online Encyclopedia for Genetic Epidemiology (www.oege.org) and chi-square test. Independent sample t-test or one-way analysis of variance (ANOVA) followed by Tukey posthoc test were used for comparisons of the basic clinical characteristics of subjects in groups and also evaluation the difference between clinical characteristics of subjects in various K121Q genotypes. In order to evaluate the effect of K121Q genotypes in the association of osteoporosis with T2DM, multinomial logistic regression was used and the related odds ratio (OR) was determined. All statistical analysis was performed using SPSS software (version 23.0) and P-values < 0.05 considered significant. 3. Results All participants in this study were classified into two big categories of diabetic (DM) and non-diabetics ones. Then, we classified these two groups to osteoporosis (OS), osteopenia (OP) and normal subgroups. Therefore we had six separate groups. Summary of clinical characteristics of subjects are presented in Table 2. As it is clear from this table, significant differences were observed in mean age and BMI in some included groups. K121Q genotype distribution was in line with Hardy–Weinberg equilibrium (minor allele frequencies: 18%, χ2 goodness of fit: 0.31 and P-value = 0.58). Allele and genotype distribution frequencies are shown in Table 3. As it is presented in Table 4, in diabetic individuals there were significant differences between frequencies of K121Q genotypes between various subgroups (according to their bone health, i.e. OS, OP and normal cases), but such a pattern did not observe in non-diabetic category. Multiple regression analysis revealed that non-diabetic carriers of KQ/QQ genotypes are at higher risk for osteopenia (OR = 1.64) and osteoporosis (OR = 1.31) compared to KK genotype; both of them not significant. But, in diabetic subjects, there were significant differences between frequencies of K121Q genotypes in three subgroups. In

Fig. 1. RFLP analysis of rs1044498 (K121Q) using Ava II restriction enzyme. Lane 1 and 5 present as native form (i.e. KK genotype); Lane 2 and 6 present as heterozygus mutant form (i.e. KQ genotype) and Lane 3 present as mutant homozygus form (i.e. QQ genotype). Lane 4 present as 50 bp DNA ladder.

product size was 327 bp and after digestion, two separate fragments resulted as follow: a 97 bp fragment and a 230 bp fragment (Fig. 1). 2.3. Statistical analysis K121Q minor allele frequencies and Hardy-Weinberg Equilibrium 102

Gene 637 (2017) 100–107

N. Neamati et al.

Table 2 Summary of clinical characteristics of participants. no-DM

Number (%) Age (year) BMI (Kg/m2) FBS (g/dl) Ca (g/dl) P (g/dl) PTH Vit.D (ng/ml) LS·BMD (g/cm2) FN·BMD (g/cm2) T-S T-F

T2DM

P-value

Control

OS

OP

Control

OS

OP

30 (6.1%) 64.1 ± 3.6 (a) 31.2 ± 4.2 (b) 102.3 ± 9.1 (a) 9.24 ± 0.4 (a) 3.9 ± 0.5 (a) 59.2 ± 28.3 (a) 36.4 ± 26 (a) 1.05 ± 0.08 (a) 0.99 ± 0.06 (a) − 0.02 + 0.73 (a) − 0.13 ± 0.58(a)

189 (38.6%) 69.6 ± 7.2 (b) 27.4 ± 4.8 (a) 97.7 ± 11.7 (a) 9.22 ± 0.4 (a) 4.0 ± 0.6 (a) 61.4 ± 35.6 (a) 35.6 ± 32.5 (a) 0.67 ± 0.11 (b) 0.77 ± 0.12 (b) −3.38 ± 0.98 (b) −2.34 ± 1.00 (b)

91 (18.6%) 66.7 ± 6.8 (a,b) 29.6 ± 4.4 (a,b) 98.9 ± 11.1 (a) 9.29 ± 0.4 (a) 4.1 ± 0.7 (a) 59.4 ± 25.9 (a) 37.2 ± 34.9 (a) 0.87 ± 0.07 (c) 0.86 ± 0.09 (c) − 1.61 ± 0.6 (c) − 1.21 ± 0.67 (c)

27 (5.5%) 65.4 ± 5.8 (a) 31 ± 4.5 (b) 138.6 ± 43.8 (b) 9.26 ± 0.4 (a) 4.0 ± 0.5 (a) 57.3 ± 28.8 (a) 39.3 ± 42.4 (a) 1.06 ± 0.09 (a) 1.00 ± 0.09 (a) 0.08 ± 0.81 (a) 0.01 ± 0.75 (a)

90 (18.4%) 69.9 ± 6.9 (b) 27.4 ± 5.2 (a) 167.1 ± 59.6 (c) 9.24 ± 0.5 (a) 4.0 ± 0.6 (a) 59.2 ± 32.4 (a) 37.2 ± 36.2 (a) 0.69 ± 1.1 (b) 0.69 ± 0.11 (b) − 3.23 ± 0.99 (b) − 2.58 ± 0.99 (b)

63 (12.9%) 65.7 ± 5.3 (a) 29.8 ± 4.4 (a,b) 163.2 ± 66.8 (c) 9.24 ± 0.4 (a) 4.1 ± 0.6 (a) 49.7 ± 32.0 (a) 38.0 ± 38.4 (a) 0.88 ± 0.08 (c) 0.84 ± 0.09 (c) −1.49 ± 0.7 (c) −1.4 + 0.76 (c)

< 0.001 < 0.001 < 0.001 0.882 0.856 0.282 0.993 < 0.001 < 0.001 < 0.001 < 0.001

no-DM: non-diabetic; T2DM: Type 2 Diabetes Mellitus; OS: Osteoporosis; OP: Osteopenia; LS·BMD: Lumbar spine BMD; FN·BMD: Femural neck BMD; T-S: T score at lumbar spine; T-F: T score at femural neck; Data have shown as mean ± SD; significant differences observed between various (a), (b) or (c).

without considering the diabetic status and only taking into account the bone health status, significant differences were observed in mean age, BMI, femoral neck/lumbar spine BMD and T-score between normal, osteopenia and osteoporosis groups (Fig. 3). We have also evaluated the risk of diabetes and osteoporosis separately. Genotype frequencies of K121Q based on the above-mentioned classifications are represented in Table 6. The frequencies of K121Q genotypes have no significant difference in diabetic versus non-diabetic group. Using regression analysis, the risk of diabetes in carriers of KQ/ QQ genotypes was approximately equal to individuals with KK genotype (OR = 0.931, 95%CI = 0.629–1.380; P = 0.723). In terms of the osteoporosis classification, we observed that frequencies of KQ/QQ genotypes in osteopenia and osteoporosis groups were significantly higher than normal groups and the differences were statistically significant (P = 0.029 and 0.041, respectively). Using multiple regression analysis, it has been shown that carriers of KQ/QQ genotypes were significantly at higher risk of osteopenia (OR = 2.32, 95%CI = 1.13–4.85; P = 0.029) and osteoporosis (OR = 2.12, 95%CI = 1.05–4.29; P = 0.041) compared to individuals with KK genotype. Comparison of the clinical characteristics of subjects between various genotypes in each group (using ANOVA analysis) showed no significant difference (data not shown).

Table 3 K121Q Allele and genotype distribution frequencies in included subgroups. Allele

no-DM

T2DM

Control OP OS Control OP OS

Genotype

K (%)

Q (%)

KK (%)

KQ (%)

QQ (%)

50 (83.3) 143 (78.6) 313 (82.8) 51 (94.5) 102 (81) 143 (79.5)

10 (16.7) 39 (21.4) 65 (17.2) 3 (5.5) 24 (19) 37 (20.5)

22 (73.3) 57 (62.6) 128 (67.7) 24 (88.9) 42 (66.7) 57 (63.3)

6 (20) 29 (31.9) 57 (30.2) 3 (11.1) 18 (28.6) 29 (32.2)

2 5 4 0 3 4

(6.7) (5.5) (2.1) (4.8) (4.4)

no-DM: non-diabetic; T2DM: Type 2 diabetes mellitus; OP: Osteopenia; OS: Osteoporosis.

addition, multiple regression analysis showed that diabetic carriers of KQ/QQ genotypes are absolutely at higher risk of developing osteoporosis and osteopenia versus individuals with KK genotype. For eliminating the effects of age and BMI on the risk of osteoporosis, as their differences were observed statistically significant in the subgroups, adjustment with these variables were performed. The results were the same and did not show a big discrepancy. Clinical characteristics of subjects in each subgroup according to the rs1044498 genotypes are summarized in Table 5. As it shows, no significant differences were observed in clinical characteristics of subjects, except in lumbar spine BMD and T-score in diabetic group that suffered from osteopenia. In addition, we were also interested if various genotypes of rs1044498 could be accounted as a risk factor for osteoporosis and diabetes, separately. Significant differences were observed at mean FBS, LS·BMD and T-S in diabetic group versus non-diabetic ones. Moreover, mean PTH levels were slightly lower (but not significant) in diabetic patients compared with non-diabetic individuals (Fig. 2). Additionally,

4. Discussion The present study aimed to determine the association between diabetes and osteoporosis regarding rs1044498 (K121Q) polymorphism of the ENPP1 gene within the female population in northern Iran. In our study, Q allele of rs1044498 was significantly associated with osteoporosis in diabetic post-menopausal women. By evaluation of the effects of this polymorphism on diabetes and osteoporosis separately, no

Table 4 Estimation of the risk of osteoporosis and osteopenia by regression analysis. Genotype

no-DM

T2DM

Control OP OS Control OP OS

Unadjusted

KK

KQ/QQ

22 57 128 24 42 57

8 34 61 3 21 33

Adjusted *

P-value

OR (CI 95%)

P-value

OR (CI 95%)

0.288 0.540

1.640 (0.66–4.09) 1.311 (0.55–3.11)

0.381 0.577

1.52 (0.59–3.94) 1.303 (0.51–3.309)

0.038 0.018

4.0 (1.08–14.82) 4.632 (1.29–16.57)

0.031 0.06

4.429 (1.14–17.14) 3.572 (0.95–13.48)

no-DM: non-diabetic; T2DM: Type 2 diabetes mellitus; OP: Osteopenia; OS: Osteoporosis; * The risk of osteoporosis and osteopenia estimated after adjustment with age and BMI.

103

104

Control KK 24 64.8 ± 5.8 31.8 ± 4.7 143 ± 43.6 9.28 ± 0.45 3.9 ± 0.57 59 ± 29.8 37 ± 44.1 1.05 ± 0.09 1.0 ± 0.09 − 0.2 ± 0.78 0.0 ± 0.76

T2DM

KQ 3 69.7 ± 5.1 31.7 ± 1.5 96 ± 5.7 9.1 ± 0.35 4.2 ± 0.31 39 ± 5.13 51 ± 28.5 1.15 ± 0.05 1.0 ± 0.1 0.9 ± 0.49 0.07 ± 0.8

KQ 6 64.5 ± 5.1 30.3 ± 4.6 101 ± 9.9 9.3 ± 0.37 4.2 ± 0.61 56 ± 36.4 30 ± 24.2 1.08 ± 0.11 0.98 ± 0.05 0.3 ± 1.02 −0.12 ± 0.36

QQ 0 – – – – – – – – – – – 0.179 0.972 0.075 0.511 0.457 0.261 0.606 0.075 0.963 0.051 0.894

P-value

QQ 2 62.5 ± 0.7 33.2 ± 1.7 103 ± 9.9 9.2 ± 0.01 3.8 ± 0.92 88 ± 67.0 22 ± 5.3 0.99 ± 0.01 1.06 ± 0.02 − 0.55 ± 0.07 0.55 ± 0.21 0.807 0.700 0.980 0.989 0.582 0.350 0.559 0.402 0.263 0.334 0.233

QQ 5 72.8 ± 7.5 26.7 ± 4.8 98 ± 10.3 9.4 ± 0.68 4.1 ± 0.97 71 ± 19.1 22 ± 12.1 0.85 ± 0.1 0.79 ± 0.08 − 1.8 ± 0.81 − 1.76 ± 0.61

QQ 3 66.0 ± 1.0 28.5 ± 7.0 181 ± 121 8.9 ± 0.25 4.2 ± 0.78 41 ± 18.6 52 ± 38.3 1.01 ± 0.21 0.92 ± 0.15 − 0.3 ± 1.09 − 0.71 ± 1.36

KQ 29 66.3 ± 7.7 29.3 ± 3.9 96 ± 9.2 9.4 ± 0.48 4.0 ± 0.74 61 ± 26.7 47 ± 36 0.87 ± 0.07 0.85 ± 0.08 −1.6 ± 0.65 −1.23 ± 0.65

KQ 18 65.9 ± 4.8 30.5 ± 5.2 151 ± 84.8 9.3 ± 0.36 3.9 ± 0.58 52 ± 32.9 40 ± 34.0 0.89 ± 0.08 0.85 ± 0.71 − 1.4 ± 0.70 − 1.3 ± 0.67

OP KK 57 66.5 ± 6.1 30.0 ± 4.6 100 ± 11.9 9.2 ± 0.4 4.1 ± 0.64 57 ± 26.1 33 ± 34.4 0.87 ± 0.06 0.86 ± 0.09 − 1.5 ± 0.56 − 1.16 ± 0.67

Op KK 42 65.6 ± 5.7 29.6 ± 3.9 167 ± 57.4 9.2 ± 0.41 4.2 ± 0.62 49 ± 32.8 36 ± 40.7 0.87 ± 0.06 0.83 ± 0.09 − 1.6 ± 0.50 − 1.5 ± 0.74

P-value

0.976 0.687 0.647 0.209 0.533 0.848 0.766 0.007 0.186 0.006 0.214

P-value

0.124 0.253 0.309 0.168 0.765 0.490 0.153 0.875 0.281 0.779 0.151

P-value

Os KK 57 69.9 ± 7.4 27.7 ± 5.3 169 ± 61.2 9.3 ± 0.57 4.0 ± 0.67 58 ± 31.4 32 ± 27 0.70 ± 0.12 0.69 ± 0.11 − 3.1 ± 1.11 − 2.5 ± 1.08

OS KK 128 69.6 ± 7.0 27.5 ± 4.8 98 ± 10.6 9.2 ± 0.4 4.1 ± 0.6 63 ± 35.9 34 ± 28.7 0.68 ± 0.11 0.72 ± 0.12 − 3.3 ± 1 − 2.37 ± 0.99

KQ 29 69.4 ± 6.1 27.7 ± 5.1 163 ± 56.7 9.2 ± 0.46 4.1 ± 0.52 60 ± 35.5 45 ± 48.6 0.68 ± 0.08 0.69 ± 0.09 − 3.3 ± 0.71 − 2.6 ± 0.72

KQ 57 70.2 ± 7.8 27.0 ± 4.9 97 ± 14.2 9.3 ± 0.4 4.0 ± 0.6 55 ± 33.4 36 ± 36.9 0.66 ± 0.1 0.72 ± 0.11 − 3.55 ± 0.94 − 2.31 ± 0.99

QQ 4 73.7 ± 5.2 22.3 ± 3.2 169 ± 72.6 9.0 ± 0.17 4.0 ± 0.34 64 ± 28.6 41 ± 44.1 0.65 ± 0.09 0.63 ± 0.17 − 3.6 ± 0.83 − 3.2 ± 1.33

QQ 4 64.0 ± 4.7 29.2 ± 7.7 92.5 ± 10 9.2 ± 0.4 3.7 ± 0.5 66 ± 55.4 60 ± 72 0.64 ± 0.07 0.79 ± 0.15 −3.62 ± 0.61 −1.72 ± 1.19

0.508 0.127 0.902 0.656 0.924 0.933 0.289 0.527 0.468 0.589 0.468

P-value

0.251 0.620 0.541 0.559 0.533 0.324 0.301 0.374 0.442 0.229 0.435

P-value

no-DM: non-diabetic; T2DM: Type 2 Diabetes Mellitus; OS: Osteoporosis; OP: Osteopenia; LS·BMD: Lumbar spine BMD; FN·BMD: Femural neck BMD; T-S: T score at lumbar spine; T-F: T score at femural neck; Data have shown as mean ± SD.

Number (%) Age (year) BMI (Kg/m2) FBS (g/dl) Ca (g/dl) P (g/dl) PTH Vit.D (ng/ml) LS·BMD (g/cm2) FN·BMD (g/cm2) T-S T-F

Number (%) Age (year) BMI (Kg/m2) FBS (g/dl) Ca (g/dl) P (g/dl) PTH Vit.D (ng/ml) LS·BMD (g/cm2) FN·BMD (g/cm2) T-S T-F

Control KK 22 64.1 ± 3.5 31.3 ± 4.3 102 ± 9.3 9.2 ± 0.47 3.9 ± 0.53 57 ± 22.3 39 ± 27.5 1.04 ± 0.08 0.98 ± 0.07 − 0.06 ± 0.65 − 0.19 ± 0.62

no-DM

Table 5 Clinical characteristics of the included participants according to the K121Q genotypes.

N. Neamati et al.

Gene 637 (2017) 100–107

Gene 637 (2017) 100–107

N. Neamati et al.

Fig. 2. Boxplot for comparison of clinical characteristics of diabetic and non-diabetics subjects.

Fig. 3. Boxplot for clinical characteristics in normal, osteopenia and osteoporosis group.

Recently, the number of investigations about genetic factors relating to bone diseases have increased. Some of those implies on their utilization as a method for prediction, screening and subsequent management of related diseases. Up to now, > 150 genetic variations

significant differences were observed in K121Q minor allele frequencies between diabetic and non-diabetic subjects. However frequencies of Q121 allele were significantly higher in patients with osteopenia and osteoporosis compared to normal subjects. 105

Gene 637 (2017) 100–107

N. Neamati et al.

Table 6 Estimation the risk of type 2 diabetes and bone disorders by logistic regression. Genotype

Osteoporosis status

Diabetic status ⁎

Normal Osteopenia Osteoporosis Non-diabetic Type 2 diabetic

Adjusted⁎

Unadjusted

KK

KQ/QQ

46 99 185 207 123

11 55 94 103 57

P-value

OR (CI95%)

P-value

OR (CI95%)

0.029 0.041

2.323 (1.11–4.85) 2.125 (1.05–4.29)

0.032 0.082

2.26 (1.07–4.75) 1.92 (0.92–4.01)

0.723

0.931 (0.63–1.38)

0.704

1.08 (0.73–1.61)

The risk of osteoporosis and osteopenia estimated after adjustment with age and BMI.

K121Q with T2DM in different ethnicities. In Indian and Chinese populations, the relation between K121Q polymorphism with diabetes has been confirmed (Jing et al., 2012; Prakash et al., 2013; Badaruddoza et al., 2014), but in Japanese (Keshavarz et al., 2006), Korean (Seo et al., 2008), Iranian (Saberi et al., 2011) and Danish (Rasmussen et al., 2000) populations such significant associations were not reported. As presented earlier, significant differences were not observed in Q121 allele frequencies between diabetic and non-diabetic individuals, but significant differences were observed in osteoporosis and osteopenia patients in comparison to normal subjects. Previous studies indicated the protective effects of higher BMI against osteoporosis. It is proposed that leptin can induce proliferation and differentiation of osteoblast cells. In addition, production of estrogen in adipose tissue from androgen (in both sexes) can be benefit to the improvement of bone mass. The main effect of estrogen (specifically in females) is the inhibition of osteoclast activity and improvement of bone formation. In line with the previous studies, in the present study we observed that patients with osteoporosis have lower BMI. However, no significant difference was observed at BMI between various K121Q genotypes in each of the study groups. After adjustment of the groups by BMI and age, the results were the same. Disregarding the effect of genetic factors on osteoporosis, some pathological conditions such as hyperparathyroidism, calcium/phosphorus dyshomeostasis and vitamin D deficiency can lead to osteoporosis (Starup-Linde et al., 2014). All of our study groups were homogenized according to the aforementioned conditions. Moreover, there were no significant differences between serum Ca, P, PTH and vit. D regarding K121Q genotypes in each study groups. In the present study, although we observed a significant association between K121Q polymorphism and osteoporosis risk, we were not able to detect any significant differences in BMD levels with various K121Q genotypes. We expected lower BMD levels in Q121 allele, but as our results showed, the main effect of K121Q polymorphism is independent from BMD. More studies are required to confirm and clarify the main effects of K121Q polymorphism on pathogenesis of diabetes and osteoporosis. In conclusion, a significant association was not observed between Q121 allele and the risk of diabetes in this study; but the results of logistic regression suggested that Q121 allele can be a better predictor for osteoporosis and osteopenia in diabetic subjects compared to nondiabetic subjects. In diabetic subjects K121Q polymorphism is promising a potential predictor for developing osteoporosis. According to our best knowledge, this is the first study concerning the association of diabetes and osteoporosis considering the K121Q polymorphism of ENPP1 gene. Our results imply that it is reasonable to propose rs1044498 polymorphism of the ENPP1 gene as a probable risk factor for prediction of osteoporosis in diabetic patients. However, more studies are needed for confirmation of these results. Potentially the results of bigger studies in this field can open a new window for better monitoring and understanding the underlying mechanism of these two important diseases.

attributed to osteoporosis have been identified by genome wide association studies. Most of these candidate genes are probably contributed in BMD variations and pathogenesis of osteoporosis. There are some reports about the functional role of the 6q23.2 region (belonging to the ENPP1 gene) in bone mineralization disorders. Babij et al. in a study on mutant mice (C397S mutation of the ENPP1 gene) observed the lower BMD, joint disorders and vascular calcification and concluded that the normal ENPP1 gene has some important roles in bone mineralization (Babij et al., 2009). Bone mineralization is strongly dependent on a balanced PPi/Pi ratio at bone matrix. It has been stated that higher concentration of PPi is related to inhibition of this process. In addition, studies have shown that a lower PPi level may be related to hypophosphatemia, vascular calcification, defect in bone mineralization and finally osteoblast differentiation (Rutsch et al., 2003, White et al., 2014). Type 2 diabetes mellitus is considered as an underlying disease relating to bone destruction. In recent studies, it has been shown that diabetic patients are much more susceptible to developing osteoporosis (Vestergaard, 2007; Karimifar et al., 2012; Jiajue et al., 2014). However, there are some conflicts about BMD and the risk of osteoporosis in diabetic patients. Jiao et al. reported that in type 1 diabetic patients, BMD is lower than healthy subjects and T2DM patients (Jiao et al., 2015). Vestergaard et al. reported that there is a significant association between diabetes and osteoporosis, but most diabetic patients have a higher BMD compared to non-diabetic subjects (Vestergaard, 2007). The results of this study are consistent with previous researches, where significant association between diabetes and osteoporosis were observed. Lumbar spine BMD was significantly higher in T2DM patients, but no significant difference was observed at femoral neck BMD (Miyata et al., 1996; Yao and Brownlee, 2010). Jiajue et al. have proposed that, although higher BMD was observed in diabetic subjects, inhibition of bone recycling may have been the probable mechanism for bone fragility and osteoporosis in diabetic patients (Jiajue et al., 2014). Jiao et al. proposed that both type 1 and 2 of diabetes are able to contribute to bone destruction by increasing osteoblasts apoptosis and decreasing the production of bone remodeling inducers, such as bone morphogenetic protein and bone specific growth factors. In addition to the above-mentioned mechanisms, it is proposed that hyperglycemia and insulin resistance can affect bone strength by producing advanced glycosylated end products (AGEs) and oxidative stress (Jiao et al., 2015). There are some evidences about genotype variations of the ENPP1 gene relating to diabetes and also to bone disorders. Ermakov et al. reported that rs1799774 and rs7754561 polymorphisms of the ENPP1 gene are significantly related to bone phenotypes, such as bone size and strength (Ermakov et al., 2010). Cheung et al. have reported that the rs1974201 polymorphism of ENPP1 is significantly associated with hip geometry parameters. In Cheung's study, rs858345, rs1044498, rs7768480, rs1799774 and rs7754561 polymorphisms of the ENPP1 genes had a significant association with bone size and strength (Cheung et al., 2010). In addition to these reports regarding genotype variations of ENPP1 in bone disorders, there are some conflicts about the association of 106

Gene 637 (2017) 100–107

N. Neamati et al.

diabetes mellitus. Osteoporos. Int. 25 (8), 1999–2005. Jiao, H., et al., 2015. Diabetes and its effect on bone and fracture healing. Curr.Osteoporos. Rep. 13 (5), 327–335. Jing, C., et al., 2012. Meta-analysis of association studies between five candidate genes and type 2 diabetes in Chinese Han population. Endocrine 42 (2), 307–320. Kannus, P., 2003. International osteoporosis foundation: The facts about osteoporosis and its impact. International Osteoporosis Foundation, Lyon. Karimifar, M., et al., 2012. Evaluation of bone loss in diabetic postmenopausal women. J. Res. Med. Sci.: the official journal of Isfahan University of Medical Sciences 17 (11), 1033. Keshavarz, P., et al., 2006. No evidence for association of the ENPP1 (PC-1) K121Q variant with risk of type 2 diabetes in a Japanese population. J. Hum. Genet. 51 (6), 559–566. Lau, E., et al., 2001. Risk factors for hip fracture in Asian men and women: the Asian osteoporosis study. J. Bone Miner. Res. 16 (3), 572–580. Maddux, B.A., et al., 1993. Inhibitors of insulin receptor tyrosine kinase in fibroblasts from diverse patients with impaired insulin action: evidence for a novel mechanism of postreceptor insulin resistance. J. Clin. Endocrinol. Metab. 77 (1), 73–79. Mejía-Benítez, A., et al., 2013. Analysis of the contribution of FTO, NPC1, ENPP1, NEGR1, GNPDA2 and MC4R genes to obesity in Mexican children. BMC Med. Genet. 14 (1), 21. Miyata, T., et al., 1996. Possible involvement of advanced glycation end-products in bone resorption. Nephrol. Dial. Transplant. 11 (supp5), 54–57. Prakash, J., et al., 2013. K121Q ENPP1/PC-1 gene polymorphism is associated with insulin resistance in a North Indian population. J. Genet. 92 (3), 571–576. Rahnavard, Z., et al., 2009. The incidence of osteoporotic Hip fracture: Iranian Multicenter Osteoporosis Study (IMOS). Res. J. Biol. Sci. 4 (2), 171–173. Rasmussen, S.K., et al., 2000. The K121Q variant of the human PC-1 gene is not associated with insulin resistance or type 2 diabetes among Danish Caucasians. DiabetesAm. Diabetes Assoc. 49 (9), 1608–1611. Rutsch, F., et al., 2003. Mutations in ENPP1 are associated with'idiopathic'infantile arterial calcification. Nat. Genet. 34 (4), 379. Saberi, H., et al., 2011. The ENPP1 K121Q polymorphism is not associated with type 2 diabetes and related metabolic traits in an Iranian population. Mol. Cell. Biochem. 350 (1–2), 113–118. Seo, H.-J., et al., 2008. The K121Q polymorphism in ENPP1 (PC-1) is not associated with type 2 diabetes or obesity in Korean male workers. J. Korean Med. Sci. 23 (3), 459–464. Starup-Linde, J., et al., 2014. Biochemical markers of bone turnover in diabetes patients—a meta-analysis, and a methodological study on the effects of glucose on bone markers. Osteoporos. Int. 25 (6), 1697–1708. Tripathi, A.K., et al., 2013. Obesity and ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) polymorphism and their association with pathophysiology diabetes type 2 in Central Indian population. J. Diabetes Endocrinol. 4 (2), 19–26. Vasudevan, R., et al., 2009. No association of TCF7L2 and ENPP1 gene polymorphisms in Malaysian type 2 diabetes mellitus with or without hypertension. Res. J. Biol. Sci. 4 (6), 703–709. Vestergaard, P., 2007. Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2 diabetes—a meta-analysis. Osteoporos. Int. 18 (4), 427–444. White, K.E., et al., 2014. Hypophosphatemic rickets: revealing novel control points for phosphate homeostasis. Curr. Osteoporos. Rep. 12 (3), 252–262. Yao, D., Brownlee, M., 2010. Hyperglycemia-induced reactive oxygen species increase expression of the receptor for advanced glycation end products (RAGE) and RAGE ligands. Diabetes 59 (1), 249–255.

Acknowledgements The present study was supported by a grant number 2928 of Babol University of Medical Sciences, Babol, Iran. The authors thank vice chancellor for research and technology affairs for the approval of this study. References Association, A. D, 2010. Diagnosis and classification of diabetes mellitus. Diabetes Care 33 (Suppl. 1), S62–S69. Association, A. D, 2014. Diagnosis and classification of diabetes mellitus. Diabetes Care 37 (Suppl. 1), S81–S90. Babij, P., et al., 2009. New Variants in the Enpp1 and Ptpn6 Genes Cause Low BMD, Crystal-Related Arthropathy, and Vascular Calcification. J. Bone Miner. Res. 24 (9), 1552–1564. Bacci, S., et al., 2011. The ENPP1 Q121 Variant Predicts Major Cardiovascular Events in High-Risk Individuals. Diabetes 60 (3), 1000–1007. Badaruddoza, B.B., Matharoo, K., Bhanwer, 2014. Role of the ENPP1 K121Q Polymorphism and Susceptibility to Type 2 Diabetes in North Indian Punjabi Population. J. Diabetes Metab. 5 (10), 1000450. Cheung, C.L., et al., 2010. Hip geometry variation is associated with bone mineralization pathway gene variants: the Framingham Study. J. Bone Miner. Res. 25 (7), 1564–1571. Cummings, S.R., Melton, L.J., 2002. Epidemiology and outcomes of osteoporotic fractures. Lancet 359 (9319), 1761–1767. Dontas, I., Yiannakopoulos, C., 2007. Risk factors and prevention of osteoporosis-related fractures. J. Musculoskelet. Neuronal Interact. 7 (3), 268–272. Ermakov, S., et al., 2010. Association of ALPL and ENPP1 gene polymorphisms with bone strength related skeletal traits in a Chuvashian population. Bone 46 (5), 1244–1250. Forriol, F., Shapiro, F., 2005. Bone development: interaction of molecular components and biophysical forces. Clin. Orthop. Relat. Res. 432, 14–33. Gaulton, K.J., et al., 2008. Comprehensive association study of type 2 diabetes and related quantitative traits with 222 candidate genes. Diabetes 57 (11), 3136–3144. Genant, H.K., et al., 1999. Interim report and recommendations of the World Health Organization task-force for osteoporosis. Osteoporos. Int. 10 (4), 259–264. Guariguata, L., et al., 2014. Global estimates of diabetes prevalence for 2013 and projections for 2035. Diabetes Res. Clin. Pract. 103 (2), 137–149. Harmey, D., et al., 2004. Concerted regulation of inorganic pyrophosphate and osteopontin by akp2, enpp1, and ank: an integrated model of the pathogenesis of mineralization disorders. Am. J. Pathol. 164 (4), 1199–1209. Hessle, L., et al., 2002. Tissue-nonspecific alkaline phosphatase and plasma cell membrane glycoprotein-1 are central antagonistic regulators of bone mineralization. Proc. Natl. Acad. Sci. 99 (14), 9445–9449. Ho, A.M., et al., 2000. Role of the mouse ank gene in control of tissue calcification and arthritis. Science 289 (5477), 265–270. Hosseini, S.R., et al., 2013. Cohort profile: The Amirkola Health and Ageing Project (AHAP). Int. J. Epidemiol. 43 (5), 1393–1400. Hui, S.L., et al., 1988. Age and bone mass as predictors of fracture in a prospective study. J. Clin. Investig. 81 (6), 1804. Jiajue, R., et al., 2014. Suppressed bone turnover was associated with increased osteoporotic fracture risks in non-obese postmenopausal Chinese women with type 2

107