Introduction

“Osteo” origin comes from the Greek word “osteon” which means bone; “porosis” comes from “poros” means hole, passage. Osteoporosis is a systemic disease characterized by reduction in skeletal mass associated with bone micro architectural deterioration, which results in increased fracture risk [1, 2]. Continuous formation and resorption follow each other in bone [3, 4]. Osteoporosis has a complex etiology and is considered a multifactorial polygenic disease in which genetic determinants are modulated by hormonal, environmental, and nutritional factors [5]. As lots of environmental factors effect on bone mineral density (BMD); the heritability of BMD at the spine and hip has been estimated to lie between 70 and 85 % [6]. Osteoporosis is a disease increasing rates; characterized by low bone mass and increased fracture risk [7, 8], which effects at some point in their lives as much as 30 % of women and 12 % of the men [2]. Therefore, to estimate the risk of osteoporosis early in life and thus increase the possibility of prevention will be important [8].

To date, research has been carried on several candidate genes that are associated with osteoporosis. The gene that causes osteoporosis in some societies, that has no effect in other society. In Turkish society, the candidate genes VDR, ESR, and COL1A1 genes was investigated, in postmenopausal osteoporotic women an important relationship was found between VDR TaqI, ESR PvuII and COL1A1 Sp1 polymorphism with BMD [9–11].

In our study, research of distribution and effect on BMD in Turkish population especially in Isparta was aimed of the gene polymorphisms COL1A1 Sp1, VDR BsmI, VDR FokI, ESR PvuII, ESR XbaI and CTR AluI that cause osteoporosis in some population.

Materials and methods

Subjects

This study was carried out on 188 people applying for BMD (g/cm2) measurements to Isparta Maternity and Children Diseases Hospital Imaging Center. All subjects (or their responsible next of kin) provided written informed consent and were studied under a protocol approved by the Isparta Local Ethic Committee. All subjects underwent a comprehensive evaluation by a research physician medical history, and questioners. After evaluation of physician patient who has disease affect BMD (malabsorption, liver or kidney disease, endocrine disease, such as hyperthyroidism, hyperparathyroidism, and the presence or metabolic abnormality, surgical menopause) and drug users that affect BMD values (glucocorticoid or hormone replacement therapy), were out of scope of work.

Measurement of BMD

Persons included in the study accepted regardless of male or female and whether they are relatives the lumbar spine (L1–L4) and femoral neck BMD, was measured using dual energy X-ray absorptiometry (DEXA) with a lunar DPX (GE Lunar Corporation, Madison, WI, USA). The unit worked with the CORE 2005 software, system test application was made daily before the measurement. Whether the participants were osteoporosis determined according to the World Health Organization (WHO) criteria.

Sample collection and DNA extraction

About 2 ml aliquots of peripheral blood samples were collected from the subjects and stored in EDTA-coated vacutainers. The blood sample brought Afyon Kocatepe University School of Medicine genetics lab in appropriate circumstances of −20 °C. DNA was extracted from the whole blood samples using Clinical Arrays® MetaBone (Genomica, Spain) DNA isolation kit. The purity and quantity of isolated DNA samples were measured by using spectrophotometer (Nanodrop ND-1000) determination. DNA samples were stored at +4 °C until used. 5 μl DNA (nearly 200 ng total DNA) used for the amplification and the remainder stored at −20 °C.

PCR and genotyping

In vitro reproduction process by PCR method of familiar sequence Genomic DNA region defined as PCR or amplification. Detection is carried out via simultaneous amplification of five genome regions, each of them of a size of 150–250 pb (154, 157, 211, 225, 246), and containing the six targeted SNPs. For this, the temperature cycles programmed on PCR device. Reaction tubes were placed in the PCR device, and the program was launched. Amplification process took about 4 h. PCR products were checked by agorose gel electrophoresis. ФX174 DNA/BsuRI (HaeIII) was used as Marker DNA. Presence of 1353, 1078, 872, 603, 310. 281, 271, 234, 194, 118 and 72 bp DNA fragments in Marker DNA provided us easily monitor our DNA between 246 and 154 bp.

Detection of the PCR amplified products is performed by means of a new technology platform, the array tube (AT). The AT platform is based on a very simple and efficient principle consisting in including a low-density micro array at the bottom of an classical 2 ml tube. The micro array technology allows the simultaneous detection of multiple diagnostic molecular markers. In this case, the detection of each SNP involves a minimum of three probes that will detect the presence of the two possible variants, and a common gene site. The AT considerably simplifies the micro array manipulation steps.

Hybridization of the amplified PCR product was detected by generation of an insoluble precipitate at the sites of the micro array where the amplified products have been captured by the probes. This is achieved by marking the amplified products with biotin during the PCR procedure. Biotinylated products hybridize to their specific probes attached to the micro array surface and become immobilized on the array surface. These immobilized biotinylated products were recognized by the streptavidin of a streptavidin–peroxidase conjugate, thus providing with peroxidase activity to the hybridized products. Peroxidase activity then metabolized 3,3′,5,5′-tetramethylbenzidine (TMB) and produced an insoluble product which precipitated in those places where hybridization occurred.

The processing of the data obtained in each analysis is completely automatic. The reading/analysis equipment provided a report with the results. The system monitor displays a three-column table; the left column displays the various genes, the central column displays the genotype diagnosed for each specific gene, while the right column displays validity determined by DNA and amplification controls. Here for the patient the result of VDR-B was accepted as invalid. The other five gene polymorphisms were evaluated but VDR-B gene excluded from evaluation Table 1.

Table 1 The system monitor displays a three-column table
Full size table

Some patient a few results were invalid. Patient all the results were invalid excluded from the study. As a result, statistical studies were made on 188 patients. Valid and invalid results, shown in Table 2.

Table 2 Validity of results
Full size table

Statistical analysis

In statistical evaluation of data SPSS 16.0 (PASW 16.0) and SPSS 18.0 (18.0 PASW) package programs were used. Categorical measurements number and percentage, numerical measurements were summarized as the mean and standard deviation. In determining the level of significance of the difference between patient and control group’s allele frequencies, two sample t tests (independent samples t test) was used. To determine the relationship between gene polymorphisms and the parameters that affect BMD values regression analysis was used. Level of statistical significance was accepted as P ≤ 0.05 Major effect and 0.05 < P ≤ 0.1 minor effect. Allele frequencies, conformity according to Hardy–Weinberg equilibrium were assessed with the χ2 test.

Results

Total 188 person working group evaluated according to WHO criteria, 78 (41.5 %) with osteoporosis and 110 (58.5 %) people were determined as a control group. Working within the scope of 188 people, genotypes designated, allele frequencies were calculated. Allele frequencies, conformity according to Hardy–Weinberg equilibrium were assessed with the χ2 test and P values shown as P1 in Table 3. If the obtained P value is higher than 0.05, our data does respect the equilibrium and might thus be considered as a good sample for our study.

Table 3 Working group genotypes and Allele frequencies
Full size table

Generally this working group genotype distribution also show us that the appropriate in Hardy–Weinberg law. In determining the level of significance of the difference between Patient and control group’s allele frequencies, two sample t tests (independent samples t test) was used and P values shown as P2 in Table 3. This P value is higher than 0.05 means no difference between patient and control group’s allele frequencies.

We think that genotypes association with femoral and lumbar vertebral BMD measurements is more important than the difference between patient and control group’s allele frequencies than we evaluate the genotypes association with femoral and lumbar vertebral BMD measurements.

The effects of gene polymorphisms on femoral and lumbar vertebral BMD

In these analyses, the separation into control and patient groups has not been done since the analysis of the results BMD (g/cm2) of the individuals taking part in this study had been done. Whether there is a difference between the BMD of different polymorphic structures is evaluated through regression analysis by determining which one they have out of three different polymorphisms for each gene (Table 4). At the end of the result, P ≤ 0.05 as major influential and 0.05 < P ≤ 0.1 as minor influential are accepted.

Table 4 The analysis of the effects of different polymorphisms of 188 individuals as a part of the study on BMD through regression analysis
Full size table

It has been seen that VDR-B gene polymorphisms affect femur bone density in a extremely meaningful way (P ≤ 0.05) and are one of the major factors while VDR-F gene polymorphisms affect significantly on the borderline and are one of the (0.05 < P ≤ 0.1) minor factors. It is seen that the effect of these gene polymorphisms on vertebral bone density is not meaningful.

In our study, the statistically meaningful relation of COL1A1, CTR, ESR1P and ESR1X genes that we have been working in the overall population has not been found neither on femur BMD nor lumbar vertebral BMD. The P levels has been found higher than 0.1 (P > 0.1).

The effect of VDR-B, VDR-F gene polymorphisms on BMD

The femur BMD averages of the ones who have each of VDR-B gene polymorphisms (BB, Bb, bb) that affect femur bone density in a notably meaningful way (P = 0.013) are shown in the Table 5. Accordingly, the results of the ones with BB and bb homozygote are low while this is higher for the ones with heterozygote. VDR-B gene heterozygosity is a protective factor against osteoporosis.

Table 5 The averages of femur BMD of the ones whose VDR-B gene polymorphisms are BB, Bb, bb and VDR-F gene polymorphisms are Ff, Ff, ff
Full size table

In this research group, it is seen in the Table 5 that the homozygous individuals who have FF genotype of VDR-F polymorphisms which affect femur bone density significantly on the borderline; that is, are among the minor factors (P = 0.082), have lower femur BMD levels than heterozygotes (Ff) and ff homozygotes. In addition, it is important that the femur BMD levels of ff homozygotes and heterozygotes (Ff) are close to each other. These results have shown us that the homozygous ones who have F allele are inclined to osteoporosis.

The effects of the factors except genes on BMD levels

The other possible factors except genes, which could affect BMD levels, have been analyzed through regression analysis to see whether they affect the bone mineral densities or not. As seen in the Table 6, it is seen that sex (P = 0.013), age (P = 0.004), body mass index (BMI) (P ≤ 0.001), the existence of an osteoporotic family member (P = 0.015) and regular nutrition (P = 0.011) have influenced the lumbar vertebral BMD levels in a meaningful way. As for femoral BMD levels, age (P ≤ 0.001), BMI (P ≤ 0.001) and regular nutrition (P = 0.019) have affected in a meaningful way.

Table 6 The analysis of the effects of the 118 individuals’ features except genes through regression analysis
Full size table

Both the averages of lumbar vertebral BMD and femoral BMD of men are higher than women. Though there is such a difference between women and men, the difference in lumbar vertebral BMD has been found to be meaningful (P = 0.013) while the difference in femoral BMD is not meaningful (P = 0.330).

Age is a critical factor in osteoporosis. It influences both femur and vertebral BMD levels significantly for femur (P = 0.000), for vertebra (P = 0.004). As age goes up and BMD levels go down, the risk of osteoporosis increases.

The BMI is also one of the factors that affect BMD. As BMI increases, the BMD levels go up, too.

The femur BMD levels of 40 individuals who have an osteoporotic family member and know that an osteoporotic family member exists are close to the levels of the individuals who do not have any osteoporotic family member while the BMD levels of lumbar vertebras is low in a meaningful way (P = 0.018) in the ones who have an osteoporotic family member.

Regular nutrition is effective in a meaningful way on femoral and lumbar vertebral BMD levels (P = 0.011 and 0.019). It is seen that regular nutrition delays the formation of osteoporosis and both femoral and lumbar vertebral BMD levels of the ones being nourished regularly is much higher than the ones being nourished irregularly.

Doing sport regularly (P = 0.058) and drinking alcohol (P = 0.099) influence femur BMD levels as a minor factor, but not the lumbar vertebral BMD levels. In our study, the effects of smoking and being subject to sunlight for both femur and lumbar vertebral BMD could not be found out and the P levels has been found higher than 0.1 (P > 0.1).

Discussion

Many studies have been done so far on the candidate gene related with osteoporosis. In some studies, the gene which has an effect on osteoporosis has been found ineffective. When the two t test samples, whose allele frequency is independent, of control and experimental groups are compared, there is no meaningful difference in allele frequency of all COL1A1, CTR, ESR1P, ESR1X, VDR-B and VDR-F gene polymorphisms. In the study carried out with ESR1P, ESR1X in Adana [11] the same results had been gained as in our study. Two values, T score 2.5 and below for the osteoporotic patients and T score above 2.5 and for the control groups have been used to find out the effects of these genes on BMD levels and to compare the allele frequencies of control and experimental groups. However, it has been thought that comparison of the femoral and lumbar vertebral BMD levels of measurement separately for each gene would be more meaningful so as to show how femoral and lumbar BMD measure levels belonging to each individual are affected by the genotype structure. In literature review [12–16] the relation between BMD levels and genotypes has been examined. As for our study, the relation of femoral and lumbar vertebral BMD levels of the genotypes has been carried out by regression analysis.

COL1A1

In our study we have found out that the COL1A1 gene polymorphisms do not have any effect on BMD levels of femur and lumbar vertebral. In 2000 McGuigan et al. [17] studied four region polymorphisms of COL1A1 gene which are Sp1, Msp1 RsaI and MnlI and found any polymorphisms were not related to osteoporosis. Besides this researches supporting our results which are related with COL1A1 gene, in some studies which have led us to include COL1A1 gene in our study at the beginning of our study the relation between BMD and COL1A1 gene polymorphism has been stated. One of the first publications on how related COL1A1 gene Sp1 polymorphism is with osteoporosis belongs to Grant et al. [12] found out that heterozygotes (Ss) whose Sp1 polymorphic area is G/T have lower BMD in a meaningful way and BMD in T/T homozygotes (ss) is low. In the studies done by Şimsek et al. [10] in an effort to evaluate the effect of COL1A1 Sp1 polymorphisms on BMD of hormone-replacement therapy among the patients, it has been gained from the results of COL1A1 Sp1 connection area polymorphism that 79 individuals (71.2 %) are SS, 30 individuals (27.0 %) are Ss and two individuals (1.8 %) are ss homozygous. Moreover, when the Ss heterozygous individuals are compared with the SS ones, the ones with Ss genotypes have been found to have low BMD levels in both lumbar spine and femur neck.

CTR

Any relation between the CTR polymorphisms and BMD levels has not been found out in the study carried out to examine the CTR gene among the total population of 188 individuals living in Isparta and in the environs of Isparta. Masi et al. [18] reported that there was not a statistically considerable difference though the number of normal women with C/C (AA) genotypes was higher than the osteoporotic women. In the study which the relation of lumbar vertebral and femur BMD levels of 663 postmenopausal and 52 perimenopausal with CTR gene was examined in Italy in 2000 [19] they claimed that CTR gene had more effect on young women; therefore, CTR genotypes were more effective in the formation of peak bone mass rather than bone loss while ageing.

ESRX and ESRP

In our study, we have come to the conclusion that ESRX and ESRP polymorphisms do not influence the femur and lumbar vertebral BMD levels. In some studies the relation between ESR gene and BMD is seen [20, 21] while in some studies any relation is not observed [22, 23].

VDR-B

The results gained from our study have been analyzed statistically through regression analysis. According to the P levels gained as a result of this analysis, it is seen that VDR-B gene polymorphisms have affected femur bone density in a highly meaningful way (P = 0.013); but their effect on vertebral bone density is not meaningful. The averages of the ones who have each VDR-B gene polymorphisms (BB, Bb, bb) show that the femur BMD levels of the ones with BB (0.873 g/cm2) are low while heterozygous ones have the highest (0.915 g/cm2). According to the BMD levels, which are close to low, of the ones with homozygote bb (0.888 g/cm2) genotypes, it can be said that the VDR-B gene heterozygosity is a protective factor against osteoporosis, and BB homozygotes are inclined to osteoporosis.

The statistical analyses in our study indicate that VDR-F gene polymorphism affects femoral BMD levels in a meaningful way (0. 05 < P ≤ 0. 1), but the effect on vertebral bone density is not meaningful. It is also important that the individuals with FF polymorphism have lower femur BMD levels than the ones with Ff and ff, and the femur BMD levels of ff homozygotes and heterozygotes (Ff) are close to each other. These results show us that the individuals with FF homozygote genotypes are inclined to osteoporosis.

Morrison et al. [24] found that VDR BB genotypes are related to low BMD. However, Hustmyer et al. [25] in their twin-studies in 1994 could not find any relation between BMD and VDR polymorphisms.

Ferrari et al. [26] found that the BB homozygotes bearing f allele have lower BMD in a meaningful way. Gennari et al. [22] found a relationship between VDR polymorphisms and lumbar vertebral BMD, osteoporosis. As Bulca [11] said, these inconsistencies in the studies can be based on low statistical capacity, the variations among ethnic groups, age, the menopause condition and environmental factors.

The factors except genes

Melton et al. [27] stated that osteoporosis is seen less among men because of the excess of their skeleton surface and maximum bone mass, the late inception of bone loss and advancing more slowly, the nonexistence of menopause and accompanying quick bone loss. Our findings are proper for the literature, too.

Ageing is an important factor in osteoporosis. It affects both femur and vertebral BMD levels in a meaningful way. In the overall population [for femur (P = 0.000), for vertebra (P = 0.004)] Kanis et al. [1] stated that osteoporosis has usually an effect on people in later years and seen more commonly among women after menopause, and men encounter osteoporosis in later years than women.

In our study BMI has been found one of the factors that affect both femoral (P ≤ 0.001) and lumbar vertebral (P ≤ 0.001) BMD levels in a meaningful way. According to the results we have found, as BMI increases, BMD levels also go up. These findings are also compatible with the literature. Armamento-Villareal et al. [28] found in their vertebral bone density study done among 63 premenopausal women whose ages changed from 19 to 46 that the individuals who had low vertebral bone density compared to the ones having normal bone density had more osteoporotic family stories. The study carried out by Paker et al. [29] indicated that there was a significant correlation between the body mass, femur neck and L2–L4 BMD levels, and body mass has a protective effect against the osteoporosis.

In our study the existence of osteoporotic family member has been found to influence the lumbar vertebral BMD levels in a meaningful way (P = 0.015). While the femur BMD levels of the 40 individuals who have an osteoporotic family member or know an osteoporotic family member exists are close to the levels of the ones who do not have any osteoporotic family, the lumber vertebral BMD levels of the ones who have an osteoporotic family member (0.934 g/cm2) and who do not have any osteoporotic patient in their family (1,021 g/cm2) are found to be significantly low.

Besides many studies claiming that the inadequate intake of calcium is a risk factor for osteoporosis [30–32], there also exist many contradictories. Cordain et al. [33] and Maclennan [34] said that the period which milk was consumed most all over the history of Great Britain islands was the moment osteoporosis occurred most commonly. In our study it has been found that regular nutrition affects both femoral BMD levels (P = 0.019) and lumbar vertebral BMD levels (P = 0.011) in a meaningful way. The both femoral (0.913 g/cm2) and lumbar vertebral BMD levels (1,031 g/cm2) of the ones having a balanced diet are higher than the ones who do not have a balanced diet (femoral: 0.895 g/cm2 and vertebral: 0.996 g/cm2).

The relation of lumbar vertebral and femur BMD levels with VDR-F, VDR-B, COL1A1, ESR1X, ESR1P and CTR polymorphisms and the clinical and definitive features of the participants has been sought out. It has been seen that COL1A1, CTR, ESR1P and ESR1X gene region polymorphisms do not affect the BMD levels in a meaningful way statistically; VDR-B and VDR-F gene polymorphisms are effective especially in femur BMD levels; moreover, the tendency towards osteoporosis in BB and FF genotypes is much higher. The sex and BMI among the factors except gene are proven to be the most significant parameters that affect BMD levels.

In the studies the various results of COL1A1, CTR, VDR-F, VDR-B, ESR1X and ESR1P gene polymorphisms and BMD levels have been gained, thereby depending on ethnic groups and population. To conclude, we believe that the number of our study group is not sufficient to give exact results about the genotype structure of our society, and increasing the number of patients in the future studies will consolidate this information.

Relationship between VDR-F, VDR-B, COL1A1, ESR1X, ESR1P and CTR gene polymorphisms and BMD was investigated among various populations, and different results were obtained. Therefore, the relationship between these polymorphisms and BMD will be well understood doing new researches. In this study, VDR-F, VDR-B, COL1A1, ESR1X, ESR1P and CTR gene polymorphisms compared with lumbar vertebra and femoral neck BMD values in Turkish population.