FokI Polymorphism in the Vitamin D Receptor Gene (VDR) and Its Association with Lumbar Spine Pathologies in the Italian Population: A Case-Control Study

Alterations in vitamin D homeostasis, mainly involving its nuclear receptor (VDR), could have a role in the pathophysiology of the spine. The association between VDR polymorphisms and spine disorders has been analyzed in different ethnic groups, focusing on the functional FokI polymorphism. However, so far, inconsistent findings were reported. The aims of this study were to evaluate, in the Italian white population, the VDR FokI polymorphism frequencies distribution in subjects with clearly defined lumbar spinal pathologies compared to asymptomatic controls and to analyze the interplay of genetic and conventional risk factors. Using a case-control design, 267 patients with spinal disorders and 220 asymptomatic controls were enrolled, evaluating their exposition to putative risk factors. Patients’ clinical assessment was performed by Magnetic Resonance Imaging. FokI polymorphism (rs2228570) was detected by PCR-RFLP. Genotypes were designated by a lowercase letter (f allele, T nucleotide) for the presence of the restriction site and by a capital letter (F allele, C nucleotide) for its absence. Family history, higher age and BMI, exposure to vibration, physical job demand, smoking habit and lower practice of leisure physical activity were associated with spinal disorders. The FF genotype and F allele represented approximately 2-fold risk factors to develop discopathies and/or osteochondrosis concomitant with disc herniation, while f allele was protective. In conclusion, the link we observed between VDR FokI variants and specific lumbar spine pathologies suggests that spinal tissue degeneration is influenced by the genetic background. Future studies should evaluate the signaling pathways involving alterations in VDR and influencing the development and/or progression of spine disorders.


Introduction
Low back disorders, in particular disc herniation, which represents by far the most prevalent pathology causing pain and sciatica, constitute an important source of disability and one of the most cost-intensive health problems [1]. In Western countries they represent the most common musculoskeletal diseases; it is estimated that 15-20% of adults experience back pain during a single year and around 50-80% have at least one episode during their lifetime [2].
Lumbar disc degeneration (LDD) is considered a primary cause of low back pain (LBP) [3,4]. Many environmental and behavioral risk factors, including age, gender, weight, occupational load, smoking and exposure to vehicular vibration are likely to contribute to the genesis or to the progression of spinal degeneration and pain onset [5,6,7,8,9,10]. In particular, occupational exposures to heavy physical loads, prolonged sitting or nonneutral work postures and vehicle driving have been involved in the disc degeneration processes [11] and considered as the primary source of the mechanical factors damaging the spine [12].
However, some epidemiologic studies and reports among families and twins highlighted that disc herniation, and particularly sciatica, may be explained to a large degree by hereditary factors with apparently a relatively minor effects of environmental and behavioral risk factors [13,14,15,16]. These findings supported the idea that there is a familiar predisposition for development of disc degeneration disorders and that such pathologic conditions may be, at least in part, genetically determined [14,17,18,19,20,21,22,23].
Vitamin D receptor gene (VDR) has been studied as genetic factor putatively predisposing to spine pathologies since 1998 [24,25]. Several single nucleotide polymorphisms (SNPs) have been identified in the VDR sequence, between them FokI (rs10735810, merged into rs2228570) represents an independent polymorphic site [26]. It is a C/T transition polymorphic site located in the VDR start codon, affecting the structure and the function of the encoded protein. The allelic variants of this polymorphism code for structurally different receptor proteins from a 424 aminoacids wild-type (F allele, C) to a 427 aminoacids long protein (f allele, T). The short and long protein forms are associated to a different ability to induce transcription of vitamin D-dependent genes [27,28,29]. Consequently, studies concerning the possible association of this SNP with disc degeneration may be particularly interesting for their potential biological significance.
Wide evidences support the notion that the vitamin D endocrine system is involved in the modulation of different biological processes, including skeletal metabolism, immunological response, proliferation and differentiation of a wide variety of cell types [30,31]. More recently some studies detected the presence of VDR also in the disc cells, highlighting the prominent role of vitamin D active metabolites in the regulation of cells proliferation, matrix genes expression and specific cytokines and proteins production [32,33].
The pleiotropic effects of vitamin D and its involvement in bony and disc metabolisms could explain why alterations in vitamin D homeostasis could be associated to several pathological conditions of the intervertebral disc (IVD).
The association of FokI polymorphism in VDR with hernia, disc degeneration [25,34,35,36] or lumbar spinal stenosis [37] and with occupational vibration, leading to the development of LDD [38,39], was analyzed in different ethnic groups. However, so far inconsistent findings were reported [40]. This can derive, at least in part, from the lack of a clear definition of the lumbar spine pathological phenotypes and/or by the poor definition or differences associated to the specific ethnic group examined.
To our knowledge, there are no studies investigating the association of FokI polymorphism in VDR and specific lumbar spine pathologies in the Italian white population.
Based on these evidences, the aims of this study were: i. to evaluate the VDR FokI alleles frequencies distribution in subjects with specific lumbar spine pathologies in comparison with asymptomatic controls in the Italian population; ii. to analyze the interplay of genetic and conventional, behavioral and environmental factors in the development of lumbar spine pathologies.

Ethics Statement
The study was approved by the Institutional Review Board ASL Città di Milano. The methods used in this study were in accordance with the Helsinki Declaration of 1975 as revised in 1996.

Subjects
Using a case-control design, a total of 487 Italian white subjects, age range 18-65 years, were enrolled after having signed a written informed consent. Inclusion criteria for cases were the presence of a lumbar spine pathology confirmed by Magnetic Resonance Imaging (MRI), while inclusion criteria for controls were the absence of LBP or confirmed severe or chronic spine pathologies. The concomitant presence of other orthopedic diseases such as osteoarthrosis, hip, knee and hand osteoarthritis, osteoporosis was recorded.
Exclusion criteria for both cases and controls were presence of a pathologic condition such as cervical discopathies, scoliosis, fibromyalgia, pregnancy at study enrollment, and chronic diseases like diabetes, cardiovascular diseases, malignancies, lupus erythematosus, and rheumatoid arthritis.
The study included 267 consecutive patients (hospitalized or outpatients) with lumbar spine disorders enrolled for the European Genodisc Project, from May 2009 to January 2013, at the Orthopedics and Traumatology Department of I.R.C.C.S. Istituto Ortopedico Galeazzi (Milan, Italy) by the same clinician. A total of 220 asymptomatic controls were enrolled from January 2011 to January 2013 among healthy volunteers, blood donors or subjects hospitalized for anterior cruciate ligament injuries or hallux valgus surgery.

Clinical Assessment
Assessment of lumbar spine disorders was performed by an expert clinician in spinal diseases by contrast-enhanced MRI 12 scans of the lumbar spine with a 1.5 T scanner (Avanto, Siemens, Erlangen, Germany). Diagnosis of disc herniation was performed when patients presented disc material protrusion/extrusion beyond the posterior margins of the adjacent vertebral bodies [41] (Figure 1a). Disc herniations were often associated with discopathies and/or osteochondrosis (Figure 1b). Diagnosis of discopathies was performed in presence of degenerative changes of the IVD, while diagnosis of osteochondrosis was performed in presence of degenerative process involving primarily the vertebral bodies structures limiting the disc (disc narrowing, subchondral sclerosis, wavy endplates, osteophytes and Schmorl's node) [42] ( Figure 1c). Spinal stenosis was diagnosed in presence of a narrowing of the central spinal canal, lateral recess or intervertebral foramina [43,44] (Figure 1d). Finally, patients with degenerative spondylolisthesis presented an acquired anterior displacement of a vertebra over the subjacent one (Figure 1e), due to degenerative changes, without an associated disruption or lysis of the pars interarticularis [45], that is present in patients with lytic/isthmic spondylolisthesis [44]. Stenosis and spondylolisthesis, which are the more concerning structural degenerative spine changes, were often associated ( Figure 1f).
Since the close linkage between discopathies, disc herniation and osteochondrosis, a further subgroups division from A to D (not mutually exclusive) was performed to better analyze the association between these pathologies and VDR FokI genotypes in our cohort of patients: Subgroup A, comprising all herniation cases i. e. Subgroup 1 grouped with Subgroup 2 (total n = 176); Subgroup B, including all discopathies and/or osteochondrosis regardless of herniation, i.e. Subgroup 2 grouped with Subgroup 3 (total n = 127); Subgroup C, comprising all discopathies concomitant with disc herniation (n = 46) grouped with subjects with discopathies alone (n = 18) (total n = 64); and Subgroup D, comprising all osteochondrosis concomitant with disc herniation (n = 37) grouped with subjects with osteochondrosis alone (n = 13) (total n = 50).

Conventional, Behavioral and Environmental Factors Evaluation
A medical history, including possible low back symptoms or spine surgery, and a questionnaire, reporting the exposition to individual behavioral, environmental, occupational and physical activity putative risk factors were obtained from each subject. The collected information included medical history of family (parents, brothers or sisters), the smoking habit, the job physical demand for the majority of the working years (evaluated by the following score: 0 = sedentary; 1 = light; 2 = medium; 3 = heavy), the hours/day spent driving or as a passenger in motorized vehicles (exposure to vibrations) and, finally, over the past year, the times a week (outside work activity) spent in vigorous physical activity or leisure exercise activities involving twisting, bending or lifting (indicated thereafter collectively as leisure physical activity).

Determination of Genotypes
Blood samples were collected from the antecubital vein with evacuated ethylenediamine tetra acetic acid (EDTA) tubes (Vacutainer Tubes, Becton-Dickinson, Franklin Lakes, NJ, USA) from the 267 cases and 220 controls. Genomic DNA was extracted from white blood cells according to the procedure of the DNeasy Midi kit (Qiagen, Duesseldorf, Germany). Polymerase chain reaction and restriction fragment length polymorphism (PCR-RFLP) methods were applied to detect the FokI polymorphism of VDR. Genomic DNA was amplified using PCR. At first DNA was denatured at 95uC for 5 minutes. Standard PCR conditions were as follows: 94uC for 1 minute, annealing temperature 63uC for 1 minute and 72uC for 2 minutes for 35 cycles and finally 96uC for 1 minute and 72uC for 5 minutes.
The FokI polymorphism of VDR was studied using previously tested primers [46].
The resulting 265 bp DNA fragment was digested with FokI restriction enzyme (Euroclone, Milano, Italy) generating two fragments of 196 and 69 bp only in presence of the f allele (T). DNA fragments were separated on poliacrylamide gel. Randomly chosen 30 samples' gel results were confirmed by DNA sequencing.
Genotypes were designated by a lowercase letter (f allele, T nucleotide, mutated) for the presence of the restriction site and by a capital letter (F allele, C nucleotide, wild-type) for its absence ( Figure 2).

Statistical Analysis
Kolmogorov-Smirnov test was used to assess the normal data distribution. Student's t-test or Mann Whitney test were used to assess the differences between the frequency distributions of variables in cases and controls. Odds ratios (ORs) were calculated to set the association between alleles or genotypes and risk of spine pathologies in cases and controls and specific Subgroups of patients. Chi squared or Fisher's exact test P values were reported as appropriate. Logistic regression was used to evaluate effects of confounders by obtaining adjusted ORs and 95% CIs for genotypes and alleles. Adjusted analysis included conventional risk factors: age, BMI, family history, smoke, physical job demand and exposure to vibrations. Leisure physical activity was not included as confounding because this kind of activity may derive both by personal habits and by absence of pain.
Significance level was held at 0.05. At variance, P values #0.10 were considered as a tendency. Statistical softwares used were GraphPad Prism version 5.00 (GraphPad software, La Jolla, CA, USA) and SPSS version 14.0 (SPSS Inc., Chicago, IL, USA).

Characteristics of Subjects
The characteristics of the overall population of cases and controls, including age, gender, BMI, spine surgery, family history, smoking habit, job physical demand, exposure to vibrations, practice of leisure physical activity and presence of other orthopedic conditions were shown in Table 1. Among the cases, there were more males (149/267, 55.8%) than females (118/267, 44.2%), while among controls there were almost equal numbers of males (106/220, 48.2%) and females (114/220, 51.8%). Due to the study design, surgery for disc pathologies (39% of cases) and presence of other orthopedic conditions (12.7% of cases) were present only in the cohort of cases.

Influences of Conventional, Behavioral and Environmental Risk Factors
Associations between lumbar spine pathologies and putative conventional risk factors were reported as continuous variables in all cases and in mutually exclusive Subgroups 1 to 4 in Table 1. In the overall cohort of cases higher age (P,0.001), higher BMI (P = 0.001), family history (OR = 3.02; 95%CI = 1.94-4.68; P, 0.001), smoking habit (OR = 1.54; 95%CI = 1.08-2.21; P = 0.020), stronger job physical demand (P = 0.001), higher exposure to vibration (P = 0.013) were all significantly associated with the development of lumbar spine pathologies. Controls subjects practiced leisure physical activity more frequently than the pathological subjects (OR = 2.33; 95%CI = 1.61-3.37; P,0.001).
A stronger job physical demand was observed in Subgroup 1 (P = 0.009) and Subgroup 2 (P = 0.023), with a tendency for Subgroup 3 (P = 0.064) and Subgroup 4 (P = 0.091). Finally, a higher exposure to vibrations was registered in Subgroup 2 (P = 0.001) in respect to controls.
Controls subjects practiced leisure physical activity more frequently than the subjects in Subgroup 1 (OR = 1.82; 95%CI = 1.

VDR Genotypes and Alleles in Controls and Cases
In our total sample of 487 Italian white subjects the frequency of FF homozigotes was 42.3% (206/487), Ff heterozigotes was 46.9% (219/487) and ff homozigotes was 12.9% (62/487). F allele had a frequency of 64.8% (631/974) and the f allele had a frequency of 35.2% (343/974). The observed genotype frequencies were consistent with Hardy-Weinberg equilibrium (X 2 = 0.102, P = 0.75). Table 2 reported frequencies of FokI genotypes and alleles in cases and controls, including crude and adjusted ORs and 95%CIs. Not significant differences in the frequencies distribution of both genotypes and alleles were observed in overall subjects, the wild homozygous FF genotype was present in 43.8% of cases versus 40.5% of controls, while the minor ff homozygous genotype was found in 14.5% controls versus 11.2% of cases, and the heterozygous Ff genotype had the same frequency between cases and controls (45.0%). F allele had a frequency of 66.3% in cases versus 63.0% of controls, while f allele had a frequency of 37.0% in controls versus 33.7% of cases. Neither controls (X 2 = 0.273, P = 0.60) nor cases (X 2 = 0.0085, P = 0.93) deviated from Hardy-Weinberg equilibrium.

Association of Specific Lumbar Spine Pathologies and VDR Genotypes and Alleles
The frequencies distribution of FokI VDR genotypes and alleles in different specific pathological subgroups (1-4 and A-D) and in controls subjects, with the relative crude and adjusted ORs and 95%CIs, were reported in Tables 3 and 4, respectively. Concerning genotypes (Table 3), the only significant association  Table 1. Characteristics of the subjects recruited and influence of risk factors of lumbar spine pathologies. 1 5 patients had missing information about intensity of physical demand at work, thus a total of 262 data were available. Physical job demand score used: 0 = sedentary; 1 = light; 2 = medium; 3 = heavy. 2 1 patient had missing information about leisure physical activity per week, thus a total of 266 data were available. 3 Orthopedic conditions included: osteoarthrosis, hip, knee and hand osteoarthritis, and osteoporosis.
Subgroup 1 = patients with disc herniation alone; Subgroup 2 = patients with discopathies and/or osteochondrosis associated with disc herniation; Subgroup 3 = patients with discopathies and/or osteochondrosis without herniation; Subgroup 4 = patients with stenosis and/or spondylolisthesis. In subjects with discopathies and/or osteochondrosis without herniation (Subgroup 3) the Ff genotype showed a tendency to be a risk factor (adjusted OR = 2.00; 95%CI = 0.96-4.16; P = 0.063) and the ff genotype had a tendency to be protective (adjusted OR = 0.30; 95%CI = 0.07-1.24; P = 0.096). No other significant finding was observed even grouping Subgroups 1, 2 and 3 altogether (thus excluding the Subgroup 4).

Discussion
To our knowledge, this is the first study that evaluated and showed an association between VDR FokI variants and specific spine pathologies in the Italian white population and the largest study showing also the concomitant influences of conventional risk factors.
Moreover, we observed that the VDR FokI polymorphism in our Italian sample was in Hardy-Weinberg equilibrium as previously found in other European populations [25].
The association between the presence of polymorphisms in the VDR and lumbar spine pathologies is a debated topic [24]. In our study, considering the very broad category of patients with lumbar spine disorders, the FokI polymorphism was not associated with disease risk. However, this polymorphism represented a risk factor to develop discopathies in general, and particularly discopathies and/or osteochondrosis concomitant with disc herniation.
Three previous studies from other authors reported an association between the FokI polymorphism in VDR and specific signs of disc degeneration in Turkish [34], Brazilian [35] and Finnish [25] populations, with subjects having Ff and ff genotypes showing a predisposition towards worse phenotypes.
By contrast, other studies found no association between FokI genotypes and disc pathologies. Specifically, no association was observed for disc herniation or lumbar spinal stenosis in the Finnish population [25,37]; nor for LDD in a Norway case/ control study [36] and neither for osteophyte formation without disc degeneration in a cohort of elderly Japanese males and females with LBP [49].
The comparison of data present in the literature is particularly difficult especially because of the study design and ethnic differences in the various research studies. Moreover, the absence of a standardized definition of pathological phenotypes hampers the comparisons and reliable interpretations of the reported data.
In our work, we classified our cases by means of the pathological features evidenced throughout detailed objective evaluation by MRI. This approach allowed us to subgroup patients accordingly to specific lumbar spine pathologies having a defined clinical significance. To evaluate the association of VDR FokI genotypes/ alleles, we first analyzed the broad sample of all patients with lumbar spine disorders, then we subgrouped them in 4 mutually exclusive subgroups (1 to 4). Moreover, we analyzed subgroups of all patients having a condition in common like hernia, discopathy, osteochondrosis, regardless of other concomitant disorders (Subgroups A to D). Considering specific subgroup of patients i.e. those suffering from discopathies and/or osteochondrosis concomitant with herniation, and in general all patients having discopathies, the FF genotype was associated with a 2-fold increased disease risk, also after adjusting for conventional, behavioral and environmental risk factors. The F allele was associated with a 1.5 to 1.8-fold increased risk in all patients having discopathies, in all patients having osteochondrosis, and in patients having discopathies and/ or osteochondrosis concomitant with herniation. On the contrary, f allele seemed to be protective for these pathological phenotypes. We confirmed these results after adjusting for conventional risk factors.
In general, genetic risk factors may interact with behavioral and environmental factors in enhancing the development of lumbar spine pathologies. In this context, the results present in the literature about a possible interplay between VDR FokI genotypes, occupational load exposure and exposition to whole-body vibration were controversial [38,39].
In our study, regarding environmental risk factors such as exposure to vibrations and job physical effort, we observed an association between lumbar spine pathologies, higher number of hours/day exposure to vibration and higher physical job demand. On the contrary, we noted that the practice of leisure physical activity was inversely associated to lumbar spine pathologies. Due to the study design, we cannot assess if this was a really protective behavior or an indication of absence of major low back pain, concomitant with personal habits. Among the other putative conventional risk factors analyzed, we observed that in our cohort of cases family history, higher age, overweight and smoking habit were associated with risk for lumbar spine pathologies. Overall, our findings highlighted that subject voluntary behaviors in addition to environmental factors are major determinants in lumbar spine pathologies.
A limitation of this study is represented by the difference in the mean age of the recruited cases and controls. We found difficulties in finding over 50 years old healthy subjects and, thus, in the future we would like to enlarge the group of controls, trying to match this difference and to confirm our results in a larger cohort of subjects.
Additionally, an increase in the number subjects with stenosis and/or spondylolisthesis (Subgroup 4), could be useful to perform a better evaluation of particular features of this subgroup.
In conclusion, the conventional, behavioral and environmental factors analyzed in this study represented determinants of risk for the development of lumbar spine pathologies in general. On the contrary, finding of genetic associations required objective characterization of lumbar spine disorders. Particularly, we evidenced that personalized evaluation through imaging techniques of each patient is necessary to determine the appropriate subgroup belonging. Our results showed that patients with FF homozygous genotype are at risk to develop discopathies in general and discopathies and osteochondrosis in association with disc herniation, independently by the influence of the conventional behavioral and environmental determinants of risk. It is of note that in our study the FF homozigosity was not a risk factor for simple herniation. Notably, the F allele was an independent risk factor for all discopathies, discopathies and osteochondrosis, discopathies and osteochondrosis combined with herniation, but not simple herniation, stenosis and spondylolisthesis. Additionally, in an adjusted analysis the F allele was a 1.7-fold risk factor for osteochondrosis.
Based on the assumption that the wild F allele is producing a more transcriptionally active receptor than the f allele, it appears that enhanced vitamin D final effects are favoring discopathies and the severe progression of discopathy and/or osteochondrosis to herniation. Interestingly, a very recent study performed in 140 Iranian subjects with diabetes evidenced that the VDR ff genotype may be regarded as ''low responders'' to vitamin D intake [50]. Thus, it is tempting to speculate that a nutrigenic approach based on specific genotypes may be needed to protect patients with specific lumbar spine disorders.
Finally, the lack of significant finding for the association of simple herniation with the genetic background could reflect the accidental/traumatic origin of this condition and/or the necessity to explore different genetic polymorphisms for this specific disorder.