Association of vitamin D receptor (Fok-I) polymorphism with the clinical presentation of calcium urolithiasis
Ming-Tsang Wu, Department of Family Medicine, Kaohsiung Medical University Hospital, TzYou 1st Road, Kaohsiung 807, Taiwan.
To investigate the effect of the vitamin D receptor (VDR) FokI polymorphism on the clinical presentation of calcium urolithiasis, as a FokI polymorphism in the VDR gene was recently reported to be associated with calcium metabolism disorders.
PATIENTS, SUBJECTS AND METHODS
In all, 235 patients with calcium urolithiasis and 231 age- and sex-matched healthy controls were recruited from Kaohsiung Medical University Hospital between June 2003 and February 2005. Clinical information on the age at first onset, stone episodes, stone severity and presence of family history were collected from patients with stones. Any VDR FokI polymorphism was detected using polymerase chain reaction-based restriction analysis.
The frequency of VDR FokI genotypes between the patients and the healthy controls was not significantly different. However, among patients, those with the FF genotype had a significantly higher risk of having more stone episodes (adjusted odds ratio 2.15, 95% confidence interval 1.02–4.54, P = 0.044) and were younger at the first onset (3.23, 1.08–9.63, P = 0.036) than those with the ff genotype.
The VDR FokI polymorphism might be important in the clinical presentation of patients with calcium urolithiasis, especially for the frequency of stone episodes and age at first onset, although it is not associated with the formation of stones.
vitamin D receptor
Kaohsiung Medical University Hospital
restriction fragment length polymorphism.
Urolithiasis is a multifactorial disease influenced by environmental, hormonal and genetic factors . In developed countries the prevalence rate of urolithiasis is 4–20%. Calcium-containing stones, including mainly calcium oxalate or calcium phosphate, account for 70–80% of urolithiasis .
The 1,25(OH)2D3/vitamin D receptor (VDR) system is central in calcium homeostasis and skeletal metabolism . Overexpression of VDR in the duodenum and kidney cortex has been found in genetic hypercalciuric stone-forming rats . In humans, a close link between idiopathic calcium stone formation and the microsatellite marker D12S339 near the VDR locus was also reported . These findings suggest that the VDR gene might also be important in the development of urinary stone disease.
The VDR gene is located on chromosome 12q12–14 and spans nine exons . Several polymorphisms of this gene have been identified and some of them, e.g. FokI and BsmI, were associated with abnormal bone turnover and calcium metabolism [7–11]. The VDR FokI polymorphism is a thymine/cytosine polymorphism and is located at the first two potential start (ATG) codons at the 5′-end of VDR. The VDR FokI polymorphism was thought to be functionally significant [12–14]. Although several studies investigated the correlation between VDR FokI polymorphism and urolithiasis, the results were not consistent [15–19]. Therefore, we conducted a hospital-based case-control study to investigate the association of FokI polymorphism with urolithiasis in a Taiwanese population.
PATIENTS, SUBJECTS AND METHODS
Between June 2003 and February 2005, 235 patients (161 men and 74 women, mean age 50.1 years, sd 12.3, range 20–85) with upper urinary tract stones treated at the Kaohsiung Medical University Hospital (KMUH) were enrolled for the study. All patients were confirmed to have stones by ultrasonography or radiography. No patients had radiolucent stones on X-ray or had suspected cystine or uric acid stones by clinical evaluation. An infrared spectroscopy (Spectrum RX I Fourier Transform-Infrared System, PerkinElmer, USA) was used to confirm the composition of calcium-containing stones, if the stone specimens were removed by surgery or obtained after medical treatment or ESWL. Another 231 healthy sex- and age-matched (±3 years) controls (159 men and 72 women, mean age 51.7 years, sd 11.1, range 22–85) were randomly selected from subjects receiving general health examinations at KMUH during the same period. The controls had no past history of stone disease and no clinical findings of stones, confirmed by plain abdominal film and abdominal ultrasonography. Patients and controls with chronic UTI, renal failure, chronic diarrhoea, gout, renal tubular acidosis, primary and secondary hyperparathyroidism, cancer, osteoporosis and the use of medications such as diuretics, vitamin D or calcium supplements, were excluded from the study. All patients and subjects lived in southern Taiwan and provided informed consent before enrolment in the study. The study protocol was approved by the Institutional Review Board of the KMUH.
Clinical information on the age at the first onset of stone formation, total numbers of stone episodes, and the presence of family history were collected from patients with stones by questionnaires and review of their medical charts. A family history of stones was considered positive if any first-degree relative had a history of urolithiasis. The severity of stone disease was determined by calculating the sum of the long-axis diameter of the largest stone (1 point <1 cm; 2 points 1–2 cm; and 3 points >2 cm) and the number of stones (1 point, one stone; 2 points, two stones; 3 points, more than two stones) from the X-ray image in the latest episode.
After overnight fasting, 20 mL of whole blood and a one-spot midstream urine sample from the first voided urine were collected from all subjects. Blood and urine samples were used to measure total calcium, phosphate, uric acid and creatinine levels. If there was UTI the urine sample was discarded for further analysis. All urinary data were corrected using urinary creatinine levels.
For assessing VDR FokI polymorphism, DNA was extracted from peripheral whole blood using a Purgene DNA Isolation Kit (Gentra System, Inc., Minneapolis, MN, USA). The VDR FokI polymorphism (dbSNP ID rs10735810) was determined using the PCR-restriction fragment length polymorphism (RFLP) method described previously . Briefly, DNA fragments were amplified with two primers: 5′- AGC TGG CCC TGG CAC TGA CTC TGC TCT-3′ (forward) and 5′- ATG GAA ACA CCT TGC TTC TTC TCC CTC-3′ (reverse). When the FokI restriction site was present (f genotype), the 265-bp fragment was digested into two 196-bp and 69-bp fragments. The F genotype was not cleaved by FokI and had a single band of 265 bp, whereas the heterozygous (Ff) genotype was 265 bp, 196 bp and 69 bp.
Before full-scale genotyping by PCR-RFLP we confirmed the three genotypes (FF, Ff and ff) by direct sequencing (data not shown). We included one positive control and one negative control sample in each genotyping set (≈ 10 samples). To re-confirm the results of genotypes, we randomly selected 24 patients and 24 controls (≈ 10% of the study samples) for re-genotyping, and the results were completely identical.
All data are shown as the mean (sd); for genotypes, the Hardy–Weinberg equilibrium tests in the controls were calculated first. Contingency table chi-square tests were used to compare VDR genotype frequencies and other qualitative variables. anova with Bonferroni correction for multiple comparisons was used to compare quantitative variables by VDR genotypes. Logistic regression was used to test the associations between disease status and VDR FokI genotypes, after adjusting for other covariates; in all tests, a two-sided P < 0.05 was considered to indicate statistical significance. The frequency of FF genotype among the 231 controls was 25.1% (58/231). Assuming 200 patients and 200 controls were available, and the risk of stone formation is 2.0 for those carrying the FF genotype, the statistical power is 89%. Therefore, the statistical power in the study (235 patients and 231 controls) was >89%.
Among the 235 patients with stones, 125 (53.2%) had more than one stone episode and 81 (34.5%) had a positive family history. Of the 235 patients, 188 had stone specimens for stone analysis, and all had calcium-containing stone components, including 60 (31.9%) with calcium oxalate, 17 (9.0%) with calcium phosphate, and 111 (59.1%) with a mixture of calcium oxalate and calcium phosphate.
FokI VDR genotype frequencies conformed to the Hardy–Weinberg equilibrium in 231 controls (chi-square 0.002; d.f. = 2; P = 1.00). There were no significant differences in genotype distribution and allele frequencies between patients with stones (58 ff, 113 Ff and 64 FF) and controls (57, 116 and 58, respectively; P = 0.91). In addition, the genotype distribution was also not significantly different in 82 familial cases with stones and 231 controls (P = 0.706), or in 125 patients with recurrent stones and the 231 controls (P = 0.424; data not shown).
Laboratory data and clinical characteristics in patients with stones and normal controls categorized by different FokI VDR genotypes are shown in Table 1. Among the patients with stones, those with the FF genotype had significantly more mean stone episodes than did those with Ff or ff genotypes (P = 0.029). In addition, patients with the FF genotype had the youngest mean age at the first episode of stone formation, compared to the other two genotypes, although it was not statistically significant (P = 0.22). Patients with the Ff genotype had intermediate values of mean stone episodes and mean age at the first episode of stone formation. There were no significant differences in the score of stone severity and the percentage of patients with a positive family history among the three different FokI VDR genotypes (Table 1). There were no significant differences in serum or urine variables, including calcium, phosphate and uric acid levels, between patients with stones and controls with different genotypes. In both patients and controls there appeared to be a positive trend in urinary calcium levels in those with the FF or Ff genotype vs those with ff genotype, but the difference was not significant.
Table 1. Clinical and laboratory data in patients with stones and in controls, by different VDR FokI genotypes, and the genotype frequency and genotype-based odds ratio of VDR FokI in patients with stones with different frequency of stone episodes and different age at first onset
|N||58|| 113||64|| |
|Age, years||49.1 (11.8)|| 50.3 (11.2)||50.1 (14.5)||0.82|
|Age at first onset||45.7 (12.4)|| 45.0 (12.2)||42.9 (13.6)||0.22|
|Stone episodes, n|| 1.61 (0.98)|| 1.97 (1.25)|| 2.23 (1.48)||0.03|
|Severity score|| 3.15 (1.23)|| 3.01 (1.30)|| 3.12 (1.14)||0.75|
|n (%) with +ve family history||18 (31.0)|| 44 (38.9)||19 (29.7)||0.32|
|Laboratory data, mean (sd):|
| uric acid|| 7.30 (1.51)|| 7.10 (1.75)|| 7.23 (1.93)||0.77|
| calcium|| 8.90 (1.26)|| 8.72 (0.58)|| 8.69 (0.51)||0.32|
| phosphate|| 4.49 (1.02)|| 4.55 (1.24)|| 4.46 (1.05)||0.87|
|Urinary, mg/mg creatinine|
| uric acid|| 0.521 (0.197)|| 0.508 (0.203)|| 0.489 (0.179)||0.68|
| calcium|| 0.122 (0.087)|| 0.126 (0.076)|| 0.130 (0.085)||0.83|
| phosphate|| 0.462 (0.223)|| 0.440 (0.235)|| 0.423 (0.171)||0.63|
|N||57|| 116||58|| |
|Laboratory data, mean (sd):|
|Serum levels, mg/dL|
| uric acid|| 6.92 (1.46)|| 6.63 (1.55)|| 6.80 (1.43)||0.55|
| calcium|| 9.43 (0.46)|| 9.47 (0.42)|| 9.41 (0.50)||0.14|
| phosphate|| 3.85 (0.42)|| 3.83 (0.40)|| 3.97 (0.38)||0.70|
|Urinary levels, mg/mg creatinine|
| uric acid|| 0.236 (0.190)|| 0.223 (0.170)|| 0.272 (0.195)||0.33|
| calcium|| 0.076 (0.075)|| 0.077 (0.074)|| 0.083 (0.053)||0.85|
| phosphate|| 0.471 (0.180)|| 0.481 (0.341)|| 0.477 (0.187)||0.98|
|Genotype frequency in patients (N), n (%)|
|One episode (110)||33 (56.9)|| 53 (46.9)||24 (37.5)|| |
|> one episode (125)||25 (43.1)|| 60 (53.1)||40 (62.5)|| |
|OR (95% CI)|| 1.00 (ref)|| 1.40 (0.73–2.69)|| 2.15 (1.02–4.54)†|| |
|P for trend|| || || 0.043|| |
|Onset ≥ 45 years (114)||32 (55.2)|| 56 (49.6)||26 (40.6)|| |
|Onset <45 years (121)||26 (44.8)|| 57 (50.4)||38 (59.4)|| |
|OR (95% CI)|| 1.00 (ref)|| 1.42 (0.56–3.57)|| 3.23 (1.08–9.63)‡|| |
|P for trend|| || || 0.037|| |
For the relationship between VDR FokI polymorphism and risk of stone recurrence among the 235 patients, those who had the FF genotype had a significantly higher risk of developing more than one stone episode than those with the ff genotype (adjusted odds ratio 2.15, 95% CI 1.02–4.54, P = 0.044), after adjusting for other potential covariates including age, sex and body mass index (Table 1). The risk for those with the Ff genotype, vs ff, of having more than one stone episode, was intermediate (1.40, 0.73–2.69), although it was not statistically significant. In addition, categorized by a median age at the first episode of 45 years, patients who had the FF genotype were significantly younger at the onset of stone formation than those with the ff genotype (3.23, 1.08–9.63, P = 0.036), after adjusting for other covariates. Again, the age at the first onset of stone formation was less, in order, in those with FF, Ff to ff genotype (P for trend, 0.037). In addition, compared to men, women had a significantly lower risk of having recurrent stones (adjusted odds ration 0.47, 95% CI 0.27–0.86).
In mammals, the 1,25(OH)2D3/VDR system influences calcium homeostasis through the pathway of intestinal calcium absorption, renal calcium excretion and bone metabolism . Several genetic polymorphisms within the human VDR gene were identified, and frequency of certain VDR alleles was related to different functional phenotypes, and further linked to the susceptibility of osteoporosis, response to vitamin D and bisphosphonate therapy [3,14,20].
The VDR FokI polymorphism can be distinguished by RFLP using the endonuclease FokI . Several studies showed that the VDR FokI polymorphism is functionally significant, and indicated that the F allele is more effective than the f allele in transactivation of the vitamin D signal [12–14]. In addition, the VDR FokI polymorphism is not in linkage disequilibrium with the 3′ end cluster of BsmI, ApaI or TaqI polymorphisms .
Although some studies [21–23] found that polymorphisms in the VDR gene, including BsmI in 3′ end, ApaI in intron 8, and TaqI in exon 9, were associated with calcium urolithiasis disease, other studies found no association [15,24,25]. Because the 3′-end polymorphisms do not alter the amino acid of the VDR protein, their functional significance might be secondary to the influence of gene transcription or mRNA stability .
Several studies investigated the association between the VDR FokI polymorphism and calcium urolithiasis [15–19], but the findings were conflicting. Chen et al. and Bid et al.17] found that there was a significantly different distribution of VDR FokI genotypes between patients with calcium stones and normal controls. Those with the F allele were at greater risk of stone disease than those with the f allele. By contrast, Relan et al. found no significant differences in the prevalence of VDR FokI genotypes between patients with stones and controls. Vezzoli et al. and Rendina et al. found no significant differences in the prevalence of VDR FokI genotypes between hypercalciuric and normocalciuric patients with stones, or between hypercalciuric patients and normal controls. By contrast, Relan et al. found that patients with the F allele had a significantly higher 24-h urinary excretion of calcium than did those with the f allele.
In the present study there was no significant difference in the prevalence of VDR FokI genotypes between patients with stones and healthy controls. However, among patients, those with F alleles had a higher risk of having stone episodes and were younger at the first onset of stone formation than were patients with the f allele. Among patients with stones and healthy controls there was a similar trend; those with the FF or Ff genotype had higher urinary calcium levels in the first morning urinary void than those with the ff genotype, although the difference was not statistically significant. The studies by Chen et al. and Bid et al. might have included patients with multiple stone episodes rather than including the general population of patients with stones, as we and Relan et al. did. Because patients with multiple stone episodes seem to carry the FF or Ff genotype, rather than the ff genotype, there might be a difference in the prevalence of FokI VDR genotypes between patients with multiple stone episodes and healthy controls. Also, previous studies indicated that the F allele is more effective than the f allele in transactivation of the 1,25(OH)2D3 signal [12–14]. Relan et al. found that the F allele increased the urinary calcium excretion rate more than did the f allele. Although we found a similar trend in the excretion of urinary calcium in both patients and controls, it was not significantly different between those with the FF, Ff or ff genotype. As we had no information about the phenotype of FokI polymorphism we cannot clarify the functional role of the F allele in stone formation. Recently, Mossetti et al. reported an association between VDR BsmI and TaqI polymorphisms and idiopathic hypocitraturia in calcium-oxalate recurrent stone formers. Whether FokI polymorphism is also associated with citrate metabolism needs to be validated. In addition, further studies are needed to clarify the role of VDR FokI polymorphism in calcium metabolism, or other possible mechanisms.
In the present study, the F allele appeared to be important in the clinical presentation of patients with stones. An alternative explanation is that the formation of urinary stones is multifactorial, and is influenced by environmental, hormonal and genetic factors [1,27]. Therefore, the VDR FokI polymorphism might need to combine with other metabolic problems or other gene polymorphisms to induce stone formation. After the formation of urinary stones, the VDR FokI polymorphism might determine the number of stone episodes, although this proposed mechanism needs to be verified.
The mean age of the present patients was ≈ 50 years, which is compatible with the findings from other epidemiological studies from Taiwan [1,28,29]. One of the limitations in the study is that we did not evaluate other VDR polymorphisms and did not measure citrate and oxalate levels. Further studies might be needed to evaluate whether FokI polymorphism is also associated with citrate metabolism and through this mechanism affect calcium stone formation.
In conclusion, there were no significant differences in VDR FokI genotypes between patients with stones and healthy controls. However, the VDR FokI polymorphism might be important in the clinical presentation of patients with stones, especially for the number of stone episodes and age at onset. Intensive dietary habits and lifestyle change could be suggested for those patients with the Ff and FF genotypes, especially FF, to decrease the risk of further stone recurrence.
This study was supported by grants from the Kaohsiung Medical UniversityGrant sponsor: Kaohsiung Medical University (M094007) and Kaohsiung Medical University Hospital (94-KMUH-003)
CONFLICT OF INTEREST
None declared. Source of funding: Kaohsiung Medical University (M094007) and Kaohsiung Medical University Hospital (94-KMUH-003).