Association Between Vitamin D Receptor Gene

Polymorphism and Relative Hypoparathyroidism in Patients

with Chronic Renal Failure

ELVIRA FERNANDEZ,* JOAN FIBLA,t ANGELS BETRIU,� JOSEP M. PIULATS,t

JAUME ALMIRALL,� AND JESUS MONTOLIU**Nephrology Service and Department of Medicine, Hospital Universitari de Lleida Arnau de Vilanova;

�Human Genetics Unit, Department of Basic Medical Sciences, Universitat de Lleida; �Sistemes Renals,

Lleida, Spain; and �Nephrology Unit, Consorci Hospitalari Parc TaulI Sabadell, Spain.

Abstract. To study the influence of vitamin D receptor (VDR)

gene polymorphism on parathyroid cell function in chronic

renal failure, 85 patients who had serum PTH levels <12

pmolfL (the low intact PTH [iPTH] group) and 46 patients who

had serum iPTH levels >60 pmoiIL (the high iPTH group)

were selected out of a total dialysis population of 170 individ-

uais. As a result of subsequent exclusions based on several

criteria in both groups (diabetic patients, serum aluminum

levels, serum calcium levels, and time on dialysis), the final

low iPTH group consisted of 34 patients and the final high

iPTH included 32 patients. A healthy control population (n =

120) and 162 of the 170-patient dialysis population served as

control groups. VDR gene polymorphism was determined by

digestion with the BsmI enzyme and single-strand conforma-

tion polymorphism analysis of PCR amplified fragments. Se-

rum iPTH levels were lower in patients with the BB genotype

than in those with the Bb or bb genotype, both in the total

dialysis population and when the various exclusion criteria

were applied. No differences in genotypic and allelic frequen-

cies were found between the healthy control population and the

high iPTH group. However, the genotypic distribution was

significantly different in the low iPTH group of patients before

and after applying all exclusion criteria (P = 0.037 and P =

0.0 1 8, respectively). In the final selected population, the bb

genotype was less frequent in the low iPTH group than in the

total dialysis population (14.7% versus 36.4%; odds ratio, 0.3;

confidence interval, 0.11 to 0.82; P = 0.01). Conversely, the

BB genotype was over-represented in the low iPTH group

(23.3% versus 19.7%; odds ratio, 1.9; confidence interval, 0.85

to 4.3; P = 0. i). In addition, the bb genotype and the b allele

frequencies were lower in the low iPTH group than in the high

iPTH group (14.7% versus 34.4%, P = 0.06, and 41 .2% versus

60.9%, P = 0.02, respectively), and the BB genotype and the

B allele were significantly more frequent in the low PTH group

than in the high iPTH group (32.3% versus 12.5%, P = 0.05,

and 58.8% versus 39.1%, P = 0.02, respectively). Thus, VDR

gene polymorphism influences parathyroid function in chronic

renal faiiure. (J Am Soc Nephrol 8: 1546-i552, 1997)

The BsmI polymorphism identified in the untranslated 3’ re-

gion of the vitamin D receptor (VDR) locus defines B and b

alleles that may determine differences in the VDR gene ex-

pression and ultimately have functional consequences. Recent

observations in a large population of healthy adults indicate

that osteocalcin serum levels are higher in individuals with the

BB genotype (1). This genotype is also associated with low

bone density and an increased risk of osteoporosis in post-

menopausal women (2,3). Although the molecular basis for

this phenomenon is largely unknown, the B allele has been

linked to increased transcriptional activity or messenger RNA

stability (2). These and other observations (3-6) suggest that

the presence of a specific allele can have clinical consequences

in bone.

Received January 2. 1997. Accepted March 14, 1997.

Correspondence to Dr. Elvira Fern#{225}ndezNephrology Service, Hospital Uni-versitari de Lleida Arnau de Vilanova, Rovira Roure 80, 25198 Lleida, Spain.

1046-6673/08010-I 546$03.00/0Journal of the American Society of Nephrology

Copyright 0 1997 by the American Society of Nephrology

The interaction of caicitriol with the VDR in the parathyroid

cell inhibits pre-pro-PTH synthesis, upregulates VDR expres-

sion, and exerts an antiproliferative effect (7,8). Among other

factors, resistance of parathyroid cells to calcitriol in uremia

piays an important role in the pathogenesis of hyperparathy-

roidism secondary to renal failure (9). No definite reason for

this resistance has been ascertained, but there is evidence of a

lower VDR density in the parathyroid cells of uremic patients

(10,1 1). For these reasons, we investigated whether ailelic

differences in the VDR gene could determine differences in

parathyroid cell function that could be at least partially respon-

sible for the wide variation in the degree of secondary hyper-

parathyroidism observed in patients with end-stage renal dis-

ease (ESRD).

Materials and MethodsPatients

Of a total maintenance hemodialysis population of 170, we selected

131 patients with ESRD because of their extreme serum intact PTH

(iPTH) values. On the basis of their average serum iPTH levels

(measured on two occasions separated by 3-mo intervals), these 131

patients were divided into two groups: the low iPTH group, repre-

VDR Gene Polymorphism and Relative Hypoparathyroidism 1547

sented by 85 patients with serum iPTH levels < 12 pmoiIL (normal in

healthy individuals up to 5.6 pmolIL), and the high iPTH group,

represented by 46 patients with serum iPTH levels >60 pmol/L.

Serum iPTH levels were measured by a two-site chemiluminometric

assay (Magic Lite iPTH immunoassay, Ciba-Corning Diagnostics

Corp., Medfield, MA). Serum calcium, phosphorus, and aluminum

levels were measured simultaneously in each iPTH determination. To

define the low iPTH group, we later excluded diabetic patients,

patients with plasma aluminum levels >2.2 mmol/L, and those with

serum calcium levels >2.75 �.tmolIL. Similarly, to define the high

iPTH group, we later excluded patients with serum calcium levels<2.4 mmollL. This left 61 patients in the low iPTH group and 45

patients in the high iPTH group. Because time on dialysis is known to

play a major role favoring the development of severe secondary

hyperparathyroidism, we subsequently excluded from the low iPTH

group those patients who had been on dialysis for <2 yr, and with this

exclusion, 34 patients remained in this group (referred to as the low

iPTH group). Following the same line of reasoning, we subsequently

excluded from the high iPTH group those patients who had been on

dialysis for >5 yr. and with this exclusion, 32 patients remained in

this group (referred to as the high iPTH group). These time periods

were chosen arbitrarily, using heuristic reasoning. No patient had

received immunosuppressive or anticonvulsant therapy.

Our treatment policy was to administer calcium compounds as a

first-choice, phosphate-binding agent. Aluminum hydroxide was the

second choice. Patients in the low iPTH group had been treated with

low-dose (0.25 p�g, three times weekly) oral calcitriol at some point

during their clinical course, but when samples for iPTH were collected

they had been off calcitriol for at least 1 yr. because low PTH levels

had already been noticed in previous tests. Patients in the high iPTH

group were either treated with intravenous calcitriol boluses (15

patients) at doses ranging from 1 to 2 �g posthemodialysis 3 times a

week, or they required parathyroidectomy because of calcitriol treat-

ment failure (17 patients). Surgically removed parathyroid glands

ranged in weight from 1 .5 to 6 g and showed nodular hyperplasia on

histological examination. Serum iPTH levels are given before para-

thyroidectomy in these patients. Dialysate calcium concentration was

1 .25 mmol/L in patients receiving intravenous calcitriol and 1.75

mmol/L in the remaining patients. As control groups, we determined

genotypic frequencies in 162 of the 170-patient dialysis population

and in a healthy control population made up of 120 individuals. All

patients and control subjects correspond to a Caucasian population

from the same geographic area.

VDR Genotyping

VDR genotyping was performed by BsmI digestion and

single-strand conformation polymorphism analysis of PCR-

amplified products. Briefly, a l9l-bp fragment of the VDR

locus containing the BsmI polymorphic site was amplified

using primers hVDR-3 (5 ‘-AGTGTGCAGGCGATCGTAG-

3’) and hVDR-4 (5’ ATAGGCAGAACCATCTCTCAG-3’). A

Perkin-Elmer 480 thermal cycier was used to carry out 30

amplification cycles with the following temperature profile:

94#{176}Cfor 1 mm, 62#{176}Cfor 1 mm, and 72#{176}Cfor 1 mm. Amplified

fragments were digested with BsmI enzyme (New England

Biolabs, Beverly, MA) according to the manufacturer’s in-

structions. We double-checked the VDR typing by single-

strand conformation polymorphism analysis (12) of the i91-bp

fragment. Single-strand bands were resolved in nondenaturing,

12% polyacrylamide gels (30:0.8 acrylamide:bis ratio), 1 X

TBE (94 niT�’I Tris-HC1, pH 8.0, 88 mM boric acid, and 2 mM

ethyienediamine tetra-acetic acid), without glycerol. Electro-

phoresis was completed in 1 .5 to 2 h in 1 X TBE at constant

i 00 V and room temperature. After electrophoresis, gels were

silver-stained (13).

Statistical Analyses

Differences between mean values for demographic and bio-

chemical parameters in patient groups were analyzed by un-

paired t test. Overall comparisons between serum PTH levels

of each genotype were evaluated by one-way ANOVA, and the

significance of differences between mean values was deter-

mined by the least significant difference test. The goodness of

fit to the Hardy-Weinberg equilibrium was calculated by the

chi-square test. Association studies between groups for geno-

types and alleles were tested by contingency table analysis and

using the chi-square test. Results are expressed as means ±

SD. Confidence interval (CI) was specified when appropriate.

ResultsDemographic and biochemical parameters and the etiology

of ESRD in both patient groups and in the total dialysis

population are given in Table 1 . No significant differences

were found within each group for the three genotypes. In the

low PTH group, serum iPTH values ranged from 1 .5 to 12

pmol/L (mean, 4.2 ± 3.3 pmolIL), and in the high PTH group

serum iPTH values ranged from 60 to 1 80 pmolIL (mean,

120.8 ± 57 pmolIL). When compared, groups differed in that

patients in the low iPTH group had been on dialysis longer and

also had lower serum aluminum levels. Women had a marked

predominance in the high iPTH group, a fact already observed

by others (14). No relevant differences in the etiology of renal

failure were encountered.

Figure 1 shows that serum iPTH levels in the general dial-

ysis population when the different selection criteria (specified

in the figure legend) were applied were lower in patients with

the BB genotype (29.2 ± 4 1 , 27.8 ± 42, and 3 1 .4 ± 48

pmol/L, respectively) than in those with the Bb (54.3 ± 68,

59.9 ± 70, and 62.8 ± 76 pmoi/L, respectively) or bb gemo-

types (37. 1 ± 46, 43.2 ± 48, and 58.4 ± 5 1 pmoilL, respec-

tively). In addition, in the final selected population, the prey-

alence of patients with low PTH levels was higher in those with

the BB genotype (68.7%; CI, 45.9 to 91.4%) than in those with

the Bb (43%; CI, 28 to 57%) or bb (22.7%; CI, 5.2 to 40.2%)

genotypes.

The genotype frequencies estimated from the total dialysis

population (BB,19.7%, CI, 13.5 to 25.8%; Bb, 43.8%, CI, 36.1

to 5 1 .4%; bb, 36.4%, CI, 28.9 to 43.8%) and the healthy

control population (BB, 13.4%, CI, 7.3 to 19.5%; Bb, 55.8%,

CI, 46.9 to 64.6%; bb, 30.8%, CI, 22.5 to 39.1%) were under

Hardy-Weinberg equilibrium, revealing values consistent with

those published in the literature for Caucasian populations.

Genotype frequencies of all subgroups of patients, after apply-

ing the different exclusion criteria, were also under Hardy-

Weinberg equilibrium. No differences in allelic frequencies

were found between the total dialysis population and the

healthy control population.

1548 Journal of the American Society of Nephrology

� �\ � \�-. -. �C C C.�

... +1 (N +1 +1 +1 +1 +1 r’�� � ‘5 r’-� tt� O�

E� �r-: � Q0� 0’ � r-�� r�1 (-�

r’� C C� �

�. - r�4 �tC C C

.0 +1 � +1 +1 +1 +1 +� _ � r-�� �

2 � � � �c r’� � 0� fl

(� �c � c�r� -�

?� � r� �

.� .- C- 0\ � �0

� .� .0 +1 � +1 +1 +1 +1 +1.� - - �CC C C

o� � � oci::r- � � Q r�r�r� -(-4 c�r-�� L(� � ocSr-� r-J -

� � r� �

�;�-, 0� r�Ci � 0�

-�‘fi - ‘tC C C+I�+I +1+1 +1 +1

..� -. --r�1� r-’fl - r�.� � � �r�r�-C- ��

cl�.� r- - OOC��� �- .---� C�C C C C).� � +1- +1 +1+1 +1 +1 �.0C- ‘fi‘� - --� E2 r-�r� ‘� oc� r-� � r- Lfl�� - �c� � �fl CC)c1�o� C)n � r- (�ri � �0 - \C r�C C CbO 0. � +1-� +1 +1+1 +1 +1� .0 � (��lIt� Q� C� �0� )- � r’� � - � C)C.� L� �0 ��n� �.�.- QC -C � � � r� � �&� .-� � +1� +1 +1+1 +1 +� � C�I ��‘Cr-�1C--.� v� �q � Q r-�C �C � (� C +10 �c �.- C)00 r’� � r’� � E� - r� -� �� +1� +1 +1+1 +1 +1 � � (�‘�I(�-� C’�)c/) � �0� � \C �0 �C �0tL� -� r-: � _� � C)�c ‘i-� 0.bI� � � � �C. 30.� - -�c.�r �flC C C C)� � +1N � � � � +1 r� - r� � r’� C C�� F2 � � 0� r�1-- -r’�C�-4- -�C.)‘5 �- C).� � �t �cr1 � 0�C � � ‘1� �fC C CC� #{149}�� +1� +1 +1+1 +1 +1 1/)rI� �. .0 ,� � � r- � oc C � ‘� ‘� � � � r-� .� r�(N,� r�r�i � _�C)� � �� � �c c� � t- 00c� � � �CC C C -C� .0 � c:�+1 +1+1 +1 +1 � ,� , � .�. -..�-� .� � � r�1r’� O� 0�’.� E� � r-Jr’� -�0 � � �0-C _r�1 �C �CU Lfl C . .0 �r-� �tC C C.-� +1� +1 +?+1 +1 +1 r- - c�i - - �flC .-0\�Cfl r’J’�: � 00C � r-i � - �\0 ‘�0 C).-� -� � � C)11Vn� � � >�c� E� �.CC)0 � CC�� � C)�n .� : �E E � �0.�- 0.�� c,�C)0 � � .�� .� .� >� � � .� CC) � � .� �#{149}hh!� � �� .� I-� � -:- - _- �.- �C)ECC��a� �--., C-) � C) EEEE�E��.. � �� C)60.- ;a-,�� �� .0 C � C C)>� C)� � .0 C-.� 0��E � � �-6 �-� � � � � �.0�O_ �0. � ,,�� � � U

A B

100

80 -

60 -

40 -

20 -

0-

One-way ANOVA p = 0,079

II1I

0

E0.

I

0�

C

100 -

80 -

60 -

40 -

20 -

0-

T 0� E

� 0.

j_ IF-0�

One-way ANOVA p = 0,065

�F1I

100 -

80 -

60 -

40 -

20 -

0-

T� 0

�1�

F-0�

BB Bb bb

GENOTYPE

I I

BB Bb bb

GENOTYPE

Figure 1. Serum intact PTH (iPTH) levels in patients with the BB, Bb, or bb genotype, when no selection criteria were applied (A), when

selection criteria other than time on dialysis were used (B), and when time on dialysis was included as a selection criterion (C).Values are

expressed as mean ± 95% confidence interval (CI) of the mean.

I I

BB Bb bb

GENOTYPE

VDR Gene Polymorphism and Relative Hypoparathyroidism 1549

One-way ANOVA p = 0,220

IiIiI

I

Table 2 shows genotypic and allelic frequencies in healthy

control subjects and in the various patient groups when no

exclusion criteria were applied; when exclusion criteria, except

time on dialysis, were applied; and when time on dialysis was

included as an exclusion criterion. No differences were found

between healthy control subjects and the high iPTH group. The

BB genotype was more frequent in the low iPTH group of

patients throughout. Differences in genotype distribution be-

tween expected and observed values in the low iPTH group

reached statistical significance when no exclusion criteria were

applied (BB, 23.5%, CI, 14.4 to 32.5%; Bb, 38.8%, CI, 28.4 to

49.1%; bb, 37.6%, CI, 27.3 to 47.8%; P = 0.037) and when all

exclusion criteria were considered for analysis (BB, 32.4%, CI,

16.6 to 48.1%; Bb, 52.9%, CI, 36.1 to 69.6%; bb, 14.7%, CI,

2.79 to 26.6%; P = 0.018). The final low iPTH group was

characterized by an excess of subjects homozygous for the

allele B and a defect of the bb genotype. Figure 2 shows

absolute values and frequencies of genotypes and alleles in the

total dialysis population and both patient groups of the final

selected population. We found that the bb genotype was less

frequent in the low iPTH group than in the total dialysis

population and was associated with a significant decreased risk

of belonging to the low iPTH group (14.7% versus 36.4%.

odds ratio, 0.3, CI, 0.11 to 0.82; P = 0.01). Conversely, the BB

genotype was over-represented in the low iPTH group when

compared with the total dialysis population (32.3% versus

19.7%, odds ratio, 1.9, CI, 0.85 to 4.3; P = 0.1). In addition,

the bb genotype and the b allele frequencies were lower in the

low iPTH group than in the high iPTH group (14.7% versus

34.4%, P = 0.06, and 41.2% versus 60.9%, P 0.02, respec-

tiveiy), and the BB genotype and the B allele frequency were

significantly higher in the low PTH group than in the high

iPTH group (32.3% versus 12.5%, P = 0.05, and 58.8% versus

39.1%, P = 0.02, respectively). Thus, a positive association

was detected between the low iPTH group and the BB geno-

type and a stronger but negative association was detected

between the low iPTH group and the bb genotype.

DiscussionIn addition to its known influence on bone turnover, VDR

polymorphism has been recently shown to modify parathyroid

cell differentiation and function. Patients with nonfamilial pri-

mary hyperparathyroidism have a significantly higher fre-

quency of the bb genotype and the b allele (15). However, the

genetic influences determining the development and severity of

hyperparathyroidism secondary to renal failure were largely

unknown until now. In ESRD patients, there are wide varia-

tions in the degree of secondary hyperparathyroidism. Whereas

some patients develop severe and uncontrollable hyperparathy-

roidism, which has its histologic counterpart in osteitis fibrosa

cystica, others develop only modestly elevated iPTH levels that

fail to promote adequate bone turnover and finally result in

adynamic bone disease ( 1 6, 17). The reasons for this heteroge-

neous clinical behavior are not well defined. The growing

frequency of adynamic bone disease has been attributed, in

different studies, to the use of calcium compounds as phos-

phate binders; high-dose caicitriol therapy; changing dialysis

populations that are older and have more diabetic patients; and

aluminum intoxication (18). On the other hand, the develop-

memt of severe secondary hyperparathyroidism has been cor-

related with the duration of renal failure and time on dialysis

(8).

The aim of our study was to determine the contribution of

genetic background to the different patterns of parathyroid

response in ESRD. Hopefully, this could, at least partially,

explain why some patients behave one way or another. With

that purpose in mind, we defined two groups of ESRD patients,

primarily on the basis of their iPTH levels. The high iPTH

group had levels >60 pmolIL. Such levels have been found in

previous studies to correlate strongly with the histologic pat-

I 550 Journal of the American Society of Nephrology

Table 2. Genotype and allele frequencies in healthy control subjects and in the various patient groupsa

CategoryHealthy Control

Subjects(11= 120)

A (n = 134) B (n = 106) C (n = 66)

Lown=85 Highn=49 Lown=6l Highn=45 Lown34 Highn=32

Genotype

BB 16 (13.4%) 20 (23.5%) 7 (14.2%) 15 (24.6%) 5 (1 1.1%) 1 1 (32.4%) 4(12.5%)

Bb 67 (55.8%) 33 (38.8%) 25 (51.0%) 26 (42.6%) 24 (53.3%) 18 (52.9%) 17 (53.1%)

bb 37 (30.8%) 32 (37.6%) 17 (34.6%) 20 (32.7%) 16 (35.6%) 5 (14.7%) 1 I (34.3%)

P versus control 0.037 0.846 0.109 0.822 0.018 0.929

subjects

P low versus 0.292 0.078 0.064

high

Allele

B 99 (41%) 73 (43.7%) 39 (41.9%) 56 (44.7%) 34 (38.0%) 40 (58.8%) 25 (39.0%)

b 141 (59%) 97 (56.2%) 59 (58.1%) 66 (55.2%) 56 (62.0%) 28 (41.1%) 39(60.9%)

P versus control 0.732 0.805 0.567 0.398 0.01 0.752

subjects

P low versus 0.615 0.237 0.02

high

a Genotypic and allelic frequencies in healthy control subjects and in the various patient groups, when no exclusion criteria were applied

(A); when exclusion criteria except time on dialysis were applied (B); and when time on dialysis was included as an exclusion criterion

(C). The BB genotype was more frequent in the low ‘PTH group of patients throughout. Differences in genotype distribution in thesepatient groups, when compared with those of the healthy control subjects, reached statistical significance before and after applying all

exclusion criteria. Differences in allele frequencies between the low iPTH group and the high iPTH group only reached statistical

significance in the final selected population.

tern of osteitis fibrosa cystica (16, 17). This group was charac-

terized by refractoriness to high-dose calcitriol therapy, a find-

ing that usually indicates disease severity (8). To limit the

impact of time on dialysis in this group over the development

of parathyroid hyperplasia, we selected only those patients

with <5 yr on dialysis. The low iPTH group had levels <12

pmoi/L. Such iPTH levels have been found in previous studies

to correlate strongly with the histologic pattern of adymamic

bone disease (I 6, 17). In this group, we excluded the patients

having any of the risk factors suggested to favor the existence

of adynamic bone disease. In this group, we also allowed

ample time for the development of parathyroid hyperplasia,

because we included only patients that had been on dialysis for

�2 yr. Time on dialysis was actually higher in the low than in

the high iPTH group and therefore cannot be fully incriminated

in the development of parathyroid hyperplasia. It is conceiv-

able, however, and our data when using progressive exclusion

criteria support this contention, that the longer the time on

dialysis, the more likely is the possibility of severe hyperpara-

thyroidism, and that this may even cloud or overcome genetic

influences. Serum aluminum levels were higher in the high

than in the low iPTH group, and thus could not have sup-

pressed PTH secretion. It is well known ( 14) that women

predominate in the high iPTH group. Otherwise, both groups

were similar in terms of age, serum calcium and phosphate

levels, and the etiology of renal disease.

The distribution of genotypic and allelic frequencies was

similar in the total dialysis population, the healthy control

subjects, and the high iPTH group. Therefore, the low iPTH

group emerged as a distinct population characterized by a

higher frequency of the B allele and the BB genotype and a

lower frequency of the b allele and the bb genotype. This is in

keeping with a “mirror image” with respect to recent data

indicating a higher frequency of the b allele and the bb geno-

type in patients with primary hyperparathyroidism (15). Con-

sistent with this formulation is the finding of higher serum

iPTH levels in patients with the bb genotype, a fact that has

been confirmed by others (19,20). Our results should be inter-

preted with some caution, because after applying all exclusion

criteria, the sample size was relatively small. However, the

overexpressiom of the BB genotype, regardless of the selected

population, supports our hypothesis. Taken together, these data

are consistent with the notion that the B allele is associated

with increased transcriptional activity or messenger RNA sta-

bility, which permits better VDR expression and therefore

interaction with calcitriol. Consequently, calcitriol could better

exert its antiproliferative effect over parathyroid cells and

inhibit PTH synthesis to a greater extent. Alternatively, one

could speculate that it is the absence of the b allele that protects

against the development of secondary hyperparathyroidism.

Additional studies about the function of these VDR gene

alleles are required to support any of the proposed hypotheses.

In summary, we have shown that individuals with the bb

genotype are at a lesser risk of having relative hypoparathy-

roidism. Conversely, those with the BB genotype seem to be

prone to low PTH levels. Therefore, our results strongly sug-

60

50

40

30

20

10

0

BB

Bb

bb

70

60

50

40

30

20

10

0

B

b

A

Cl)ci)ci)Cu

�6

.�

E

z

VDR Gene Polymorphism and Relative Hypoparathyroidism 1551

B

Figure 2. Genotype and allele distribution in the final selected pop-

ulation after applying all exclusion criteria. (A) Odds ratios (OR) are

for the relative risk of belonging to the low iPTH group. Individualswith the bb genotype have a significantly decreased risk of having low

iPTH levels. In contrast, the BB genotype was associated with anincreased, although not significant, risk of belonging to the low iPTH

group. The BB genotype (indicated by *) was more frequent and the

bb genotype (indicated by &) was less frequent in the low iPTH group

than in the high iPTH group (P = 0.05 and P = 0.06, respectively).U, BB genotype; 0, Bb genotype; �, bb genotype. (B) Odds ratios

have the same meaning as in Panel A. The B allele frequency (mdi-

cated by *) was higher, and, consequently, the b allele frequency(indicated by &) was lower in the low iPTH group than in the highiPTH group (P = 0.02). Absolute values for the genotype and allele

frequencies are given at the bonom of the figure. U. B allele; EL b

allele.

gest that genetic background, in addition to environmental

factors, may induce relative hypoparathyroidism in renal fail-

ure. These data could also help clarify why some patients with

ESRD have relatively low iPTH levels, never develop signif-

icant hyperparathyroidism, and are at risk for adymamic bone

disease.

AcknowledgmentsThis work was supported by grants “La Paeria” to the University of

Lleida and FIS 97/0641 . The authors thank Drs. G. Cao and M. C.Rivas (Clinical Chemistry Service), Dr. L. Perez (Surgery Service),

and Drs. M. I. Panad#{233}sand I. Ramos (Pathology Service) for their

cooperation (all are affiliated with the Hospital Universitari Arnau de

Vilanova, Lleida, Italy). Dr. Armando Torres (University of la La-

guna) and Dr. A. Sorribas (University of Lleida) provided expert

advice.

References1. Morrison NA. Yeoman R, Kelly PJ. Eisman IA: Contribution of

transacting factor alleles to normal physiological variability:Vitamin D receptor gene polymorphisms and circulating osteo-

calcin. Proc Nail Acad Sci USA 89: 6665-6669, 19922. Morrison NA. Qi IC, Tokita A, Kelly P1, Croft L, Nguyen TV,

Sambrook PN, Eisman IA: Prediction of bone density from

vitamin D receptor alleles. Nature 367: 284-287, 1994

3. Yamagata Z, Miyamura T, Lijima 5, Asaka A, Sasaki M, Kato I,Koizumi K: Vitamin D receptor gene polymorphism and bonemineral density in healthy Japanese women. Lancet 344: 1027,

I994

4. Krall EA, Parry P. Lichter JB, Dawson-Hughes B: Vitamin D

receptor alleles and rates of bone loss: Influences of years sincemenopause and calcium intake. J Bone Res 10: 978-984, 1995

5. Riggs BL, Nguyen TV. Melton U 3rd, Morrison NA, O’Fallom

WM, Kelly PJ, Egan KS, Sambrook PN, Muhs IM, Eismam JA:

The contribution of vitamin D receptor gene alleles to the deter-

mimation of bone mineral density in normal and osteoporotic

women. J Bone Res 10: 99 1-996, 1995

6. Farrow 5: Allelic variation and the vitamin D receptor. Lancet

343: 1242, 1994

7. Lowe KE. Maiyar AC, Norman AW: Vitamin D-mediated geneexpression. Crit Rev Eukaryotic Gene Expression 2: 65-109,

I992

8. Drueke TB: The pathogenesis of parathyroid gland hyperplasia

in chronic renal failure. Kidney mt 48: 259-272, 1995

9. Fukagawa M, Fukuda N, Yi H, Kurokawa K: Resistance of

parathyroid hyperfunction in chronic renal failure. Nephrol Dial

Transplant 10: 316-319, 1995

10. Korkor AB: Reduced binding of(3H) 1,25-dihydroxyvitamin D3

in the parathyroid glands of patients with renal failure. N Engl

JMed3l6: 1573-1577, 1987

11. Fukuda N, Tanaka H, Tominaga Y, Fukagawa K, Seino Y:Decreased 1,25-dihydroxyvitamim D3 receptor density is associ-

ated with a more severe form of parathyroid hyperplasia inchronic uremic patients. J C/in Invest 92: 1436-1443, 1995

12. Warren W, Hovig E, Smith-Fosorrensen B, Borresen AL: De-

tection of mutations by single-strand conformation polymor-

phism analysis. In: Current Protocols in Human Genetics, edited

by Dracopoli NC, Haines IL, Korf BR, Moir DT, Morton CC,Seidman CE, Seidman 1G. Smith DR. New York, John Wiley &Sons, 1994, pp 7.4.1-7.4.6

I 552 Journal of the American Society of Nephrology

13. Beidler J. Hilliard P. Rub R: Ultrasensitive staining of nucleic

acids with silver. Anal Biochem 126: 374-380. 1982

14. Cundy T, Hand DI, Oliver DO, Woods CG, Wright FW. Kanis

JA: Who gets renal bone disease before beginning dialysis? Br

MedJ29O: 271-275, 1985

15. Carling T, Kindmark A, Hellman P. Lundgren E, Ljunghall 5,

Rastad J. Akerstr#{246}m G, Melhus H: Vitamin D receptor genotypes

in primary hyperparathyroidism. Nat Med I : I 309- 13 1 1 . I 995

16. Quarles D, Lobaugh B, Murphy G: Intact parathyroid hormoneoverestimates the presence and severity of parathyroid-mediated

osseus abnormalities in uremia. J C/in Endocrino/ Metab 75:

145-150. 1992

17. Torres A, Lorenzo V. Hem#{225}ndez D, Rodriguez IC. Concepci#{243}n

MT. Rodriguez AP, Hern#{225}ndezA, De Bonis E, Darias E, Gonza-

lez-Posada JM, Losada M, Rufino M, Felsenfeld AJ, Rodriguez

M: Bone disease in predialysis, hemodialysis, and CAPD pa-

tients: Evidence of a better bone response to PTH. Kidney mt 47:

1434-1442, 1995

18. Hruska KA, Teitelbaum SL: Renal osteodystrophy. N Eng/

JMed333: 166-174. 1995

19. Aterini 5, Salvadori M, Ippolito E. Petrocelli P. Pacini 5, Sineo

L, Martini R. Failli M, Amato M, Ruggiero M: The role ofvitamin D receptor gene alleles in the secondary hyperpara-

thyroidism of hemodialysis patients. J Nephrol 9: 201-206,1996

20. Tsukamoto Y, Heishi M, Nagaba Y, Kobayashi N, Nomura Y,

Takahashi K, Tazawa H: More on hyperparathyroidism and the

vitamin D receptor. Nat Med 2: 1 162, 1996