Bone 69 (2014) 6–11

Contents lists available at ScienceDirect

Bone

j ourna l homepage: www.e lsev ie r .com/ locate /bone

Case Report

A human vitamin D receptor mutation causes rickets and impairedTh1/Th17 responses

Bram C.J. van der Eerden a,⁎, Josine C. van der Heyden b, Jan Piet van Hamburg c, Marijke Schreuders-Koedam a,Patrick S. Asmawidjaja c, Sabine M. de Muinck Keizer-Schrama b, Annemieke M. Boot d, Erik Lubberts c,Stenvert L.S. Drop b, Johannes P.T.M. van Leeuwen a

a Department of Internal Medicine, Erasmus MC, Rotterdam, The Netherlandsb Department of Pediatric Endocrinology, Erasmus MC, Rotterdam, The Netherlandsc Department of Rheumatology, Erasmus MC, Rotterdam, The Netherlandsd Department of Pediatrics, University Medical Center Groningen, Groningen, The Netherlands

⁎ Corresponding author at: Erasmus MC, DepartmenEe585b, 's-Gravendijkwal 230, 3015 CE Rotterdam, T7044862.

E-mail address: b.vandereerden@erasmusmc.nl (B.C.J.

http://dx.doi.org/10.1016/j.bone.2014.08.0058756-3282/© 2014 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 28 March 2014Revised 5 August 2014Accepted 13 August 2014Available online 6 September 2014

Edited by: Bente Langdahl

Keywords:RicketsVitamin DCalciumPhosphorusT cellsVitamin D receptor

We present a brother and sister with severe rickets, alopecia and highly elevated serum levels of 1,25-dihydroxyvitamin D (1,25-(OH)2D3). Genomic sequencing showed a homozygous point mutation (A133G) inthe vitamin D receptor gene, leading to an amino acid change in the DNA binding domain (K45E), which wasdescribed previously. Hereditary vitamin D resistant rickets (HVDRR) was diagnosed. Functional studies in skinbiopsy fibroblasts confirmed this. 1,25-(OH)2D3 reduced T helper (Th) cell population-specific cytokine expres-sion of interferon γ (Th1), interleukins IL-17A (Th17) and IL-22 (Th17/Th22) in peripheral blood mononuclearcells (PBMCs) from the patient's parents, whereas IL-4 (Th2) levels were higher, reflecting an immunosuppres-sive condition. None of these factors were regulated by 1,25-(OH)2D3 in PBMCs from the boy. At present, bothpatients (boy is 23 years of age, girl is 7) have not experienced any major immune-related disorders. Althoughboth children developed alopecia, the girl did so earlier than the boy. The boy showed complete recovery fromthe rickets at the age of 17 and does not require any vitamin D supplementations to date.In conclusion, we characterized two siblings with HVDRR, due to a mutation in the DNA binding domain of VDR.Despite a defective T cell response to vitamin D, no signs of any inflammatory-related abnormalities were seen,thus questioning an essential role of vitamin D in the immune system. Despite the fact that currentlymedicine isnot required, close monitoring in the future of these patients is warranted for potential recurrence of vitamin Ddependence and diagnosis of (chronic) inflammatory-related diseases.

© 2014 Elsevier Inc. All rights reserved.

Introduction

Calcium homeostasis is primarily regulated by vitamin D. Vitamin Dis produced in the skin by photosynthesis form 7-dehydroxycholesterolby UV-B light or is derived from food sources such as fatty fish. The ac-tivemetabolite of vitaminD is 1,25-(OH)2D3,which binds to the vitaminD receptor (VDR) to modulate its actions through altering target geneexpression. The VDR is located in the nucleus of many target tissues, in-cluding the intestine, kidney, bone and skin. Resistance to 1,25-(OH)2D3

can cause the rare disorder Hereditary Vitamin D Resistance Rickets(HVDRR). Approximately 35mutations in the VDR have been identified[1]. The majority of the pathogenic mutations are located in the DNABinding Domain (DBD) and the Ligand Binding Domain (LBD). DBD

t of Internal Medicine, Roomhe Netherlands. Fax: +31 10

van der Eerden).

mutations prevent the VDR from activating gene transcription eventhough 1,25-(OH)2D3 binding to the VDR is normal. LBD mutationslead to reduced hormone affinity or interference in heterodimerization,which results in complete or partial hormone insensitivity. Patientswith HVDRR develop severe rickets in the first months of life. Typicalserum characteristics include hypocalcemia, hypophosphatemia, ele-vated serum alkaline phosphatase, high serum 1,25-(OH)2D3 and sec-ondary hyperparathyroidism. All reported cases with a DBD mutationare associated with alopecia, which has only been observed in a fewcases with LBD mutations. Alopecia is persistent in HVDRR, even whenserum biochemistry has been restored to normal [1]. Cases of HVDRRwith alopecia generally reflect more severe grades of resistance and re-spond poorly, which require intensive calcium therapy to stimulateparacellular calcium uptake in the intestine [2].

Accumulating evidence indicates a high prevalence of vitamin Ddeficiency in the general population and this has been linked to preva-lence of not only autoimmune diseases but also bone diseases and can-cer [3,4]. Epidemiological studies have significantly correlated vitaminD

Table 1Biochemical analyses of and differences between patient #1 and #2 at presentation ofHVDRR.

Patient #1 Patient #2 Reference range

First signs of alopecia (yr) 3 1.5Calcium (mmol/l) 1.87a 2.23 2.2–2.6Phosphate (mmol/l) 1.28 0.61a 1.0–1.8Alkaline phosphatase (U/l) 954a 975a 55–250Parathyroid hormone (pmol/l) 19a 75.1a 1–725-Hydroxy vitamin D (nmol/l) 47 55 30–1001,25-Dihydroxyvitamin D3 (pmol/l) 790a 430.8a 39–102

a Different from reference range.

7B.C.J. van der Eerden et al. / Bone 69 (2014) 6–11

intake during early life with the incidence of type I diabetes at later ageand with the risk of developing rheumatoid arthritis [5,6]. Moreover,high serum 25-hydroxyvitamin D (25-(OH)D) levels (N99.1 nmol/l)were associatedwith reduced risk of multiple sclerosis [7]. Finally, diag-nosed systemic lupus erythromatosis patients had lower 25-(OH)Dlevels compared to controls [8]. As HVDRR patients lack a functionalVDR, an increased prevalence of chronic inflammatory diseases wasanticipated.

In this study, we performed VDR gene sequence analysis on DNAfrom a brother with 20-year follow-up and his sister (aged 6) withHVDRR. In addition, several assays on skin fibroblasts were performedto determine the precise consequences of the mutation on 1,25-(OH)2D3 signaling. Finally, peripheral blood mononuclear cells (PBMC)-derived T cell populationswere analyzed to obtain an adaptive immunesystem-related profile in HVDRR patients.

Material and methods

Patients

The first patient (#1) is a boy born in 1991 who was referred at theage of 1 year and 7 months because of unsteady walking. His parents,both nonsymptomatic, are first cousins of Moroccan origin (Fig. 1A).Studying the family history revealed one sister of the mother withshort stature and alopecia. Apart from the patient and his sister, bothparents and 3 other siblings did not have any symptoms. His heightwas 78 cm (−2.03 SDS) and his weight 13 kg (+0.65 SDS). His legswere in varus position while standing. Body and scalp hair was normal.Biochemical analyses (Table 1) showed decreased serum total calciumlevels and normophosphatemia whereas levels of alkaline phosphatase

1 2 3 4 5

I

II

1 2

A

B

II-2, II-3

II-1, II-4

I-1, I-2II-5

Fig. 1. Pedigree and DNA sequencing of VDR mutation. A) Pedigree of the family. The af-fected boy and girl are family members II-1 and II-4, respectively. B) DNA profiles of themutation region in the affected patients, their siblings and both parents.

(ALP) and parathyroid hormone (PTH) were increased. Serum 25-(OH)D level was normal and the level of 1,25-(OH)2D3 was elevated. Thesefindings implicated HVDRR. 1,25-(OH)2D3 (calcitriol, up to 30 μg/day)and calcium supplementation (500 mg 3 times daily) was initiated.At the age of three years, calcitriol was changed to 1 alpha-hydroxy vi-tamin D (alfacalcidol). The dosage was adjusted based on serum calci-um, phosphate and ALP levels as well as the clinical features of genuvarum deformity. Clinical, biochemical and radiological signs of ricketsnormalized at a dosage of 40 μg/day alfacalcidol around the age of 3.In the following years, the dosage of alfacalcidol was further decreased.From the age of 17, the boy becamemetabolically normal as was appar-ent from serum analyses at the age of 19 (Table 2), which was the lastyear of follow-up in our clinic. Interestingly, a progressive thinning ofhis scalp hair, especially at the temples, was observed from the age of3 years onwards and resulted in partial alopecia, reflecting areas oftotal baldness, adjacent to normal hair and regions of scant hair.

No (auto)-immune related diseases developed so far. X-rays ofhand and knee up to the age of 19 did not show any abnormalities.The second patient (#2) was a younger sister, born in 2007, who wasfollowed at the outpatient clinic because of the medical history of herbrother. Initially, she developed normally and biochemical analysesdid not reveal any abnormalities. However, at the age of 1.5 years, shepresented with growth retardation, genu varum and signs of partial al-opecia, similar to her brother. Biochemical analyses (Table 1) showednormocalcemia, hypophosphatemia, increased serum PTH and ALPlevels, normal 25-(OH)D and an elevated 1,25-(OH)2D3 level. An X-rayof thewrist showed rickets and HVDRRwas diagnosed. Following treat-ment with increasing dosages of alfacalcidol, normophosphatemia wasreached at the age of 2.5 years, while remaining normal serum calciumlevels. At present, at the age of 7 years, clinical rickets persists, as shownby the genu varumand elevated levels of serumALP, PTH and1,25-(OH)2D3 at a current dosage of 80 μg/day alfacalcidol. There was no report ofoligodontia in either patient.

Written informed consent was obtained from the parents.

Sequencing

Genomic DNA was isolated from peripheral blood of the two af-fected children, their siblings and their parents according to themanufacturer's protocol, using a Hamilton STAR (Hamilton RoboticsGmbh, Martinsried, Germany) apparatus and buffers from LGC ge-nomics (Agowa, Berlin, Germany). Oligonucleotide primers were

Table 2Biochemical analyses of patient #1 at 19 years of age.

Patient #1 Reference range

Calcium (mmol/l) 2.44 2.1–2.6Phosphate (mmol/l) 1.3 0.9–1.5Alkaline phosphatase (U/l) 116 b125Parathyroid hormone (pmol/l) 3.7 2–725-Hydroxy vitamin D (nmol/l) 54 30–1001,25-Dihydroxyvitamin D3 (pmol/l) 171a 39–102

a Different from reference range.

Table 3List of real-time qPCR primers used.

Gene Forward primer Reverse primer

CYP24A1 CATCATGGCCATCAAAACAA GTGTTGAGGCTCTTGTGCAGGAPDH AGCCACATCGCTCAGACAC GCCCAATACGACCAAATCC

8 B.C.J. van der Eerden et al. / Bone 69 (2014) 6–11

generated to amplify the 9 different exons of the human VDR gene.Sequencing was performed (BaseClear, Leiden, the Netherlands)and sequences were analyzed, using freely available software(Chromas lite, http://www.technelysium.com.au/chromas_lite.html).

GATA3 ACTACGGAAACTCGGTCAGG GGTAGGGATCCATGAAGCAGIL-17F GGCATCATCAATGAAAACCA TGGGGTCCCAAGTGACAGIL-2 AAGTGAAAGTTTTTGCTTTGAGCTA AAGTGAAAGTTTTTGCTTTGAGCTAIL-4 CACCGAGTTGACCGTAACAG GCCCTGCAGAAGGTTTCC

Functional studies in fibroblasts

Functional studies were performed in isolated fibroblasts fromskin biopsies of the boy, the mother and a healthy control. Firstly, theproliferative response of the skin fibroblasts was assessed followingtreatment with varying concentrations of 1,25-(OH)2D3. After 6 days,proliferation was measured, using a neutral red assay, [9,10]. Next, weassessed VDR number by performing binding studies, both in non-stimulated cells and following treatment with 10−8 M 1,25-(OH)2D3,as described previously [11]. Finally, the responsiveness of the fibro-blasts was assessed by preincubation of the fibroblasts with differentconcentrations of 1,25-(OH)2D3 for 4 h, extensive washing steps and1 h incubationwith 25-(OH)D. The formation of 24,25-dihydroxyvitaminD3 (24(R),25-(OH)2D3), as measured by high pressure liquid chroma-tography [12,13], reflected the activity of the 1,25-(OH)2D3-induced en-zyme 24-hydroxylase in the cells.

Cell preparation and stimulation of PBMC-derived T cells

PBMC samples were obtained from the boy with HVDRR, both hisparents and 2 healthy volunteers. PBMCs were prepared from heparin-ized blood by Ficoll–Hypaque Plus (GE Healthcare, Piscataway, NJ)density-gradient centrifugation. The cells were suspended in RPMI sup-plemented with 10%DMSO and 20% fetal calf serum (FCS) and stored inliquid nitrogen.

The cells were resuspended in phenol red-free Iscove's ModifiedDulbecco's Medium (IMDM; Gibco, Breda, the Netherlands) supple-mented with 10% charcoal-treated FCS, 100 units/ml penicillin, and100 μg/ml streptomycin (BioWhittaker, Walkersville, MD). Cells(100,000/well) were cultured for 72 h in 96-well plates (200 μl/well).Anti-CD3 at a dose of 300 ng/ml and anti-CD28 at a dose of 400 ng/ml(1XE and 15E8, respectively; Sanquin, Amsterdam, The Netherlands)were used to stimulate the cells towards the T-cell lineage in the pres-ence or absence of different concentrations (0.1 μM, 1 μM, and 10 μM)of 1,25-(OH)2D3 (kindly donated by Dr. L. Binderup, Leo PharmaceuticalProducts, Ballerup, Denmark). The 1,25-(OH)2D3 was dissolved in etha-nol at an initial concentration of 1 mM and serially diluted to the work-ing concentrations with IMDM medium.

Cytokine quantification

Cell culture supernatantswere stored at−20 °C formeasurement ofIL-17A, IL-22 (R&D Systems, Minneapolis, MN), TNFα, IFNγ, and IL-4(BioSource, Breda, The Netherlands) by specific enzyme-linked immu-nosorbent assay. The minimum detection limits for IL-17A, IL-22, IFNγ,IL-10 and IL-4 were 4, 4, 4, 2 and 2 pg/ml, respectively.

Real-time PCR

Themethods used for RNA extraction and complementary DNA syn-thesis have been described previously [14]. Primers were designedusing ProbeFinder software (Table 3), and probes were chosen fromthe Universal ProbeLibrary (Roche Applied Science). Real-time qPCRwas performed using the ABI Prism 7900 sequence detection system(Applied Biosystems), and the results were analyzed using SDS version2.3 software (Applied Biosystems). The obtained Ct valueswere normal-ized against Ct values for GAPDH.

Statistical analysis

The statistical significance of the data was analyzed with Prism 4.02software (GraphPad Software, San Diego, CA) using a Student's t-testunless stated otherwise. Data are given as the mean ± s.e.m. P valuesless than 0.05 were considered statistically significant.

Results

Sequencing

VDR gene sequencing of the patients, their siblings and parents re-vealed that both phenotypically affected children had a homozygous Ato Gmutation at position 133, leading to an amino acid change from ly-sine (K) to glutamic acid (E) at position 45 in exon 2 of the VDR gene(Fig. 1B). This mutation was identical to one of the 2 mutations previ-ously described by Rut and coworkers [15] but as far as we know, thepatients are not related. In the other sequenced exons, no mutationswere found. Both parents and a brother were heterozygous for thesame mutation, whereas two sisters were genetically unaffected(Fig. 1B). This exon is part of the DNA binding domain of VDR geneand indicated to be damaging in differentweb-based predictionmodels(Supplementary Table 1).

Skin fibroblast cultures

Skin fibroblasts were collected from the male patient, his motherand from a healthy control to study the cellular response to 1,25-(OH)2D3 and functionality of the VDR. 1,25-(OH)2D3 did not alterfibro-blast proliferation in the patients' fibroblasts, whereas in control fibro-blasts proliferation was strongly inhibited (Fig. 2A). Proliferation wasalso inhibited in the fibroblasts from the mother, but this was lesspronounced compared to the control (Fig. 2A). Next, the effect of1,25-(OH)2D3 treatment on the 24 hydroxylase activity was measured.24(R),25-(OH)2D3 was dose-dependently synthesized upon 1,25-(OH)2D3 stimulation in fibroblasts from both the mother and thehealthy control (Fig. 2B). No 24(R),25-(OH)2D3 was synthesized in thefibroblasts from the patient, despite incubation with high concentra-tions of 1,25-(OH)2D3 (1 μM), (Fig. 2B). VDR number was determinedin the fibroblasts on basis of a binding assay. VDR was expressed bothin the patient and the mother and appeared even more numerousthan in the control fibroblasts (Fig. 2C), showing that VDR is able tobind 1,25-(OH)2D3. 1,25-(OH)2D3 induced homologous up-regulationof the VDR in mothers' fibroblasts but not those of the patient(Fig. 2D). Collectively, these data show that despite its presence andbinding of 1,25-(OH)2D3 VDR fails to elicit a downstream response.

Absent Th1, Th17 and Th2 cell response in HVDRR patient

Weexplored the effect of vitaminD resistance in T-cell differentiationfrom the patient's PBMCs. 1,25-(OH)2D3 strongly induced CYP24A1mRNA expression in the parent's PBMCs but not in those obtainedfrom the patient, confirming the fibroblast observations (Table 4).

We assessed expression levels and protein production of pro-inflammatory (Th1 andTh17), anti-inflammatory (Th2) andT regulatory

C

B

10-8 10-7 10-6

% o

f con

trol

405060708090

100110120 Child

MotherControl

1,25(OH)2D3concentration (M)

0

10

20

30

40

50

60

70 ControlChildMother

VD

R (f

mol

boun

d/m

g pr

otei

n)

0

5

10

15

20 ChildMotherControl

10-8 10-610-10024,2

5(O

H) 2

D3

(% o

f tot

aldp

ms)

1,25(OH)2D3concentration (M)

0

10

20

30

40

50

60

70

Child MotherVD

R (f

mol

boun

d/m

g pr

otei

n)

Basal10-8M D3

D

A

Fig. 2. Functional analyses in skin fibroblasts. Skin fibroblasts from the patient and his mother were compared to control fibroblasts with respect to A) 1,25-(OH)2D3-inhibited cell prolif-eration, B) 24-hydroxylase activity and C) VDR protein expression. Finally, D) 1,25-(OH)2D3-stimulated VDR expression was compared between the patient and his mother.

9B.C.J. van der Eerden et al. / Bone 69 (2014) 6–11

cell markers. 1,25-(OH)2D3 strongly reduced protein and/or RNA ex-pression levels of the pro-inflammatory cytokines IFNγ, IL-17A, IL-2,IL-17 F, IL-22 in PBMCs of the parents but not in those of the patient(Fig. 3 and Table 4). The mRNA and protein level of the anti-inflammatory cytokine IL-4, representing the Th2 cell population,were increased by 1,25-(OH)2D3 in the parents but unaffected in the pa-tient (Table 4). IL-10 production, representative for Tregs, was unaffect-ed by 1,25-(OH)2D3 treatment both in the patient and his parents(Table 4).

Discussion

In this study we present two children from one family fromMoroccan descent with a homozygous mutation (Lys45Glu) in theDNA binding domain of the human VDR, rendering them resistant to

Table 41,25-(OH)2D3 treatment of PBMC-derived T-cells from a patient with HVDRR.

T/C ratio's Patient Father Mother

Protein expressionPro-inflammatory T-cell

IFNγ 1.00 ± 0.21 0.15 ± 0.03 0.11 ± 0.03IL-17A 1.12 ± 0.58 0.17 ± 0.06 0.32 ± 0.01IL-22 0.58 ± 0.13 0.05 ± 0.00 0.12 ± 0.00

Anti-inflammatory T-cellIL-4 0.43 ± 0.05 1.80 ± 0.17 2.20 ± 0.20

Regulatory T-cellIL-10 1.20 ± 0.55 0.91 ± 0.45 0.80 ± 0.34

Gene expressionCYP24A1 1.09 ± 0.58 3937 ± 608 1358 ± 476Pro-inflammatory T-cell

IL-2 1.48 ± 0.09 0.35 ± 0.08 0.44 ± 0.05IL-17F 1.50 ± 0.35 0.38 ± 0.12 0.50 ± 0.03

Anti-inflammatory T-cellIL-4 1.11 ± 0.35 1.98 ± 0.06 3.29 ± 0.39

Values indicate average ratio's ± s.e.m. (n = 3) of 10 μM 1,25-(OH)2D3 treatment overcontrol (T/C). Gene expression was corrected for the housekeeping gene GAPDH.

1,25-(OH)2D3. The patients suffered from classical features of HVDRRincluding the well-described effects on calcium homeostasis and bone,although some differences between the siblings in serum biochemistryand onset of alopecia were noted. Despite the incapability of immunecells to respond to vitamin D, to date no immune-related challengeshave occurred in these patients.

Mutation in VDR leads to vitamin D resistance

The Lys45Glu mutation led to 1,25-(OH)2D3 resistance as shownby unaltered fibroblast proliferation, absent VDR upregulation andfailure to stimulate 24-hydroxylase activity, the most sensitive 1,25-(OH)2D3 target. In fact, all 9 mutations discovered so far in the DBD ofthe humanVDRdisplay these characteristics, of which themajority con-veys normal 1,25-(OH)2D3 binding, just as in our study [1]. The muta-tion described in this study has been published before by Rut andcolleagues [15], who demonstrated absent 24-hydroxylase activity aswell as a failure to induce osteocalcin gene expression as measured bychloramphenicol activity in skin fibroblasts [15].With VDR being analo-gous to the crystal structures of the glucocorticoid and estrogen recep-tor, this group elegantly utilized these receptors as models to study indetail the interaction of the VDR DNA binding domain to the binding el-ement (VDRE). The Lys45Glumutation is located in theDNA recognitionhelix of the DNA binding domain. The lysine residue has a stabilizingrole by donating hydrogen bonds to both a neighboring glutamic acidand a guanine molecule at the corresponding position of the VDRE,and this is lost when mutated to glutamate. Besides, the mutationdisrupts its own binding to DNA by attracting a hydrogen bond froma neighboring histidine residue. Collectively, the data point to 1,25-(OH)2D3 resistance due to failure of VDR binding to vitamin D responseelements in the DNA.

Serum calcium/phosphate differences between the siblings

A few biochemical and phenotypical differences were noted whencomparing the siblings. Firstly, where the boy was initially diagnosed

A BIFNy

Patient Father Mother0.0

0.5

1.0

1.5

**

T/C

IL-17A

Patient Father Mother0.0

0.5

1.0

1.5

2.0

**

T/C

Fig. 3. Protein levels of PBMC-stimulated Th1 and Th17 cell markers. Protein production of A) IFNγ and B) IL-17A was assessed in 1,25-(OH)2D3-stimulated PBMC-derived T-cells (blackbars) vs controls (white bars) from patient #1 and both parents. Values indicate average ratio's ± s.e.m. of 10 μM 1,25-(OH)2D3 treatment over control (T/C).

10 B.C.J. van der Eerden et al. / Bone 69 (2014) 6–11

with hypocalcemia, the girl presentedwith low serumphosphate levels.Despite phosphate being normal in the boy and calcium in the girl, bothlevels were in the lower reference range. We have neither explanationas to why the biochemical phenotypes were different between thesiblings with HVDRR, nor any reports describing this in the literature.Secondly, the girl had alopecia at a younger age compared to the boy.One other study reported on two siblings, where the girl developed al-opecia at young age (b1 year), while the boy had sparse hair on thehead but did not develop total alopecia [16]. However, given the smallnumber of cases, no statements can be made regarding sex and onsetof alopecia.

Normalized calcium and phosphate homeostasis without currentmedication

Intriguingly, the boy showed complete recovery (except for alope-cia) from the age of 17 and did not require 1,25-(OH)2D3 therapy todate, being 22 years old. Moreover, no abnormalities were detected inthe skeleton, using X-rays. This is in line with the findings by Tiosanoet al., who did not detect skeletal differences between HVDRR patientsand controls, using dual-energy X-ray absorption [17]. An explanationfor the recovery of calcium homeostasis may result from the observa-tion that fractional calcium absorption in late adolescents and youngadults is elevated compared to young children, which is only evidentin subjects on low calcium intake [17]. The exact mechanism is yetunclear but intestinal calcium uptake can be stimulated by estrogen,independently of 1,25-(OH)2D3, as shown in a rat study following17β-estradiol treatment [18]. These data suggest that when calcium ho-meostasis is under pressure, such as in HVDDR patients, both male andfemale young adolescents including our patient are capable of absorbinga surplus of calcium, potentially by estrogens, thereby achieving calciumhomeostasis in the absence of 1,25-(OH)2D3 signaling [17]. Closemonitoring is required though, since Tiosano et al. also reported that el-evated fractional calcium uptake potential in these patients lasts untilthe end of the third decade of life, after which vitamin D dependencemay return [17]. Besides increased gastrointestinal calcium uptake, theage of the boy at the time of complete recovery coincided with theend of puberty, when skeletal growth ceases and the need for calciumdecreases. Although there is no evidence to support that reduced calci-um requirement at the end of puberty directly alleviates dependence onvitamin D administration in HVDRR, it may partially explainmedicationindependence in late adolescent HVDRR patients. This, however, seemsto contradict the observations by Tiosano et al. [15] that recurrence ofvitamin D-dependency ensues in the third decade of life of HVDRRpatients.

No signs of inflammatory episodes despite vitamin D deficiency

A few years ago, the Institute of Medicine (IOM) installed a commit-tee that issued a vitaminD guideline demonstrating that vitaminDhas a

beneficial effect on skeletal health [19]. Unlike the skeleton, thecommittee concluded that currently there is no proof for beneficial ef-fects of vitamin D for nonskeletal outcomes, such as cardiovascular dis-ease, quality of life and (auto)immune diseases. In fact, randomized,double-blinded, controlled, prospective studies are needed to providean objective link between vitamin D levels and nonskeletal conditions,which are crucially lacking to date. Only when clinical studies fulfilthese criteria, health-related institutions are inclined to accept and in-stall observed associations.

Several studies have investigated a potential link between vitaminD and the prevalence of (auto)immune diseases. Epidemiological stud-ies have significantly correlated vitamin D intake during early life withthe incidence of type I diabetes at a later age and with the risk of devel-oping rheumatoid arthritis [5,6].Moreover, high serum25-(OH)D levels(N99.1 nmol/l) were associated with reduced risk of multiple sclerosis[7]. However, due to the nature of these studies, no causative relation-ships can bemade. Along these lines,we anticipated a higher prevalenceof chronic inflammatory diseases in HVDRR. VitaminD appears tomain-tain a balance between inflammatory Th1/Th17 cells and immunosup-pressive Th2/Treg cells [20]. Hence, suggesting an important role ofvitamin D in the immune system and follow up of diagnosed vitaminD deficient patients seems relevant [21]. Our patient with HVDRR didnot respond to vitamin D stimulation in terms of pro-inflammatory(Th1 and Th17) and anti-inflammatory (Th2) cell markers, whereashis parents demonstrate a 1,25-(OH)2D3-initiated reduction of Th1and Th17 cell markers and induction of Th2/Treg markers. In concor-dancewith our data, impairments in a proinflammatory cytokine profilein lymphocytes of HVDRR patients were recently reported by Tiosanoand colleagues [22]. These findings indicate that the patient may beprone to immunogenic challenges. However, the long-term follow upof this patient has not revealed any immune-related problems to date.This questions the necessity of vitamin D for the immune system and/or it points to compensatory mechanisms that are recruited to protectagainst (chronic) inflammatory-related diseases. One of these mecha-nismsmay be non-genomic actions of vitamin D, which take place inde-pendent of the classical VDR. Although no evidence is provided for thismechanism in immune-related disorders, a membrane-bound 1,25D3membrane associated rapid response-binding protein (1,25D3MARSS)also known as protein-disulfide isomerase-associated 3 (Pdia3) wasfound to play a role in rapid responses to 1,25-(OH)2D3[23]. In addition,there are reports describing non-genomic effects of vitamin D indepen-dent of vitamin D binding in mouse osteoblasts as well as in basal cellcarcinomas ([24–26]).

Conclusions

In conclusion, we have characterized two siblings with HVDRR, har-boring a homozygousmutation in the DNAbinding domain of VDR, ren-dering them vitamin D resistant. The adaptive immune system wasunresponsive to vitamin D but to date no history of infections is evident

11B.C.J. van der Eerden et al. / Bone 69 (2014) 6–11

for the siblings. Intriguingly, the boy does not need medication atpresent at the age of 22. Nevertheless, we believe that close monitoringof these patients is warranted for the potential recurrence of vitamin Ddependency of the calcium and phosphate homeostasis and occurrenceand early diagnosis of (chronic) inflammatory diseases.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.bone.2014.08.005.

Acknowledgments

We thank Cok Buurman for the technical assistance.

References

[1] Malloy PJ, Tiosano D, Feldman D. Hereditary 1,25-dihydroxyvitamin-D-resistantrickets. In: Feldman D, Pike JW, Adams JS, editors. Vitamin D. 3rd ed. Amsterdam:Elsevier; 2011. p. 1197–232.

[2] Hochberg Z, Tiosano D, Even L. Calcium therapy for calcitriol-resistant rickets. JPediatr 1992;121:803–8.

[3] Holick MF. Vitamin D, deficiency. N Engl J Med 2007;357:266–81.[4] CantornaMT, Mahon BD. Mounting evidence for vitamin D as an environmental fac-

tor affecting autoimmune disease prevalence. Exp Biol Med (Maywood) 2004;229:1136–42.

[5] Hypponen E, Laara E, Reunanen A, JarvelinMR, Virtanen SM. Intake of vitamin D andrisk of type 1 diabetes: a birth-cohort study. Lancet 2001;358:1500–3.

[6] Merlino LA, Curtis J, Mikuls TR, Cerhan JR, Criswell LA, Saag KG. Iowa Women'sHealth S. Vitamin D intake is inversely associated with rheumatoid arthritis: resultsfrom the Iowa Women's Health Study. Arthritis Rheum 2004;50:72–7.

[7] Munger KL, Levin LI, Hollis BW, Howard NS, Ascherio A. Serum 25-hydroxyvitamin Dlevels and risk of multiple sclerosis. JAMA 2006;296:2832–8.

[8] Kamen DL, Cooper GS, Bouali H, Shaftman SR, Hollis BW, Gilkeson GS. Vitamin D de-ficiency in systemic lupus erythematosus. Autoimmun Rev 2006;5:114–7.

[9] Abe T, Hara Y, Abe Y, Aida Y, Maeda K. Serum or growth factor deprivation inducesthe expression of alkaline phosphatase in human gingival fibroblasts. J Dent Res1998;77:1700–7.

[10] Abe T, Abe Y, Aida Y, Hara Y, Maeda K. Extracellular matrix regulates induction of al-kaline phosphatase expression by ascorbic acid in human fibroblasts. J Cell Physiol2001;189:144–51.

[11] Pols HA, van Leeuwen JP, Schilte JP, Visser TJ, Birkenhager JC. Heterologous up-regulation of the 1,25-dihydroxyvitamin D3 receptor by parathyroid hormone(PTH) and PTH-like peptide in osteoblast-like cells. Biochem Biophys Res Commun1988;156:588–94.

[12] Pols HA, Schilte HP, Nijweide PJ, Visser TJ, Birkenhager JC. The influence of albuminon vitamin D metabolism in fetal chick osteoblast-like cells. Biochem Biophys ResCommun 1984;125:265–72.

[13] Pols HA, Birkenhager JC, Schilte JP, Bos MP, van Leeuwen JP. The effects of MC903 on1,25-(OH)2D3 receptor binding, 24-hydroxylase activity and in vitro bone resorp-tion. Bone Miner 1991;14:103–11.

[14] van Hamburg JP, Mus AM, de Bruijn MJ, de Vogel L, Boon L, Cornelissen F, et al.GATA-3 protects against severe joint inflammation and bone erosion and reducesdifferentiation of Th17 cells during experimental arthritis. Arthritis Rheum 2009;60:750–9.

[15] Rut AR, HewisonM, Kristjansson K, Luisi B, Hughes MR, O'Riordan JL. Twomutationscausing vitamin D resistant rickets: modelling on the basis of steroid hormone re-ceptor DNA-binding domain crystal structures. Clin Endocrinol (Oxf) 1994;41:581–90.

[16] Malloy PJ, Weisman Y, Feldman D. Hereditary 1 alpha,25-dihydroxyvitamin D-resistant rickets resulting from a mutation in the vitamin D receptor deoxyribonu-cleic acid-binding domain. J Clin Endocrinol Metab 1994;78:313–6.

[17] Tiosano D, Hadad S, Chen Z, Nemirovsky A, Gepstein V, Militianu D, et al. Calcium ab-sorption, kinetics, bone density, and bone structure in patients with hereditary vita-min D-resistant rickets. J Clin Endocrinol Metab 2011;96:3701–9.

[18] Colin EM, Van Den Bemd GJ, Van AkenM, Christakos S, De Jonge HR, Deluca HF, et al.Evidence for involvement of 17beta-estradiol in intestinal calcium absorption inde-pendent of 1,25-dihydroxyvitamin D3 level in the rat. J Bone Miner Res 1999;14:57–64.

[19] Institute of Medicine. Dietary Reference Intakes for Calcium and Vitamin D.Washington: National Academies Press; 2011.

[20] Hewison M. Vitamin D, and immune function: an overview. Proc Nutr Soc 2012;71:50–61.

[21] Adorini L, Penna G. Control of autoimmune diseases by the vitamin D endocrine sys-tem. Nat Clin Pract Rheumatol 2008;4:404–12.

[22] Tiosano D,Wildbaum G, Gepstein V, Verbitsky O,Weisman Y, Karin N, et al. The roleof vitamin D receptor in innate and adaptive immunity: a study in hereditary vita-min D-resistant rickets patients. J Clin Endocrinol Metab 2013;98:1685–93.

[23] Nemere I, Safford SE, Rohe B, DeSouza MM, Farach-Carson MC. Identification andcharacterization of 1,25D3-membrane-associated rapid response, steroid (1,25D3-MARRS) binding protein. J Steroid Biochem Mol Biol 2004;89–90:281–5.

[24] Wali RK, Kong J, Sitrin MD, Bissonnette M, Li YC. Vitamin D receptor is not requiredfor the rapid actions of 1,25-dihydroxyvitamin D3 to increase intracellular calciumand activate protein kinase C inmouse osteoblasts. J Cell Biochem 2003;88:794–801.

[25] Tang JY, Xiao TZ, Oda Y, Chang KS, Shpall E, Wu A, et al. Vitamin D3 inhibits hedge-hog signaling and proliferation in murine basal cell carcinomas. Cancer Prev Res(Phila) 2011;4:744–51.

[26] Uhmann A, Niemann H, Lammering B, Henkel C, Hess I, Nitzki F, et al. Antitumoraleffects of calcitriol in basal cell carcinomas involve inhibition of hedgehog signalingand induction of vitamin D receptor signaling and differentiation. Mol Cancer Ther2011;10:2179–88.