CYP2R1 Mutations Impair Generation of 25-hydroxyvitamin D and Cause an Atypical Form of Vitamin D...

CYP2R1 Mutations Impair Generation of 25-hydroxyvitamin D and Cause an Atypical Form ofVitamin D Deficiency

Tom D. Thacher, M. D,1, Philip R. Fischer, M.D,2, Ravinder J. Singh, Ph.D,3,Jeffrey Roizen, M.D, Ph.D4, Michael A. Levine, M.D4

1 Department of Family Medicine, Jos University Teaching Hospital, Jos, Nigeria and the Department ofFamily Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905; 2 Department of Pediatric andAdolescent Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905; 3 Department of LaboratoryMedicine and Pathology, Mayo Clinic, 200 First St SW, Rochester, MN 55905; 4 The Children’s Hospitalof Philadelphia and Department of Pediatrics, University of Pennsylvania Perelman School of Medicine,34th St & Civic Center Blvd, Philadelphia, PA 19104 Key words: metabolic bone, rickets, genetics,vitamin D, nutrition

Context: Production of the active vitamin D hormone 1,25-dihydroxyvitamin D requires hepatic25-hydroxylation of vitamin D. The CYP2R1 gene encodes the principal vitamin D 25-hydroxylasein humans.

Objective: To determine the prevalence of CYP2R1 mutations in Nigerian children with familialrickets and vitamin D deficiency and assess the functional effect on 25-hydroxylase activity.

Design and Participants: We sequenced the CYP2R1 gene in subjects with sporadic rickets andaffected subjects from families in which more than one member had rickets.

Main outcome measures: Function of mutant CYP2R1 genes as assessed in vivo by serum 25-hydroxyvitamin D values after administration of vitamin D and in vitro by analysis of mutant formsof the CYP2R1.

Results: CYP2R1 sequences were normal in 27 children with sporadic rickets, but missense muta-tions were identified in affected members of 2 of 12 families, a previously identified L99P and anovel K242N. In silico analyses predicted that both substitutions would have deleterious effects onthe variant proteins, and in vitro studies showed that K242N and L99P had markedly reduced orcomplete loss of 25-hydroxylase activity, respectively. Heterozygous subjects were less affectedthan homozygous subjects, and oral administration of vitamin D led to significantly lower increasesin serum 25-hydroxyvitamin D in heterozygous than in control subjects, while homozygous subjectsshowed negligible increases.

Conclusion: These studies confirm that CYP2R1 is the principal 25-hydroxylase in humans, anddemonstrate that CYP2R1 alleles have dosage-dependent effects on vitamin D homeostasis.CYP2R1 mutations cause a novel form of genetic vitamin D deficiency with semidominantinheritance.

Vitamin D deficiency is prevalent worldwide (1–3) andremains the most common cause of rickets in chil-

dren and osteomalacia in adults. Moreover, low vitamin Dstatus is associated with reduced bone density and in-

creased fracture risk in the elderly (4, 5). A growing num-ber of association studies have also implicated low vitaminD status as a potential risk factor for diabetes mellitus,hypertension, malignancy, infection, and immune diseases

ISSN Print 0021-972X ISSN Online 1945-7197Printed in U.S.A.Copyright © 2015 by the Endocrine SocietyReceived March 24, 2015. Accepted May 1, 2015.

Abbreviations:

O R I G I N A L A R T I C L E

doi: 10.1210/jc.2015-1746 J Clin Endocrinol Metab jcem.endojournals.org 1

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 July 2015. at 17:43 For personal use only. No other uses without permission. . All rights reserved.

(3, 6–10), and optimal vitamin D status is a topic of activeinvestigation and ardent controversy (11, 12). Vitamin D,produced in the skin as vitamin D3 (cholecalciferol) afterexposure to ultraviolet (UV) light, or supplied in the dietas vitamin D3 or vitamin D2 (ergocalciferol), must un-dergo 25-hydroxylation in the liver (13) by CYP2R1, togenerate 25-hydroxyvitamin D(25(OH)D). Further hy-droxylation in the kidneys by the 25(OH)D-1�-hydrox-ylase enzyme CYP27B1 generates 1,25-dihydroxyvitaminD [1,25(OH)2D], the active hormone responsible for mostof the physiological actions of vitamin D.

Because few foods naturally contain or are fortifiedwith vitamin D, the principle source of vitamin D for mostpopulations is cutaneous production of vitamin D3 fromsunlight. Thus, it has been surprising that vitamin D de-ficiency remains prevalent in sun-enriched parts of theworld (14–17), and that rickets occurs in children who livein tropical countries (1). We sought to identify potentialgenetic causes of rickets in a cohort of children in centralNigeria (latitude 10°N) that responds more effectively tocalcium supplementation than to conventional doses ofvitamin D (18, 19). These subjects had low levels of25(OH)D, which suggested a potential defect in 25-hy-droxylation of vitamin D. Hence, we examined theCYP2R1 gene, which encodes the principle 25-hydroxy-lase, as recessive mutations in the human CYP2R1 gene(20–22) and disruption of murine Cyp2r1 (23) by con-ventional gene targeting result in markedly reduced serumlevels of 25(OH)D.

Materials and Methods

Subjects. Our cohort of rachitic Nigerian children included 27children with sporadic rickets (serum concentration of 25(OH)D12.1 � 4.8 ng/mL, mean � SD) and 12 subjects from families(serum concentration of 25(OH)D 11.4 � 4.5 ng/mL) with morethan one first-degree relative with rickets. Twenty-two first-de-gree relatives of 12 index cases had a history of leg deformitiesconsistent with rickets, and 14 were included in this study. Weconfirmed rickets in all index cases and most siblings with ricketsusing a rickets radiographic score of greater than 1.5 on a 10-point severity scale (24). The control group consisted of 21 nor-mal children between 19 and 59 months of age (mean � SD,35.7 � 11.9 months) who had been previously characterized(25). We collected medical and demographic data from eachsubject, with particular emphasis on skeletal deformities consis-tent with rickets, and measured biochemical parameters of boneand mineral metabolism. All studies were approved by the ap-propriate institutional review boards, and written informed con-sent was received from all patients or their parents prior to in-clusion in the study.

Laboratory analyses and gene sequencing. We measured serumelectrolytes and creatinine using routine methods. For measure-ments of vitamin D metabolites, blood samples were centrifuged

within 30 minutes after collection, and serum samples werestored at –70°C until shipped frozen to the Mayo Clinic foranalysis. Measurements of vitamin D3, vitamin D2, 25(OH)D3,and 25(OH)D2 were performed by isotope-dilution liquid chro-matography tandem mass spectrometry (LC-MS/MS) using anAPI 4000 instrument (Applied Biosystems, Forest City, CA). Wemeasured serum concentrations of 1,25(OH)2D by radioimmu-noassay (RIA) (DiaSorin, Stillwater, MN). Intact PTH was mea-sured using the Immulite 2000 PTH assay (Diagnostics ProductCorporation, Los Angeles, CA).

We extracted genomic DNA from dried blood spots or salivausing standard methods. All five exons containing the codingregion, the exon-intron junctions, and the promoter of CYP2R1(11p15.2) were sequenced as previously described (26). Nucle-otide position was numbered according to the starting point ofthe ATG codon in the complementary DNA (cDNA) (sequenceaccession number NM 024514.4). The predicted effect of thevariants was determined by in silico analyses using a web-basedtool, Condel (27), which assesses the outcome of amino acidchanges using a consensus deleteriousness score that combinesvarious tools (eg, SIFT, Polyphen2, MutationAssessor) (28–30).We used the HOPE tool (31) to determine the effect of variationon the 3D structure of the CYP2R1 protein.

We performed linkage analysis to determine potential asso-ciations of CYP2R1 alleles with rickets by genotyping membersof all families for single nucleotide polymorphisms (SNPs) lo-cated within the CYP2R1 gene (rs12794714, rs7936142,rs1993116, rs1740157, rs10500804, rs11023373) (32).Genomic DNA from 59 unrelated subjects of Nigerian originwas genotyped for mutations and used as a population control.In addition, we analyzed CYP2R1 sequence data from 628 un-related subjects in the 1000 Genomes Project.

Assessment of 25-hydroxylase activity in vivo. To assess thefunctional capacity of 25-hydroxylase in vivo, we administered50 000 IU orally of vitamin D2 (Pliva, Inc. East Hanover, NJ) orvitamin D3 (Bio-Tech, Fayetteville, AR) on separate occasions atleast three months apart to patients with CYP2R1 mutations,their unaffected first-degree relatives, and control subjects (25).

Cell culture, analysis of Cyp2r1 in HEK293T cells, and statisti-cal methods. See online supplemental methods.

Results

Molecular genetic studies. CYP2R1 sequences were nor-mal in all 27 children with sporadic rickets (data notshown). By contrast, probands from two of twelve Nige-rian families with more than one member affected by rick-ets carried CYP2R1 mutations. In a micro linkage analy-sis, there was concordance between CYP2R1 alleles andrickets and/or low serum levels of 25(OH)D in affectedmembers of both of these two families (Figure 1). Table 1lists relevant characteristics of the patients in these twofamilies. By contrast, probands from the other ten familiesdid not carry CYP2R1 mutations, and there was discor-dance between CYP2R1 haplotypes and transmission ofrickets in these kindreds (data not shown).

2 CYP2R1 Mutations and Vitamin D Deficiency Rickets J Clin Endocrinol Metab


We identified the previously described c296T�C(L99P) CYP2R1 mutation (21) in eight subjects in twogenerations in both Families 1 and 2. The mutation was inlinkage disequilibrium with a consistent haplotype across45 kb of gDNA on the CYP2R1 locus. This haplotype wasnot present in any of the 59 subjects in the 10 other familiesthat we analyzed and hence is rare. Accordingly, these twoapparently unrelated Nigerian pedigrees show identity bystate at the CYP2R1 locus on the mutant chromosome,and likely share a common origin with the L99P allele wepreviously identified in an affected member of a third fam-ily with rickets that originated in Nigeria (21).

The mother of the proband in family 2 was heterozy-gous for a substitution of cytosine for adenine at nucleo-tide position 726 (c.726A3C) in exon 3 of the CYP2R1gene (Figure 3C). This missense mutation replaces lysinewith asparagine at amino acid 242 of the Cyp2r1 protein(p.K242N).

Both the L99P and K242N mutations were absent inCYP2R1 alleles from 59 unrelated subjects of Nigerianorigin and 628 unrelated subjects in the 1000 GenomesProject, thereby excluding the possibility that the identi-fied mutations represent population-specific sequencevariants. Moreover, Leucine99 and Lysine 242 are bothconserved in the CYP2R1 enzymes of mammals, chickens,and fish, suggesting that these residues play importantroles in protein function or structure. The effects of thetwo point mutations were analyzed using the bioinfor-matics software Condel, which predicted that the L99Pand K242N amino acid replacements were both deleteri-ous. Replacement of Leucine 99 has been proposed toimpair CYP2R1 folding (33), while our in silico analysis ofthe K242N mutation predicts that this amino acid substi-tution disturbs the interaction surface (31). Replacementof leucine99 by proline, a known helix breaker, disruptsthe hydrogen bond network and sterics of the helix. Re-placement of lysine242 by asparagine is predicted to de-

stabilize positioning of Phe240 withconsequent decreased interaction ofCYP2R1 with its substrate, parentvitamin D.

Clinical characterization of family1. The index case in Family 1 (Figure1A) was homozygous for a missensemutation c.296T�C in exon 2 of theCYP2R1 gene (Figure 2D), which re-sults in substitution of proline forleucine at position 99 (p.L99P). Inaddition to the proband (II-1), hisbrother with rickets (II-2), and theirfather (I-1) were also homozygousfor the L99P mutation, and two sis-

ters and the mother were heterozygous for the L99P allele.Both brothers who were homozygous for the L99P mu-tation had severe rickets.

The proband (Figure 1A, II-1; Figure 2A and 2B; Table1) presented at age 12.5 years with leg pain, and hadmarked anterior tibial bowing that was first noted at age2 years. He had begun to walk at age 9 months but stoppedat age 3 years. His typical daily dairy product calciumintake was 50 mg. Rib beading and wrist enlargementwere prominent, and radiographs and biochemistries wereconsistent with severe rickets. The serum 25(OH)D con-centration was 8 ng/mL. As part of a clinical trial (34), hehad a minimal response to six months of supplementalcalcium (reduction in radiographic severity score from10.0 to 8.0), and three months after additional treatmentwith intramuscular (IM) vitamin D3 600 000 IU (Hamex-medica Ltd., London, UK) his radiographic score declinedfrom 8 to 3.5. A second dose of vitamin D3 was given, and3 months later the radiographic score was 0.5, indicatingthe rickets was healed. He began walking unaided, but hisdeformities persisted.

The proband’s brother (Figure 1A, II-2 and Figure 2C;Table 1) presented simultaneously at age 11 years with legpain and genu valgum that had been first noted at age 6.5years. Walking had been delayed until age 3 years. Histypical daily dairy product calcium intake was only 15 mgand his serum level of 25(OH)D was 4.1 ng/mL. Radio-graphs showed severe rickets, and the biochemical fea-tures were similar to those of his brother. In the sameclinical trial as the proband (34), he had a favorable clin-ical response to treatment with calcium supplementationwith reduction in the radiographic severity score from 9.0to 1.0 (near complete healing), resolution of leg pain andimprovement in genu valgum. The serum 25(OH)D con-centration remained low, and other biochemical features

Figure 1. Pedigrees and Single Nucleotide Polymorphism Haplotype Analysis. (A)Family 1. (B) Family 2. The presence of a CYP2R1 mutation is noted under each subject, and thegenotypes for single nucleotide polymorphisms in and around the CYP2R1 gene are shown in thehaplotype boxes at the bottom of each pedigree. The CYP2R1 allele carrying the L99P mutationis shaded in gray. Probands are indicated by the arrows.

doi: 10.1210/jc.2015-1746 jcem.endojournals.org 3


of rickets remained unchanged, however. He developedscoliosis in adolescence.

The 49 year-old father (Table 1 and Figure 1A, I-1) hadno history of childhood skeletal deformities. At age 24years, he developed spastic paraparesis due to tuberculosis(TB) of the spine. He recovered after antituberculosistreatment, but he had a relapse of paraparesis at age 44years and has remained unable to walk. A spine radio-graph at age 46 years showed reduced mineralization andmultiple thoracic vertebral compression fractures; radio-graphs of the pelvis, hips, femurs, and scapulae were nor-mal, without pseudofractures of osteomalacia. His serum25(OH)D concentration was 4.9 ng/mL and PTH wasmodestly elevated.

One of the proband’s sisters (Figure 1A, II-3) had ahistory of mild genu varum that had resolved spontane-ously as a toddler, and the mother and the other sister hadno history of bone deformity or pain

Clinicalcharacterizationoffamily2.Theprobandoffam-ily 2 (Figure 1B, II-1) was heterozygous for the c.296T�Cmissense mutation, p.L99P, as was her affected sister (Fig-ure 1B, II-2) and father (Figure 1B and Figure 3B). Bothchildren had clinical (Figure 3A), biochemical (Table 1),and radiographic (Figure 2A and 2B) evidence of rickets.Their father (I-1) had a history compatible with childhoodrickets. A younger sister (II-3) was clinically unaffected.The proband and her affected sister had presented withbowed legs at ages 62 and 32 months with radiographic

scores of 5.0 and 2.0, respectively.The typical diet of the proband pro-vided a daily calcium intake of 175mg. Her serum 25(OH)D concentra-tion was 16.4 ng/mL. Both affectedsisters were treated with calcium andthe proband had also received oralvitamin D2 50 000 IU monthly for 6months in a clinical trial (35). Aftertreatment, rickets had resolved inboth subjects (radiographic scores of1.0 and 1.5, respectively). Themother had no history of bone de-formity, but she had an elevated se-rum PTH concentration and a lowserum 25(OH)D concentration.

Response to oral vitamin Dchallenge. Affected members ofFamilies 1 and 2 were given a singledose of 50 000 IU orally of vitaminD2 or vitamin D3 on separate occa-sions. Serum concentrations of par-ent vitamin D2 and vitamin D3 were

less than 5 ng/mL in all study and control subjects at base-line, and showed robust increases one day after adminis-tration of the respective vitamin D compound (Figure 4A).The three subjects who were homozygous for L99P hadlower baseline 25(OH)D3 concentrations (3.1 � 1.7 ng/mL; P � .001) and showed significantly blunted responsesto oral vitamin D3 compared with individuals who wereheterozygous for the L99P or K242N mutations or normalsubjects (Figure 4B). Subjects who were heterozygous forthe L99P mutations had lower baseline 25(OH)D3 con-centrations than control children (13.4 � 2.3 ng/mL vs.25.9 � 6.1 ng/mL, respectively; P � .001), and showedsubnormal responses to administration of vitamin D3. Thebaseline 25(OH)D3 concentration for the subject who washeterozygous for the K242N mutation was 15.6 ng/mL.The peak 25(OH)D3 values were observed 3 days afteroral vitamin D3, and represented incremental increases inserum 25(OH)D3 of 9.8 � 2.7 ng/mL, 19.9 � 9.6 ng/mL,and 30.6 � 16.0 ng/mL in L99P homozygous, L99Pheterozygous, and control subjects, respectively (P � .03).The incremental increase in serum 25(OH)D3 in theK242N heterozygous subject was 10.5 ng/mL.

Serum levels of 25(OH)D2 were undetectable prior toadministration of vitamin D2, and peaked at day 3 withincremental responses that were lower in individuals whowere homozygous (9.0 � 0.5 ng/mL) than those who wereheterozygous (19.0 � 7.8 ng/mL) for the L99P mutation(P � .025) (Figure 4C). Moreover, heterozygous individ-

Figure 2. Skeletal Deformities and Mutation Analysis in Affected Members ofFamily 1. (A and B) The proband of family 1 shows marked residual anterior tibial bowing atage 20 years. (C) His brother shows residual genu valgum at age 18 years. (D) A portion of theDNA sequence chromatogram of exon 2 from the proband. The nucleotide sequence of CYP2R1is noted above each chromatogram. This reveals a homozygous thymidine (T) to cystosine (C)substitution at c.296 (arrow) that leads to replacement of amino acid leucine at codon 99 withproline (p.99L�P).



uals had lower peak values than control children (38.2 �13.3 ng/mL, P � .008) (Figure 4C). In the K242Nheterozygous subject, the peak 25(OH)D2 concentrationwas 12.4 ng/mL.

Peak serum levels of 25(OH)D on day 3 positively cor-related with peak serum concentrations of vitamin D onday 1 in control and heterozygous subjects (Figure 5) butnot in subjects who were homozygous for the L99P mu-tation. The relatively lower vitamin D levels achieved inhomozygous and heterozygous subjects generally resultedin greater 25(OH)D concentrations at similar vitamin Dlevels in healthy controls. Among the L99P homozygoussubjects, the relatively low concentrations of 1,25(OH)2Dat baseline increased in response to administration of vi-tamin D2 or D3, consistent with a functional vitamin Ddeficiency).

Expression and functional analyses. The abundance ofboth mutant and wild-type CYP2R1 recombinant pro-teins in transiently transfected HEK293T cells was similar(Figure 4D), indicating that neither mutation affected ex-

pression of the protein. In order todetermine the consequence of theidentified mutations on CYP2R1 en-zyme function, we transiently trans-fected HEK293T host cells withempty vector, wild type or mutantCYP2R1 cDNAs and compared thegeneration of 1,25(OH)D3 fromsubstrate 1�-hydroxyvitamin D3 us-ing a luciferase reporter gene that isunder the control of the VDR. In theabsence of 1�-hydroxyvitamin D3 orrecombinant CYP2R1, very little lu-ciferase activity was detected (Figure4E). Cells expressing wild-typeCYP2R1 showed robust increases inluciferase activity compared to cellstransfected with vector DNA, indi-cating that synthesis of1,25(OH)2D3 was dependent uponforced expression of 25-hydroxylaseactivity. Moreover, the catalyticproperties of the recombinantCYP2R1 enzyme with 1�-hy-droxyvitamin D as a substrate (Km9.6 �M) compared favorably to theproperties of the purified enzyme(Km 11.3 �M) (33). By contrast, thep.L99P CYP2R1 did not induce ac-tivity of the reporter gene, indicatingthat this mutation abolished 25-hy-droxylase activity. The K242N mu-

tant induced a blunted response to 1�-hydroxyvitamin D3

(Km 10.4 � 7.7 �M, Vmax 0.55 � 0.12 relative units)compared to the wild type enzyme (Km 9.6 � 2.8 �M,Vmax 0.93 � 0.08 relative units), indicating a significantreduction in 25-hydroxylase activity. Because recent stud-ies show that the closely related CYP2C8 exists as a dimerwhen bound to mammalian membranes, and that thisstructure has functional significance (36), we considered itconceivable that the dimeric structure of CYP2R1 thatwas observed after crystallization of solubilized proteinmight also exist in the membrane-bound form of CYP2R1(33). Hence we assessed whether the L99P and K242Nproteins might inhibit activity of the wild type enzyme.However, when coexpressed with equal amount of wild-type CYP2R1, neither mutant enzyme showed a dominantnegative effect (data not shown).

Discussion

Rickets generally results from inadequate cutaneous syn-thesis or insufficient dietary supply of vitamin D. Less

Figure 3. Skeletal Deformities and Mutation Analysis in Affected Members ofFamily 2. (A) The proband and her affected sister from family 2, who are heterozygous for thep.L99P mutation, both had marked genu varum deformity. (B) A portion of the DNA sequencechromatogram of exon 2, which demonstrates heterozygous replacement of T by C at positionc.296 (arrow) that leads to replacement of amino acid leucine at codon 99 with proline(p.99L�P) on one allele. (C) A portion of the DNA sequence chromatogram of exon 3 from theirmother, with the arrow denoting heterozygous replacement of adenine (A) by cytosine (C) atc.726 (A�C) that leads to replacement of amino acid lysine at codon 242 by asparagine(p.242K�N).



commonly, rickets arises from pseudovitamin D defi-ciency, in which severe hypocalcemia and rickets developwithin the first year of life and do not respond to treatmentwith standard doses of vitamin D and calcium. The initialcharacterization of this hereditary form of pseudovitaminD deficiency rickets noted that to maintain health, intakeof vitamin D had to be consistently in vast excess of the

recommended daily allowance (RDA), hence the term ‘vi-tamin D dependency’ was proposed to describe the newsyndrome (37). Later reports characterized a second formof pseudovitamin D deficiency that was not responsive toeven high doses of vitamin D (38). The mechanisms andresponsible genes for these two types of pseudovitamin Ddeficiency have been characterized. Mutations in the

Table 1. Biochemical and Radiographic Characteristics of Five Subjects with CYP2R1 Mutations.

SubjectSubject II-1(Family 1)

Subject II-2(Family 1)

Subject I-1(Family 1)

Subject II-1(Family 2)

Subject I-2(Family 2)

Mutation Homozygous Homozygous Homozygous Heterozygous Heterozygous

L99P/L99P L99P/L99P L99P/L99P L99P/wt K242N/wt

Age 12.5 yr 13 yr 13.5 yr 20 yr 11 yr 11.5 yr 18 yr 49 yr 5 yr 5.5 yr 24 yr

Characteristic Before After Calcium* After Vitamin D3** Before After Calcium* Before After Calcium Reference Ranges

Calcium (mg/dL) 5.9 5.7 6.7 6.1 6.3 7.8 9.6 9.6 9.9 9.2 8.5–10.6

Phosphorus (mg/dL) 2.6 3.4 4.0 3.9 4.3 5.5 3.2 2.0 4.6 3 2.5–5.4 ages 5–14 y

2.5–4.5 adults

Albumin (g/dL) 3.7 3.9 3.7 3.4 4.4 3.9 4.3 3.5–5.0

Alkaline phosphatase (U/liter) 4866 5029 551 2391 2131 413 109 714 275 172 162–587 ages 5–14 y

45–115 adults

25(OH)D (ng/mL) 8.0 2.2 1.5 4.1 3.7 3.1 4.9 16.4 25.9 18.7 �20

1,25(OH)2D (pg/ml) 18 22 17 26 180 241 22–67

PTH (pg/ml) 123 382 339 208 199 182 60 107 42 85 11–67

Radiographic score 10 8 0.5 9 1 5 1 0

* Calcium was given as ground fish including bones 10 g twice daily, providing approximately 952 mg of elemental calcium daily for 6 months.

** Vitamin D3 600 000 IU (15 mg) was given intramuscularly twice at 3-monthly intervals.

*** To convert values for calcium to millimoles per liter, multiply by 0.25. To convert values for phosphorus to millimoles per liter, multiply by0.32. To convert values for 25(OH)D to nanomoles per liter, multiply by 2.50. To convert values for 1,25(OH)2D to picomoles per liter, multiply by2.40. To convert the values for PTH to picomoles per liter, multiply by 0.11.

Figure 4. Effect of CYP2R1 Mutations on 25-hydroxylase Activity. (A) Peak serum vitamin D2 and D3 levels 24 hours afteradministration of vitamin D2 or D3. (B) Serum 25(OH)D3 concentrations after administration of 50 000 IU of vitamin D3. (C) Serum 25(OH)D2

concentrations after administration of 50 000 IU of vitamin D2. (D) Immunoblot of HEK293T cell extracts. Cells were transfected with plasmidscontaining vector sequences only (lane 1) or cDNAs corresponding to wild type (lane 2), L99P (lane 3) or K242N (lane 4) Cyp2r1 with FLAG epitopetag. Blots were sequentially incubated with anti-FLAG antibody an anti-J-actin antibody to assess protein loading of each lane. (E) Response ofnormal and mutant Cyp2r1 enzymes to increasing concentrations of 1�-hydroxyvitamin D3. The expression plasmids were introduced into HEK293 cells with DNAs constituting the VDR-Gal4/GAL4-UAS-Luciferase hybrid reporter gene system. Relative activity is expressed as the ratio ofinduced firefly luciferase to control (constitutive) Renilla luciferase. Points on the graphs represent means of triplicate values established at eachconcentration of secosteroid. In all figures, error bars represent standard error of the mean, and stars represent significant (P � .05) differencesbetween subjects homozygous for the L99P mutation and control subjects, using the Kruskal-Wallis test with Dunn’s post-test analysis. Onesubject in Family 1 (II-4) did not participate in the in vivo vitamin D response assessment.



CYP27b gene encoding 25(OH)D 1�-hydroxylase are thebasis for vitamin D-dependent rickets type 1A (VDDR1A,MIM 264 700) (39, 40), and explain the inability to syn-thesize the fully active form of vitamin D, 1,25(OH)2D. Bycontrast, mutations in the VDR gene encoding the vitaminD receptor are the basis for vitamin D-dependent ricketstype 2A (VDDR2A, MIM 277 440) (41, 42), and accountfor resistance to high doses of vitamin D or even1,25(OH)2D. More recently, overexpression of a hetero-geneous nuclear ribonucleoprotein (hnRNP)-like domi-nant-negative protein that binds to the VDR response el-ement in vitamin D target genes has been described as asecond mechanism for VDDR2 (VDDR2B, MIM600 785), although the precise molecular defect is un-known (43). In this report we describe another mechanismfor pseudovitamin D deficiency that is uniquely associatedwith low serum concentrations of 25(OH)D, vitamin Ddependent rickets type 1B (VDDR1B, MIM 600 081).

We identified two different CYP2R1 missense muta-tions in affected members of two families with VDDR1B.Linkage analysis excluded an association between theCYP2R1 locus and rickets in ten additional families.Moreover, we did not identify CYP2R1 mutations in 27unrelated patients with sporadic rickets. Although ourmutation-detection strategy may have missed some mu-tations in CYP2R1, our results are consistent with geneticheterogeneity in VDDR1B and/or undisclosed environ-mental effects.

Our data support a gene dosage effect of CYP2R1 onconversion of vitamin D to 25(OH)D in humans. Semid-ominant inheritance is signified by a more severe diseasephenotype when the mutant allele is homozygous and aless severe phenotype when the allele is heterozygous.While VDDR1A is an autosomal recessive condition,VDDR1B is a more complex disorder in which the phe-notype is dependent upon the number of defective alleles.Patients with one defective CYP2R1 allele showed a mildform of VDDR1B. These patients produce less 25(OH)D

than control subjects after administration of either vita-min D2 or vitamin D3, but are able to maintain near nor-mal mineral homeostasis as adults. By contrast, patientswho are homozygous for CYP2R1 mutations have a moresevere form of VDDR1B. These patients show minimalincreases in serum 25(OH)D after vitamin D, and clinicalimprovement is effected only with very high doses of vi-tamin D or calcium (21, 44). Similarly, two Saudi siblingswith low serum concentrations of 25(OH)D due to com-pound heterozygosity for two other CYP2R1 mutationsshowed very modest responses to high dose vitamin Dsupplementation (20). Interestingly, subject I-1 in family1, an adult, showed near normal serum levels of PTH,calcium and phosphorus despite homozygosity for thenonfunctional L99P allele. This is similar to the naturalhistory of many patients with VDDR2A, who during andafter puberty and into adulthood are able to maintain nor-mal mineral metabolism with modest oral calcium sup-plements (45), in accordance with a proposed mechanismfor vitamin D-independent calcium absorption from theintestine (46). Nutritional rickets in Nigerian childrencommonly results from the interaction of very low dietarycalcium intakes and suboptimal vitamin D status (35). Thesusceptibility of children to develop rickets with theCYP2R1 mutation may be augmented by dietary calciumdeprivation or by vitamin D deficiency. The clinical re-sponse to calcium supplementation in some of the subjectswith CYP2R1 mutations indicates the critical role of pro-viding adequate calcium.

We demonstrated in one L99P homozygous subjectthat radiographic healing of rickets could be achieved withtwo large doses of IM vitamin D (600,000 IU), indicatingthat the vitamin D deficiency could be overcome. Thesmall but significant increases in serum 25(OH)D in pa-tients who have biallelic loss of function mutations inCYP2R1 may reflect the activity of other cytochromeP450 enzymes that can also serve as 25-hydroxylases (26,47, 48). These data are similar to observations in mice inwhich Cyp2r1 alleles have been genetically ablated, andwhich show both a gene-dosage effect on circulating levelsof 25(OH)D and markedly reduced but detectable levels of25(OH)D in homozygous Cyp2r1 knockout mice (23,49). The identity of the auxiliary 25-hydroxylase(s) is un-known. The increase in 1,25-dihydroxyvitamin D afteroral administration of 50 000 IU of vitamin D2 or D3 in theL99P homozygous subjects was also consistent with func-tional vitamin D deficiency, characterized by induction ofrenal CYP27B1 by elevated PTH (25). Whether lowerdoses of vitamin D would raise the substrate concentrationsufficiently to overcome the 25-hydroxylation deficit isunknown.

This report has several limitations. The two kindreds

Figure 5. Relationship of Peak 25-hydroxyvitamin D withPeak Vitamin D Concentrations. (A) Relationship of peak serum25(OH)D2 value at day 3 with peak vitamin D2 concentrations on day 1after vitamin D2 administration. (B) Relationship of peak serum25(OH)D3 value at day 3 with peak vitamin D3 concentrations on day 1after vitamin D3 administration.



whose responses to vitamin D2 and vitamin D3 were stud-ied included adult parents and older children, whereascontrol subjects were all children enrolled in a study ofvitamin D. It is possible that vitamin D absorption isgreater in children than in adults, possibly accounting forsome of the observed differences. The blunted rise in25(OH)D observed in subjects possessing the CYP2R1mutation may, in part, reflect impaired absorption of vi-tamin D rather than impaired CYP2R1 25-hydroxylation.However, even in subjects with the CYP2R1 mutation, thepeak vitamin D concentrations achieved should have beensufficient to produce a greater increase in 25(OH)D valuesthan we observed. Although consistent with a model ofsemidominant inheritance, we concede that our findingsare not definitive, and that the dosage effects that we haveobserved may be due to differences in age or genetic back-ground of the subjects we studied.

We did not treat any subjects with the CYP2R1 muta-tion with 25(OH)D, in order to bypass the enzymatic de-fect of 25-hydroxylation. This would merit further studyand potentially strengthen our conclusion regarding thefunctional significance of CYP2R1 mutations in vivo andthe benefit of treatment with 25(OH)D.

In the context of recent reports of inactivating muta-tions of CYP24A1 (50–52), which encodes an enzyme thatdegrades 25(OH)D and 1�,25(OH)2D, this study pro-vides new insights into the contribution of genetic vari-ability to human vitamin D homeostasis. The L99P alleleprevalence is nearly 1% in subjects of African descent (53).Given the abundance of other nonsynonymous polymor-phisms distributed throughout the coding region of thehuman CYP2R1 gene (28), and recent genome wide as-sociation studies showing an association between circu-lating 25(OH)D concentrations and the CYP2R1 locus(54), we suggest that variation in CYP2R1 expression mayhave an important role in determining vitamin D require-ments. Hence, our study raises important questions re-garding the effectiveness of standardized doses of supple-mental vitamin D for maintaining optimal vitamin Dstatus and encourages additional study of the relationshipbetween CYP2R1 genotype and vitamin D status.

Acknowledgments

We gratefully acknowledge the expert technical assistance of Dr.Anna Dang and Dr. Changlin Ding in performing the molecularand biochemical studies. We are grateful to Brian Netzel fordetermination of serum concentrations of vitamin D2 and vita-min D3 and their metabolites.

Address all correspondence and requests for reprints to: TomD. Thacher, M.D., Department of Family Medicine, Mayo

Clinic, 200 First Street SW, Rochester, MN, 55 902. Phone:507–284-5307, Fax: 507–284-5067, E-mail:[email protected].

This work was supported by the Lerner Research Institute ofthe Cleveland Clinic Foundation (to MAL) and the Friedman andSadikoglu families.

Disclosures: PRF, RJS, JR, and MAL have nothing to declare.TDT is a consultant for Biomedical Systems.

References

1. Thacher TD, Fischer PR, Strand MA, Pettifor JM. Nutritional rick-ets around the world: causes and future directions. Ann Trop Pae-diatr. 2006;26:1–16.

2. Holick MF, Chen TC. Vitamin D deficiency: a worldwide problemwith health consequences. Am J Clin Nutr. 2008;87:1080S–1086S.

3. Bendik I, Friedel A, Roos FF, Weber P, Eggersdorfer M. Vitamin D:a critical and essential micronutrient for human health. FrontPhysiol. 2014;5:248.

4. Cranney A, Horsley T, O’Donnell S, Weiler HA, Puil L, Ooi DS,Atkinson SA, Ward LM, Moher D, Hanley DA, Fang M, Yazdi F,Garritty C, Sampson M, Barrowman N, Tsertsvadze A, MamaladzeV. Effectiveness and safety of vitamin D in relation to bone health.Evidence Report/Technology Assessment No 158 (Prepared by theUniversity of Ottawa Evidence-based Practice Center under Con-tract No 290–02–0021) AHRQ Publication No 07-E013. Rock-ville, MD: Agency for Healthcare Research and Quality 2007.

5. Bischoff-Ferrari HA, Willett WC, Wong JB, Stuck AE, Staehelin HB,Orav EJ, Thoma A, Kiel DP, Henschkowski J. Prevention of non-vertebral fractures with oral vitamin D and dose dependency: ameta-analysis of randomized controlled trials. Arch Intern Med.2009;169:551–561.

6. Reis JP, von Muhlen D, Miller ER, 3rd, Michos ED, Appel LJ. Vi-tamin D status and cardiometabolic risk factors in the United Statesadolescent population. Pediatrics. 2009;124:e371–379.

7. Zipitis CS, Akobeng AK. Vitamin D supplementation in early child-hood and risk of type 1 diabetes: a systematic review and meta-analysis. Arch Dis Child. 2008;93:512–517.

8. Bolland MJ, Grey A, Gamble GD, Reid IR. Calcium and vitamin Dsupplements and health outcomes: a reanalysis of the Women’sHealth Initiative (WHI) limited-access data set. Am J Clin Nutr.2011;94:1144–1149.

9. Urashima M, Segawa T, Okazaki M, Kurihara M, Wada Y, Ida H.Randomized trial of vitamin D supplementation to prevent seasonalinfluenza A in schoolchildren. Am J Clin Nutr. 2010;91:1255–1260.

10. Camargo CA, Jr., Ingham T, Wickens K, Thadhani R, Silvers KM,Epton MJ, Town GI, Pattemore PK, Espinola JA, Crane J, NewZealand A, Allergy Cohort Study G. Cord-blood 25-hydroxyvita-min D levels and risk of respiratory infection, wheezing, and asthma.Pediatrics. 2011;127:e180–187.

11. Bischoff-Ferrari HA. Optimal serum 25-hydroxyvitamin D levelsfor multiple health outcomes. Adv Exp Med Biol. 2014;810:500–525.

12. Rosen CJ, Abrams SA, Aloia JF, Brannon PM, Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G, Kovacs CS, MansonJE, Mayne ST, Ross AC, Shapses SA, Taylor CL. IOM committeemembers respond to Endocrine Society vitamin D guideline. J ClinEndocrinol Metab. 2012;97:1146–1152.

13. Henry HL. Regulation of vitamin D metabolism. Best Pract Res ClinEndocrinol Metab. 2011;25:531–541.

14. van Schoor NM, Lips P. Worldwide vitamin D status. Best Pract ResClin Endocrinol Metab. 2011;25:671–680.

15. Arabi A, El Rassi R, El-Hajj Fuleihan G. Hypovitaminosis D indeveloping countries-prevalence, risk factors and outcomes. NatRev Endocrinol. 2010;6:550–561.



mailto:[email protected]

16. Wahl DA, Cooper C, Ebeling PR, Eggersdorfer M, Hilger J, Hoff-mann K, Josse R, Kanis JA, Mithal A, Pierroz DD, Stenmark J,Stocklin E, Dawson-Hughes B. A global representation of vitamin Dstatus in healthy populations: reply to comment by Saadi. Arch Os-teoporos. 2013;8:122.

17. Hilger J, Friedel A, Herr R, Rausch T, Roos F, Wahl DA, Pierroz DD,Weber P, Hoffmann K. A systematic review of vitamin D status inpopulations worldwide. Br J Nutr. 2014;111:23–45.

18. Thacher TD, Fischer PR, Pettifor JM, Lawson JO, Isichei CO, Read-ing JC, Chan GM. A comparison of calcium, vitamin D, or both fornutritional rickets in Nigerian children. N Engl J Med. 1999;341:563–568.

19. Thacher TD, Fischer PR, Pettifor JM, Lawson JO, Isichei CO, ChanGM. Case-control study of factors associated with nutritional rick-ets in Nigerian children. J Pediatr. 2000;137:367–373.

20. Al Mutair AN, Nasrat GH, Russell DW. Mutation of the CYP2R1vitamin D 25-hydroxylase in a Saudi Arabian family with severevitamin D deficiency. J Clin Endocrinol Metab. 2012;97:E2022–2025.

21. Cheng JB, Levine MA, Bell NH, Mangelsdorf DJ, Russell DW. Ge-netic evidence that the human CYP2R1 enzyme is a key vitamin D25-hydroxylase. Proc Natl Acad Sci USA. 2004;101:7711–7715.

22. Dong Q, Miller WL. Vitamin D 25-hydroxylase deficiency. MolGenet Metab. 2004;83:197–198.

23. Zhu JG, Ochalek JT, Kaufmann M, Jones G, Deluca HF. CYP2R1is a major, but not exclusive, contributor to 25-hydroxyvitamin Dproduction in vivo. Proc Natl Acad Sci USA. 2013;110:15650–15655.

24. Thacher TD, Fischer PR, Pettifor JM, Lawson JO, Manaster BJ,Reading JC. Radiographic scoring method for the assessment of theseverity of nutritional rickets. J Trop Pediatr. 2000;46:132–139.

25. Thacher TD, Fischer PR, Obadofin MO, Levine MA, Singh RJ,Pettifor JM. Comparison of metabolism of vitamins D2 and D3 inchildren with nutritional rickets. J Bone Miner Res. 2010;25:1988–1995.

26. Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW. De-orphani-zation of cytochrome P450 2R1: a microsomal vitamin D 25-hy-droxilase. J Biol Chem. 2003;278:38084–38093.

27. Gonzalez-Perez A, Lopez-Bigas N. Improving the assessment of theoutcome of nonsynonymous SNVs with a consensus deleteriousnessscore, Condel. Am J Hum Genet. 2011;88:440–449.

28. Kumar P, Henikoff S, Ng PC. Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm.Nat Protoc. 2009;4:1073–1081.

29. Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A,Bork P, Kondrashov AS, Sunyaev SR. A method and server for pre-dicting damaging missense mutations. Nat Methods. 2010;7:248–249.

30. Reva B, Antipin Y, Sander C. Predicting the functional impact ofprotein mutations: application to cancer genomics. Nucleic AcidsRes. 2011;39:e118.

31. Venselaar H, Te Beek TA, Kuipers RK, Hekkelman ML, Vriend G.Protein structure analysis of mutations causing inheritable diseases.An e-Science approach with life scientist friendly interfaces. BMCBioinformatics. 2010;11:548.

32. SNP linked to gene CYP2R1 (geneID:120227) via contig annota-tion. NCBI (accessed June 19, 2012).

33. Strushkevich N, Usanov SA, Plotnikov AN, Jones G, Park HW.Structural analysis of CYP2R1 in complex with vitamin D3. J MolBiol. 2008;380:95–106.

34. Thacher TD, Bommersbach TJ, Pettifor JM, Isichei CO, Fischer PR.Comparison of limestone and ground fish for treatment of nutri-tional rickets in Nigerian children. J Pediatr. 2015:. In press.

35. Thacher TD, Fischer PR, Pettifor JM. Vitamin D treatment in cal-cium-deficiency rickets: a randomised controlled trial. Arch DisChild. 2014;99:807–811.

36. Hu G, Johnson EF, Kemper B. CYP2C8 exists as a dimer in naturalmembranes. Drug Metab Dispos. 2010;38:1976–1983.

37. Scriver CR. Vitamin D dependency. Pediatrics. 1970;45:361–363.38. Marx SJ, Spiegel AM, Brown EM, Gardner DG, Downs RW, Jr.,

Attie M, Hamstra AJ, DeLuca HF. A familial syndrome of decreasein sensitivity to 1,25-dihydroxyvitamin D. J Clin Endocrinol Metab.1978;47:1303–1310.

39. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL,Portale AA. Cloning of human 25-hydroxyvitamin D-1 alpha-hy-droxylase and mutations causing vitamin D-dependent rickets type1. Mol Endocrinol. 1997;11:1961–1970.

40. Wang X, Zhang MY, Miller WL, Portale AA. Novel gene mutationsin patients with 1alpha-hydroxylase deficiency that confer partialenzyme activity in vitro. J Clin Endocrinol Metab. 2002;87:2424–2430.

41. Malloy PJ, Pike JW, Feldman D. The vitamin D receptor and thesyndrome of hereditary 1,25-dihydroxyvitamin D-resistant rickets.Endocr Rev. 1999;20:156–188.

42. Malloy PJ, Wang J, Peng L, Nayak S, Sisk JM, Thompson CC,Feldman D. A unique insertion/duplication in the VDR gene thattruncates the VDR causing hereditary 1,25-dihydroxyvitamin D-re-sistant rickets without alopecia. Arch Biochem Biophys. 2007;460:285–292.

43. Chen H, Hewison M, Hu B, Adams JS. Heterogeneous nuclear ri-bonucleoprotein (hnRNP) binding to hormone response elements: acause of vitamin D resistance. Proc Natl Acad Sci USA. 2003;100:6109–6114.

44. Casella SJ, Reiner BJ, Chen TC, Holick MF, Harrison HE. A pos-sible genetic defect in 25-hydroxylation as a cause of rickets. J Pe-diatr. 1994;124:929–932.

45. Tiosano D, Hadad S, Chen Z, Nemirovsky A, Gepstein V, MilitianuD, Weisman Y, Abrams SA. Calcium absorption, kinetics, bonedensity, and bone structure in patients with hereditary vitamin D-resistant rickets. J Clin Endocrinol Metab. 2011;96:3701–3709.

46. Van Cromphaut SJ, Rummens K, Stockmans I, Van Herck E, DijcksFA, Ederveen AG, Carmeliet P, Verhaeghe J, Bouillon R, CarmelietG. Intestinal calcium transporter genes are upregulated by estrogensand the reproductive cycle through vitamin D receptor-independentmechanisms. J Bone Miner Res. 2003;18:1725–1736.

47. Gupta RP, Hollis BW, Patel SB, Patrick KS, Bell NH. CYP3A4 is ahuman microsomal vitamin D 25-hydroxylase. J Bone Miner Res.2004;19:680–688.

48. Yamasaki T, Izumi S, Ide H, Ohyama Y. Identification of a novel ratmicrosomal vitamin D3 25-hydroxylase. J Biol Chem. 2004;279:22848–22856.

49. Wang Y, Marling SJ, McKnight SM, Danielson AL, Severson KS,Deluca HF. Suppression of experimental autoimmune encephalo-myelitis by 300–315nm ultraviolet light. Arch Biochem Biophys.2013;536:81–86.

50. Schlingmann KP, Kaufmann M, Weber S, Irwin A, Goos C, John U,Misselwitz J, Klaus G, Kuwertz-Broking E, Fehrenbach H, WingenAM, Guran T, Hoenderop JG, Bindels RJ, Prosser DE, Jones G,Konrad M. Mutations in CYP24A1 and idiopathic infantile hyper-calcemia. N Engl J Med. 2011;365:410–421.

51. Dauber A, Nguyen TT, Sochett E, Cole DE, Horst R, Abrams SA,Carpenter TO, Hirschhorn JN. Genetic defect in CYP24A1, thevitamin D 24-hydroxylase gene, in a patient with severe infantilehypercalcemia. J Clin Endocrinol Metab. 2012;97:E268–274.

52. Tebben PJ, Milliner DS, Horst RL, Harris PC, Singh RJ, Wu Y,Foreman JW, Chelminski PR, Kumar R. Hypercalcemia, hypercal-ciuria, and elevated calcitriol concentrations with autosomal dom-inant transmission due to CYP24A1 mutations: effects of ketocona-zole therapy. J Clin Endocrinol Metab. 2012;97:E423–427.

53. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM,Sirotkin K. dbSNP: the NCBI database of genetic variation. NucleicAcids Res. 2001;29:308–311.

54. Wang TJ, Zhang F, Richards JB, Kestenbaum B, van Meurs JB,Berry D, Kiel DP, Streeten EA, Ohlsson C, Koller DL, Peltonen L,Cooper JD, O’Reilly PF, Houston DK, Glazer NL, Vandenput L,Peacock M, Shi J, Rivadeneira F, McCarthy MI, Anneli P, de Boer



IH, Mangino M, Kato B, Smyth DJ, Booth SL, Jacques PF, BurkeGL, Goodarzi M, Cheung CL, Wolf M, Rice K, Goltzman D, Hid-iroglou N, Ladouceur M, Wareham NJ, Hocking LJ, Hart D, ArdenNK, Cooper C, Malik S, Fraser WD, Hartikainen AL, Zhai G, Mac-donald HM, Forouhi NG, Loos RJ, Reid DM, Hakim A, DennisonE, Liu Y, Power C, Stevens HE, Jaana L, Vasan RS, Soranzo N,

Bojunga J, Psaty BM, Lorentzon M, Foroud T, Harris TB, HofmanA, Jansson JO, Cauley JA, Uitterlinden AG, Gibson Q, Jarvelin MR,Karasik D, Siscovick DS, Econs MJ, Kritchevsky SB, Florez JC,Todd JA, Dupuis J, Hypponen E, Spector TD. Common geneticdeterminants of vitamin D insufficiency: a genome-wide associationstudy. Lancet. 2010;376:180–188.



CYP2R1 Mutations Impair Generation of 25-hydroxyvitamin D and Cause an Atypical Form of Vitamin D...

Documents

Transcript of CYP2R1 Mutations Impair Generation of 25-hydroxyvitamin D and Cause an Atypical Form of Vitamin D...