Targeted Inactivation of the 25-Hydroxyvitamin D 3 -1 α -Hydroxylase Gene (CYP27B1) Creates an...

Targeted Inactivation of the 25-Hydroxyvitamin D3-1a-Hydroxylase Gene (CYP27B1) Creates an Animal Modelof Pseudovitamin D-Deficiency Rickets*

OLIVIER DARDENNE, JOSEE PRUD’HOMME, ALICE ARABIAN,FRANCIS H. GLORIEUX, AND RENE ST-ARNAUD†

Genetics Unit (O.D., J.P., A.A., F.H.G., R.St.-A.), Shriners Hospital for Children, Montreal (Quebec)Canada H3G 1A6; and Departments of Surgery and Human Genetics (F.H.G., R.St.-A.), McGillUniversity, Montreal (Quebec) Canada H3A 1B1

ABSTRACTPseudovitamin D-deficiency rickets is caused by mutations in the

cytochrome P450 enzyme, 25-hydroxyvitamin D3-1a-hydroxylase (1a-OHase). Patients with the disease exhibit growth retardation, rickets,and osteomalacia. Serum biochemistry is characterized by hypocal-cemia, secondary hyperparathyroidism, and undetectable levels of1a,25-dihydroxyvitamin D3. We have inactivated the 1a-OHase genein mice after homologous recombination in embryonic stem cells.Serum analysis of homozygous mutant animals confirmed that they

were hypocalcemic, hypophosphatemic, hyperparathyroidic, and thatthey had undetectable 1a,25-dihydroxyvitamin D3. Histological anal-ysis of the bones from 3-week-old mutant animals confirmed theevidence of rickets. At the age of 8 weeks, femurs from 1a-OHase-ablated mice present a severe disorganization in the architecture ofthe growth plate and marked osteomalacia. These results show thatwe have successfully inactivated the 1a-OHase gene in mice andestablished a valid animal model of pseudovitamin D-deficiency rick-ets. (Endocrinology 142: 3135–3141, 2001)

VITAMIN D3 IS activated to its hormonal form, 1a,25-dihydroxyvitamin D3 [1,25(OH)2D3], by two succes-

sive hydroxylation reactions. The first of these takes place inthe liver, giving rise to 25-hydroxyvitamin D3 [25(OH)D3].This inactive metabolite is converted to 1,25(OH)2D3 by therenal cytochrome P450 enzyme, 25-hydroxyvitamin D3-1a-hydroxylase (CYP27B1, hereafter referred to as 1a-OHase).The kidney is the major site of 1a-OHase activity, but ex-pression of the gene has been detected in other cell types suchas keratinocytes, osteoblasts, chondrocytes, and macro-phages (1, 2).

Once synthesized, 1,25(OH)2D3 plays a key role in theregulation of calcium homeostasis by promoting calciumabsorption from the intestinal lumen. The homeostatic feed-back loop involves both 1,25(OH)2D3 and PTH. Decreases inblood calcium stimulate synthesis and secretion of PTH,which, in turn, leads to increased expression of the 1a-OHasegene, 1,25(OH)2D3 synthesis, and intestinal calcium absorp-tion. To prevent sustained production of 1,25(OH)2D3 thatwould lead to hypercalcemia, the vitamin D hormone, inturn, inhibits PTH and 1a-OHase gene expression (3).

Pseudovitamin D-deficiency rickets (PDDR) is a rare au-tosomal recessive disease characterized by growth retarda-tion, failure to thrive, rickets, and osteomalacia (4, 5). Serumbiochemistry reveals hypocalcemia, secondary hyperpara-

thyroidism, and undetectable levels of 1,25(OH)2D3. Mea-surements of circulating vitamin D metabolite levels and1a-OHase enzymatic activity have long suggested that thedisease is caused by inactivating mutations in the cyto-chrome P450 gene responsible for the synthesis of1,25(OH)2D3 (4, 5). The cloning of the 1a-OHase comple-mentary DNA (cDNA) and gene (6–10) has confirmed thishypothesis, first by mapping of the gene to the disease locus(7, 8, 11–14), then by the identification of mutations in af-fected patients (4, 6, 11, 12, 15–17).

We have generated an animal model of PDDR by targetedinactivation of the 1a-OHase gene in mice. Homozygousmutant animals were phenotypically normal at birth butprogressively developed the full range of PDDR symptoms.Hypocalcemia, hyperparathyroidism, and rickets were evi-dent as soon as weaning, whereas osteomalacia was detectedin young adult mutant animals.

Materials and MethodsTargeting vector

A partial murine 1a-OHase cDNA was amplified by PCR, usingprimers based on the rat cDNA sequence (8), and used as probe to clonea 12-kb fragment from a 129-SV genomic library (Stratagene, La Jolla,CA). A 927-bp KpnI genomic fragment containing a portion of exon 7,intron 7, and exon 8 was subcloned in pGEM3Zf(-). This template wasused to introduce a first loxP site, 35-bp into intron 7, by inverse PCR(primer information available on demand). The loxP-neo-loxP selectioncassette was engineered by subcloning the 1.6-kb Mst II fragment (con-taining the SV40 promoter, neo resistance gene, and thymidine kinasepolyadenylation signal) of pBK-CMV (Stratagene) into the HindIII siteof pBS246 (Life Technologies, Inc., Burlington, Ontario, Canada). Thecassette was excised as a NotI fragment and subcloned into the Msc I sitedownstream from exon 8 in the pGEM3Zf(-) subclone. The engineeredKpnI subfragment containing exons 7 and 8, the neo cassette, and thethree loxP sites was then reintroduced into the 12-kb 1a-OHase genomic

Received February 15, 2001.Address all correspondence and requests for reprints to: Rene St-

Arnaud, Genetics Unit, Shriners Hospital for Children, 1529 Cedar Av-enue, Montreal (Quebec), Canada H3G 1A6. E-mail: [email protected].

* This work was supported by the Shriners of North America.† A Chercheur-Boursier from the Fonds de la Recherche en Sante du

Quebec.

0013-7227/01/$03.00/0 Vol. 142, No. 7Endocrinology Printed in U.S.A.Copyright © 2001 by The Endocrine Society

3135

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2014. at 05:31 For personal use only. No other uses without permission. . All rights reserved.

fragment at the corresponding site. A linear 8.8-kb ApaLI fragment fromthis targeting vector (Fig. 1a) was electroporated into R1 embryonic stem(ES) cells (18) using conventional protocols (19). Targeted ES cell cloneswere identified by Southern blotting of BamHI-digested DNA using a450-bp PCR-amplified probe from upstream of the region of homology(upstream primer: 59-ACACACACACACACACCAATATG-39; down-stream primer: 59-TGCACCACCACGCCCGGC-39) (Fig. 1a). We thenexcised exon 8 and the neo selection cassette by transfecting targeted ES

cells with the Cre expression vector, pBS185 (Life Technologies, Inc.). Weconfirmed the in vitro excision using Southern blotting of BamHIgenomic digests with the previously described probe. For routine geno-typing of tail DNA samples (19), we used an internal PCR-amplified374-bp probe (upstream primer: 59-ACCATTTTAGACCCACCCA-CAGT-39; downstream primer: 59-GAGGAATGATCAGGAGAG-GCAC-39), yielding a wild-type fragment of 5428 bp and a diagnosticdeleted allele fragment of 4797 bp.

FIG. 1. Targeted inactivation of the 1a-OHase gene in mice. a, Schematic representation of the targeting replacement-type vector, 1a-OHasewild-type locus, and targeted 1a-OHase locus obtained after Cre-mediated excision in ES cells. Recombination between the two most distantloxP sites deletes exon 8 encoding the heme binding domain, effectively creating a null allele. Lower panel, Genotyping of a litter from arepresentative heterozygous x heterozygous intercross. Tail DNA was digested with BamHI, and Southern blot analysis was performed witha PCR-generated probe. The 5450-bp wild-type allele and the 4600-bp deleted allele are identified. Genotypes (1/1, 1/2, 2/2) are shown atthe top of each lane. HBD, Heme-binding domain. b, RT-PCR analysis of total RNA from kidney of 8-week-old animals. Left panel, 1a-OHaseamplimers using primers derived from intron 7 and intron 8. Wild-type (1/1) and heterozygotes (1/2) show a fragment of 677 bp that is absentfrom homozygous mutants (2/2) because of the deletion of exon 8 and part of introns 7 and 8. b-actin control amplimers are shown below. Rightpanel, 1a-OHase amplimers were obtained using upstream and downstream primers derived from exon 6 and exon 9, respectively. Wild-typeand heterozygous animals show a 451-bp fragment, whereas homozygous mutants express an inactive truncated transcript yielding an amplimerof 253 bp. M, Size markers; genotypes indicated below each lane. c, Circulating levels of 1,25(OH)2D3 as determined by RIA. The absence ofcirculating 1,25(OH)2D3 in 2/2 animals confirms that we engineered a true null allele. ND, Not detectable. d, Serum levels of 24,25(OH)2D3were measured by RIA in adult (8-week-old) mice. e, RIA measurement of circulating levels of 25(OH)D3 in mice of all three genotypes killedat 8 weeks of age. For c–e: ***, P , 0.001.

3136 ANIMAL MODEL OF PDDR Endo • 2001Vol. 142 • No. 7


The injection of the ES cells carrying the deleted 1a-OHase allele intoC57BL/6 embryos at the blastocyst stage used standard methodology(19). The resulting chimeras and their progeny were housed in a virus-and parasite-free barrier facility. They were exposed to a 12-h light, 12-hdark cycle and were fed tap water and regular chow at libitum. Allprocedures involving animals were previously approved by the Insti-tutional Animal Care and Use Committee.

RT-PCR analysis of 1a-OHase transcripts

The RT-PCR protocol used total RNA prepared with the Trizol reagent(Life Technologies, Inc.) from kidney of 1/1, 1/-, and 2/2 animals. TheRT reaction contained oligo-dT and Superscript II reverse transcriptase (LifeTechnologies, Inc.). PCR amplification required two pairs of primers. Thefirst pair (59-CTGCGAGGAGGGGTAAGGTGTT-39 and 59-GGAAA-CGGGGGAGGGGA-39) allowed detection of exon 8-containing transcripts.The second pair (59-TCTATGAGCTTTCCCGGCACCCC-39 and 59-TCAG-GTAGCTCTTCAAAATGGGTCAA-39) permitted detection of truncated

transcripts. Control b-actin amplification reactions used the followingprimer pair: 59-GCTGCGTGTGGCCCCTAGG-39 and 59-CAAGAAG-GAAGGCTGGAAAAGAG-39. Amplimers were detected by ethidium bro-mide staining of agarose electrophoresis gels.

Serum biochemistry

Circulating 1,25(OH)2D3 levels were measured from serum samplesusing a specific RIA (ImmunoDiagnostic Systems Ltd., Boldon, UK).Serum concentrations of 25(OH)D3 and 24,25(OH)2 D were assessedusing the 25-hydroxyvitamin D (125) I RIA kit for the quantitativedetermination of vitamin D hydroxylated metabolites (DiaSorin, Inc.,Stillwater, MN). Total calcium and phosphate were measured using aMonarch automated analyzer. We measured serum PTH levels usingthe mouse intact PTH enzyme-linked immunosorbent assay kit (Im-munotropics, San Clemente, CA). All serological data are shown asmean 6 sem.

FIG. 2. 1a-OHase 2/2 animals are hypocalcemic and hypophosphatemic, and they exhibit secondary hyperparathyroidism. a, Total serumcalcium was measured in all three genotypes, from weaning (3 weeks) to the time they were killed at 8 weeks. b, Serum phosphate levels weremeasured in 3-week-old and 8-week-old wild-type (1/1), heterozygous (1/2), and homozygous mutant (2/2) animals. c, Immunoreactive PTHlevels were assessed in animals of all three genotypes at 3 and 8 weeks. For A–C: *, P , 0.05; **, P , 0.01; ***, P , 0.001. d, Histology of theparathyroid glands in 8-week-old heterozygous (1/2) and homozygous mutant (2/2) littermates. Hematoxylin and eosin counter-stainedsections are shown in bright field. The arrows point to the parathyroid gland in each section presented. Magnification, 1003; scale bar, 0.2 mm.

ANIMAL MODEL OF PDDR 3137


Northern blot assays

We used the Trizol Reagent (Life Technologies, Inc.) to isolate total RNAfrom the intestine, kidney, and tibia of 8-week-old 1/2 and 2/2 animals,and we performed Northern blots using standard methodology. The probesused were the 170-bp EcoRI fragment from the mouse calbindin D9k cDNA(20), the 1.1-kb BamHI-EcoRI fragment from the mouse osteopontin cDNA(21), the 470-bp EcoRI-PstI fragment from the mouse osteocalcin cDNA (22),the 263-bp HincII-KpnI fragment from the rat CYP24 cDNA (23), the PstIfragment of the rat glyceraldehyde-3-phosphate-dehydrogenase cDNA(24), and a 420-bp PCR fragment from the mouse vitamin D receptor (VDR)cDNA (upstream primer: 59-AGGGTTTCTTCAGGCGGAGCAT-39; down-stream primer: 59-CATGTCCAGTGAGGGGGTGTAC-39).

Histology

The thyroid, parathyroids, trachea, and heart were dissected en blocfrom 8-week-old littermates to facilitate orientation of the specimens andsectioning in the same plane. Samples were fixed overnight in 4% para-formaldehyde before paraffin embedding. Six-micrometer sections wereprepared and stained with hematoxylin-eosin. Bones were dissected,fixed overnight in 4% paraformaldehyde, and embedded in methyl-methacrylate. Sections of 6 mm were deplastified and stained by vonKossa or Goldner (25).

ResultsGenetic ablation of the 1a-OHase gene

We cloned the 1a-OHase gene from a 129-SV mousegenomic library to construct a targeting vector in which exon8, encoding the heme binding domain (26), was flanked bya 59-loxP recognition site and by a 39-loxP-neo-loxP selectioncassette (Fig. 1a). Homologous recombination at the 1a-OHase locus, followed by transient transfection of the EScells with the Cre recombinase, generated a targeted allele inwhich exon 8 was deleted (Fig. 1A). The engineered mutationwas transmitted to the progeny (Fig. 1A), with the expectedMendelian ratio (not shown). Heterozygous animals had nodiscernible phenotype and were fertile. Homozygous mutantanimals were phenotypically normal at birth but exhibitedretarded growth, as measured by weight gain from 3–8weeks of age and femur length at 8 weeks (data not shown).

We did not detect 1a-OHase transcripts that containedexon 8 in homozygous 2/2 animals (Fig. 1b, left panel);however, mutant animals expressed a truncated 1a-OHasemessenger RNA (mRNA) (Fig. 1b, right panel). The productof this aberrant transcript was inactive, as evidenced byundetectable 1,25(OH)2D3 levels in the blood of 2/2 animals(Fig. 1c), confirming that we have engineered a true nullallele. In adult animals, circulating concentrations of 24,25-dihydroxyvitamin D3 [24,25(OH)2D3] were negligible (1.4 60.2 vs. 11.1 6 0.4 ng/ml for mutants and heterozygotes,respectively, P , 0.001) (Fig. 1d), whereas the levels of25(OH)D3 were elevated (21.9 6 5.6 vs. 12.0 6 0.4 ng/ml formutants and heterozygotes, respectively, P , 0.001) (Fig. 1e).

Serum biochemistry

A slight, but statistically significant (P , 0.01), hypocal-cemia was detected in 2/2 animals as early as weaning (3weeks, Fig. 2a). The hypocalcemic status worsened in thefollowing week, remained constant for a while, before wors-ening again when the animals were killed at 8 weeks (1.30 60.04 vs. 2.33 6 0.02 mm for 2/2 and 1/2 animals, respec-tively, P , 0.001) (Fig. 2a). In homozygous mutant pups,

marked hypophosphatemia was detected at 3 weeks of age(Fig. 2b). The difference in serum phosphate levels was lesspronounced in adult 2/2 animals, compared with wild-type and heterozygous littermate controls, because of thedecrease in circulating phosphate levels in control animals(Fig. 2b). Nonetheless, the serum phosphate levels measuredin 2/2 animals at 8 weeks remained significantly lower thanin 1/2 littermates (2.5 6 0.3 vs. 3.2 6 0.2 mm, respectively,P , 0.05) (Fig. 2b). When compared with control littermates(1/1 or 1/2), serum alkaline phosphatase activity wassignificantly elevated in 2/2 animals at all times (data notshown).

Decreases in blood calcium levels increase PTH expressionand secretion, and the absence of a concomitant rise in cir-culating 1,25(OH)2D3 leads to secondary hyperparathyroid-ism in patients with PDDR (3, 5). Similarly, 1a-OHase-ablated pups showed increased PTH blood levels as early asweaning, and the secondary hyperparathyroidism was evenmore pronounced when the animals were killed at 8 weeks(1444 6 123 vs. 20 6 2 pg/ml for mutant and wild-type,respectively, P , 0.001) (Fig. 2c). The parathyroid glands alsoincreased in size in 8-week-old homozygous mutant mice(Fig. d; the arrows point to the parathyroids).

Vitamin D-dependent gene expression

We measured the expression levels of vitamin D-regulatedgenes (27) in several target tissues, using Northern blot as-says. Expression of the calbindin D9k gene in the duodenumwas completely inhibited in 1a-OHase 2/2 animals (Fig. 3,left panel), as was the expression of CYP24 (25-hydroxyvita-min D-24-hydroxylase) in kidney (data not shown). On thecontrary, the expression levels of the osteocalcin and os-teopontin genes in tibia were unaffected in mutant animals(Fig. 3, center and right panels). Expression of the vitamin Dreceptor gene in kidney from homozygous mutants was re-duced (not shown).

Bone histology

We collected femurs from 2/2 mutants and 1/2 controllittermates at 3 and 8 weeks of age. Von Kossa staining servedto assess mineralization, whereas Goldner-stained sections

FIG. 3. Expression of vitamin D-regulated genes. Left-most panel,Northern blot hybridization of total RNA extracted from kidney of8-week-old heterozygous (1/2) and homozygous mutant (2/2) ani-mals; C9K, calbindin D9k; center and right panels, tibia RNA from thesame animals was probed with an osteocalcin (OC) and an osteopontin(OPN) probe; GAPDH, glyceraldehyde phosphate dehydrogenase.



FIG. 4. Hypomineralization, rickets,and osteomalacia in 1a-OHase mutantmice. Sagittal sections through theepiphysis of femurs from 3-week-old(a–d) and 8-week-old (e–h) homozygousmutant (2/2) (b, d, f, and h) and het-erozygous control (1/2) (a, c, e, and g)littermates. Sagittal sections from thefemoral cortex of 8-week-old animalsare also shown (i and j). Clear evidenceof rickets is visible at the epiphysealgrowth plate with hypomineralization(b, d, f, and h) and thicker osteoid seams(d and h) in 1a-OHase 2/2 animals.Osteomalacia (increased osteoid) is ev-ident in the cortex from the femur ofadult 2/2 mice (j). a, b, e, and f, vonKossa staining that shows the mineralas a black stain; c, d, and g–j, Goldnerstain that highlights the mineral ingreen and the osteoid in red; gp, growthplate; scale bar: 0.4 mm (a–h), 0.2 mm(i and j).



allowed for comparative histology. Long bones from3-week-old 2/2 pups were hypomineralized (Fig. 4, a andb). The femurs presented evidence of rickets, as demon-strated by the impaired calcification of the maturing car-tilage, disorganization of the columnar alignment of hy-pertrophic chondrocytes, increased width of the growthplate, and accumulation of osteoid (unmineralized matrix)in trabecular bone (Fig. 4, c and d). At 8 weeks of age, thedefect in mineralization was more profound (Fig. 4, e andf), and the 2/2 animals showed a more severe disorga-nization in the architecture of the growth plate (Fig. 4,g and h). The chondrocytes completely lost their columnarorganization. Marked osteomalacia (accumulation of os-teoid) was evident when we compared the femoral cortexfrom 2/2 and 1/2 littermates (Fig. 4, i and j).

Discussion

We have successfully targeted the 1a-OHase gene. Theengineered mouse strain represents a valid animal model forPDDR. The homozygous mutant animals presented with thesame clinical phenotype as patients with the disease: re-tarded growth, failure to thrive, undetectable 1,25(OH)2D3,hypocalcemia, secondary hyperparathyroidism, and boneanomalies that included rickets and osteomalacia.

Other aspects of the phenotype of 1a-OHase 2/2 animalsalso matched the clinical manifestations of PDDR. Theseincluded x-ray features, hypophosphatemia, and elevatedserum alkaline phosphatase. Contact radiography of femursfrom mutant animals revealed diffuse osteopenia (hypomin-eralization) and rachitic metaphyseal changes (data notshown). Serum phosphate concentrations may be normal orlow in PDDR patients (4, 5). Similarly, we have measured amarked hypophosphatemia in young 2/2 animals, whichtended toward normal at later stages, an observation remi-niscent of the human pathology. Individuals affected byPDDR consistently have elevated serum alkaline phospha-tase activity (4, 5), and we observed the same feature inhomozygous mutant animals (not shown).

Patients with PDDR have normal serum levels of25(OH)D3 (28–30), and circulating levels of 24,25(OH)2D3also remain within the normal range (31). This observationcontrasts with our findings, which show elevated levels of25(OH)D3 and very low levels of 24,25(OH)2D3 in 1a-OHasemutant animals. The 1,25(OH)2D3 hormone is the main invivo regulator of the expression of the CYP24 enzyme thatcatalyzes the synthesis of 24,25(OH)2D3 (32). It was thus notsurprising to measure low circulating 24,25(OH)2D3 levelsand undetectable CYP24 expression in 1a-OHase mutantanimals. The discrepancy with the human disease remains tobe explained but could result from species differences. Thisobserved inhibition of CYP24 expression in mice, combinedwith the targeted ablation of the 1a-OHase enzyme, leads toa metabolic block in mutant animals and an accumulation ofthe unprocessed 25(OH)D3 substrate.

The osteocalcin gene promoter is modularly organizedand contains several positive and negative regulatory ele-ments (33). Similarly, osteopontin gene expression is regu-lated by many agents acting via diverse signaling pathwaysin specific cell types (34). Though 1,25(OH)2D3 has been

shown to regulate the expression of these genes both in vitroand in vivo (33, 34), the unaltered mRNA levels for both genesmeasured in 1a-OHase mutant animals suggest that the vi-tamin D hormone is not the main modulator of the steady-state expression of these genes in murine bone tissue.

Two independent laboratories engineered strains of ani-mals deficient for the vitamin D receptor (35, 36). The mutantmice are valid animal models for another hereditary type ofrickets: hereditary vitamin D-resistant rickets (4, 37). Someaspects of the phenotype of these mice are identical to thephenotype of the mutant 1a-OHase mice that we engineered:retarded growth, hypocalcemia, secondary hyperparathy-roidism, and rickets. The VDR-null mice also develop alo-pecia, a clinical feature shared with some hereditary vitaminD-resistant rickets patients but not exhibited by PDDR pa-tients or the 1a-OHase-ablated mice. It seems that the hy-pocalcemia measured in 1a-OHase homozygous 2/2 pups,statistically significant as early as weaning, manifests itselfearlier than in the VDR-ablated mice (36). This is difficult toprecisely evaluate from the literature, however, and maysimply reflect different diets and housing conditions in var-ious laboratories around the world.

The 1a-OHase 2/2 animals developed clear histologicalevidence of rickets and osteomalacia. The severe disorgani-zation of the growth plate of the adult 2/2 mice exceededwhat is normally observed in PDDR patients, because thoseare usually treated before the symptoms reach their fullmanifestation. The treatment of choice for PDDR is long-termreplacement therapy with 1,25(OH)2D3 (4, 5, 30). Vitamin Dand 25(OH)D3, at high doses, have been used, in the past,with some success (38). In this case, it is likely that massiveconcentrations of 25(OH)D3 can bind to the vitamin D re-ceptor and induce some response of the target organs. Phy-sicians in some countries also treat PDDR patients with themonohydroxylated analog, 1a(OH)D3, which is activated tothe hormonal form by hydroxylation at position 25 in theliver (39). It will be interesting to compare the various ther-apeutic interventions using the animal model of PDDR thatwe have engineered. Such rescue experiments are currentlyunderway.

Although the kidney is the main site of expression of the1a-OHase gene, its expression has been documented in othercell types, including osteoblasts, chondrocytes, keratino-cytes, and macrophages (6, 40–44). It is thought that localproduction of 1,25(OH)2D3 could play an important auto-crine or paracrine role in the differentiation or function ofthese tissues (45). The conditional 1a-OHase allele that wehave engineered will provide an invaluable genetic tool totest these hypotheses in the context of normocalcemic ani-mals. Mice carrying the loxP-bearing (floxed) allele are nowbeing bred to transgenic animals expressing the Cre recom-binase in relevant target tissues to address these questions.

Acknowledgments

We thank Dr. Sylvia Christakos for the calbindin D9k probe and Dr.Marie Demay for useful advice and for recommending the PTH assaykit. Anna Lis and Mireille Dussault provided technical assistance for theserum calcium, phosphate, 25(OH)D3, and 24,25(OH)2D3 measure-ments. Mark Lepik and Guylaine Bedard prepared the figures.



References

1. Miller WL, Portale AA 2000 Vitamin D 1a-hydroxylase. Trends EndocrinolMetab 11:315–319

2. Jones G, Strugnell SA, DeLuca HF 1998 Current understanding of the mo-lecular actions of vitamin D. Physiol Rev 78:1193–231

3. Heaney RP 1997 Vitamin D: role in the calcium economy. In: Feldman D,Glorieux FH, Pike JW (ed) Vitamin D. Academic Press, San Diego, pp 485–498

4. St-Arnaud R, Glorieux FH 2000 Hereditary defects in vitamin D metabolismand action. In: DeGroot LJ, Jameson JL (ed) Endocrinology, ed 4. W.B. SaundersCompany, Philadelphia, pp 1154–1168

5. Glorieux FH, St-Arnaud R 1997 Vitamin D pseudodeficiency. In: Feldman D,Glorieux FH, Pike JW (ed) Vitamin D. Academic Press, San Diego, pp 755–764

6. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL, Portale AA1997 Cloning of human 25-hydroxyvitamin D-1 a-hydroxylase and mutationscausing vitamin D-dependent rickets type 1. Mol Endocrinol 11:1961–1970

7. Fu GK, Portale AA, Miller WL 1997 Complete structure of the human genefor the vitamin D 1a- hydroxylase, P450c1a. DNA Cell Biol 16:1499–1507

8. St-Arnaud R, Messerlian S, Moir JM, Omdahl JL, Glorieux FH 1997 The25-hydroxyvitamin D 1-a-hydroxylase gene maps to the pseudovitamin D-deficiency rickets (PDDR) disease locus. J Bone Miner Res 12:1552–1559

9. Shinki T, Shimada H, Wakino S, Anazawa H, Hayashi M, Saruta T, DeLucaHF, Suda T 1997 Cloning and expression of rat 25-hydroxyvitamin D3–1a-hydroxylase cDNA. Proc Natl Acad Sci USA 94:12920–12925

10. Takeyama K, Kitanaka S, Sato T, Kobori M, Yanagisawa J, Kato S 199725-Hydroxyvitamin D3 1a-hydroxylase and vitamin D synthesis. Science277:1827–1830

11. Yoshida T, Monkawa T, Tenenhouse HS, Goodyer P, Shinki T, Suda T,Wakino S, Hayashi M, Saruta T 1998 Two novel 1a-hydroxylase mutationsin French-Canadians with vitamin D dependency rickets type II. Kidney Int54:1437–1443

12. Kitanaka S, Takeyama K, Murayama A, Sato T, Okumura K, Nogami M,Hasegawa Y, Niimi H, Yanagisawa J, Tanaka T, Kato S 1998 Inactivatingmutations in the 25-hydroxyvitamin D3 1a-hydroxylase gene in patients withpseudovitamin D-deficiency rickets. N Engl J Med 338:653–661

13. Labuda M, Morgan K, Glorieux FH 1990 Mapping autosomal recessive vi-tamin D dependency type I to chromosome 12q14 by linkage analysis. Am JHum Genet 47:28–36

14. Labuda M, Labuda D, Korab-Laskowska M, Cole DE, Zietkiewicz E, Weis-senbach J, Popowska E, Pronicka E, Root AW, Glorieux FH 1996 Linkagedisequilibrium analysis in young populations: pseudo-vitamin D-deficiencyrickets and the founder effect in French Canadians. Am J Hum Genet59:633–643

15. Wang JT, Lin CJ, Burridge SM, Fu GK, Labuda M, Portale AA, Miller WL1998 Genetics of vitamin D 1a-hydroxylase deficiency in 17 families. Am J HumGenet 63:1694–1702

16. Smith SJ, Rucka AK, Berry JL, Davies M, Mylchreest S, Paterson CR, HeathDA, Tassabehji M, Read AP, Mee AP, Mawer EB 1999 Novel mutations in the1a-hydroxylase (P450c1) gene in three families with pseudovitamin D-defi-ciency rickets resulting in loss of functional enzyme activity in blood-derivedmacrophages. J Bone Miner Res 14:730–739

17. Kitanaka S, Murayama A, Sakaki T, Inouye K, Seino Y, Fukumoto S, ShimaM, Yukizane S, Takayanagi M, Niimi H, Takeyama K, Kato S 1999 Noenzyme activity of 25-hydroxyvitamin D3 1a-hydroxylase gene product inpseudovitamin D deficiency rickets, including that with mild clinical mani-festation. J Clin Endocrinol Metab 84:4111–4117

18. Nagy A, Rossant J, Nagy R, Abramow-Newerly W, Roder JC 1993 Derivationof completely cell culture-derived mice from early-passage embryonic stemcells. Proc Natl Acad Sci USA 90:8424–8428

19. Hogan B, Beddington R, Costantini F, Lacy E 1994 Manipulating the MouseEmbryo: a Laboratory Manual, ed 2. Cold Spring Harbor Laboratory Press,Cold Spring Harbor, NY

20. Mifflin TE, Pearson WR, Reinhardt J, Bruns DE, Bruns ME 1988 Molecularcloning and sequencing of calbindin-D9k cDNA from mouse placenta. In:Norman AW, Schaefer K, Grigoleit HG, Herrath DV (ed). Vitamin D: Molec-ular, Cellular and Clinical Endocrinology. Proceedings of the Seventh Work-shop on Vitamin D. de Gruyter, New York, pp 507–508

21. Craig AM, Smith JH, Denhardt DT 1989 Osteopontin, a transformation-associated cell adhesion phosphoprotein, is induced by 12-O-tetradecanoyl-phorbol 13-acetate in mouse epidermis. J Biol Chem 264:9682–9689

22. Celeste AJ, Rosen V, Buecker JL, Kriz R, Wang EA, Wozney JM 1986 Isolationof the human gene for bone gla protein utilizing mouse and rat cDNA clones.EMBO J 5:1885–1890

23. Ohyama Y, Noshiro M, Okuda K 1991 Cloning and expression of cDNAencoding 25-hydroxyvitamin D3 24-hydroxylase. FEBS Lett 278:195–198

24. Piechaczyk M, Blanchard JM, Marty L, Dani C, Panabieres F, El Sabouty S,Fort P, Jeanteur P 1984 Post-transcriptional regulation of glyceraldehyde-3-phosphate-dehydrogenase gene expression in rat tissues. Nucleic Acids Res12:6951–6963

25. Dickson GR 1984 Methods of calcified tissue preparation. Elsevier, New York26. Panda DK, Al Kawas S, Seldin MF, Hendy GN, Goltzman D 2001 25-

hydroxyvitamin D 1a-hydroxylase: structure of the mouse gene, chromosomalassignment, and developmental expression. J Bone Miner Res 16:46–56

27. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Sel-znick SH, Dominguez CE, Jurutka PW 1998 The nuclear vitamin D receptor:biological and molecular regulatory properties revealed. J Bone Miner Res13:325–349

28. Scriver CR, Reade TM, DeLuca HF, Hamstra AJ 1978 Serum 1,25-dihy-droxyvitamin D levels in normal subjects and in patients with hereditaryrickets or bone disease. N Engl J Med 299:976–979

29. Rosen JF, Finberg L 1972 Vitamin D-dependent rickets: actions of parathyroidhormone and 25- hydroxycholecalciferol. Pediatr Res 6:552–562

30. Delvin EE, Glorieux FH, Marie PJ, Pettifor JM 1981 Vitamin D dependency:replacement therapy with calcitriol. J Pediatr 99:26–34

31. Mandla S, Jones G, Tenenhouse HS 1992 Normal 24-hydroxylation of vitaminD metabolites in patients with vitamin D-dependency rickets type I. Structuralimplications for the vitamin D hydroxylases. J Clin Endocrinol Metab74:814–820

32. Shinki T, Jin CH, Nishimura A, Nagai Y, Ohyama Y, Noshiro M, Okuda K,Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3–24-hydrox-ylase mRNA expression stimulated by 1 a,25-dihydroxyvitamin D3 in ratkidney but not in intestine. J Biol Chem 267:13757–13762

33. Lian JB, Stein GS, Stein JL, van Wijnen AJ 1998 Osteocalcin gene promoter:unlocking the secrets for regulation of osteoblast growth and differentiation.J Cell Biochem Suppl 31:62–72

34. Denhardt DT, Noda M 1998 Osteopontin expression and function: role in boneremodeling. J Cell Biochem Suppl 31:92–102

35. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997Targeted ablation of the vitamin D receptor: an animal model of vitaminD-dependent rickets type II with alopecia. Proc Natl Acad Sci USA94:9831–9835

36. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y,Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A,Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit im-paired bone formation, uterine hypoplasia and growth retardation after wean-ing. Nat Genet 16:391–396

37. Malloy PJ, Pike JW, Feldman D 1999 The vitamin D receptor and the syn-drome of hereditary 1,25-dihydroxyvitamin D-resistant rickets. Endocr Rev20:156–188

38. Balsan S, Garabedian M, Lieberherr M, Gueris J, Ulmann A 1979 Serum1,25-dihydroxyvitamin D concentrations in two different types of pseudo-deficiency rickets. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G,Coburn JW, DeLuca HF, Mawer EB, Suda T (eds) Vitamin D: Basic Researchand Its Clinical Application. Walter de Gruyter, Berlin, pp 1143–1149

39. Reade TM, Scriver CR, Glorieux FH, Nogrady B, Delvin E, Poirier R, HolickF, DeLuca HF 1975 Response to crystalline 1a-hydroxyvitamin D3 in vitaminD dependency. Pediatr Res 9:593–599

40. Turner RT, Puzas JE, Forte MD, Lester GE, Gray TK, Howard GA, BaylinkDJ 1980 In vitro synthesis of 1 a,25-dihydroxycholecalciferol and 24,25-dihy-droxycholecalciferol by isolated calvarial cells. Proc Natl Acad Sci USA77:5720–5724

41. Puzas JE, Turner RT, Howard GA, Brand JS, Baylink DJ 1987 Synthesis of1,25-dihydroxycholecalciferol and 24,25- dihydroxycholecalciferol by calvarialcells. Characterization of the enzyme systems. Biochem J 245:333–338

42. Reichel H, Bishop JE, Koeffler HP, Norman AW 1991 Evidence for 1,25-dihydroxyvitamin D3 production by cultured porcine alveolar macrophages.Mol Cell Endocrinol 75:163–167

43. Cadranel J, Garabedian M, Milleron B, Guillozo H, Akoun G, Hance AJ 19901,25(OH)2D2 production by T lymphocytes and alveolar macrophages recov-ered by lavage from normocalcemic patients with tuberculosis. J Clin Invest85:1588–1593

44. Schwartz Z, Brooks B, Swain L, Del Toro F, Norman A, Boyan B 1992Production of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 bygrowth zone and resting zone chondrocytes is dependent on cell maturationand is regulated by hormones and growth factors. Endocrinology130:2495–2504

45. Feldman D, Glorieux FH, Pike JW 1997 Vitamin D. Academic Press, San Diego



Targeted Inactivation of the 25-Hydroxyvitamin D 3 -1 α -Hydroxylase Gene (CYP27B1) Creates an...

Documents

Transcript of Targeted Inactivation of the 25-Hydroxyvitamin D 3 -1 α -Hydroxylase Gene (CYP27B1) Creates an...