Pseudo-vitamin D deficiency rickets
F H Glorieux
Genetics Unit, Shriners Hospital and Departments of Surgery, Pediatrics and Human Genetics, McGill University, Montr\l=e'\al, Qu\l=e'\bec, Canada H3G 1A6
(Requests for offprints should be addressed to F H Glorieux)
In 1961, Prader et al. (1961) reported on a new form of protein which belongs to the steroid-thyroid-retinoic acid
vitamin D resistant rickets, which differed from the classic receptor superfamily of genes (Evans 1988). It is comprised
hypophosphatemic type (Albright et al. 1937) by its early of two functional domains, a steroid hormone binding
onset (within the first year of life), the development of domain and a DNA-binding domain. Mutations affecting
severe hypocalcemia with tetany, moderate hypophos- both have been found in VDD2 families. In other in¬
phatemia and hyperaminoaciduria both reflecting second- stances there appears to be a VDR intranuclear retention
ary hyperparathyroidism, enamel hypoplasia and the defect (Barsony et al. 1990).
complete correction of all clinical and biochemical evi-
dence of rickets (including stunted growth rate) with high Serum vitamin D metabolite levels
daily doses of vitamin D. In the initial report identifying
this hereditary form of pseudo-vitamin D deficiency In most cases, serum calcitriol values are very low in
rickets (PDDR), transmission was reported to be auto- untreated PDDR patients (Delvin et al. 1981). Even when
somal dominant. Later, recessive inheritance was estab- treated with large doses of vitamin D causing major
lished (Fanconi & Prader 1969). In view of the fact that to increases in the circulating levels of calcidiol, calcitriol
maintain health, intake of vitamin D had to be consistently values do not reach the normal range (Scriver et al. 1978,
in vast excess of the recommended daily allowance Delvin et al. 1981). Right after birth PDDR infants have
(400 IU), the term 'vitamin D dependency' (VDD) very low calcitriol levels, months before any clinical
was proposed to describe the new syndrome (Scriver evidence of rickets develops. Some investigators have
1970). reservation at making low calcitriol levels an absolute
In 1973, through the availability of la,25(OH)2D3 as a criterion. Indeed, Balsan (1991) observed PDDR patients
therapeutic agent, came the demonstration that PDDR with normal calcitriol levels. Those values, however,
was an inborn error of vitamin D metabolism involving the should be considered as inappropriate in the face of rickets,
defective conversion of calcidiol (25(OH)D) to calcitriol hypocalcemia and secondary hyperparathyroidism. It may
(la25(OH)2D) (Fraser et al. 1973, and the biosynthetic well be that these differences reflect a degree of genetic
pathway shown in Fig. 1). Several groups subsequently heterogeneity among PDDR pedigrees, that will only be
showed that small (physiologic) daily amounts of calcitriol resolved at the molecular level.
or its analog, la(OH)D3, allowed excellent and prolonged Infant death by hypocalcemia or pulmonary infections
control of the abnormal phenotype (Balsan et al. 1975, was not infrequent in the past when the diagnosis was
Reade et al. 1975, Delvm et al. 1981). Thus the basic either missed (confused for a neurologic or respiratory
PDDR abnormality appears to be an altered activity of disease) or made too late. Treatment with large doses of
the renal la-hydroxylase (la-OHase) enzyme, although vitamin D remains the only solution in countries where
the exact nature of the mutation remains to be identified. calcitriol or its analogue are not available. However, the
As is often the case in autosomal recessive disorders, therapeutic doses are close to the toxic ones and renal
hétérozygotes for the trait cannot be characterized either damage remains a threat. Furthermore, adjustments made
clinically or biochemically, but only by analysis of their necessary by immobilization or pregnancy are much more
progeny. difficult to monitor because of the large accumulated stores
In 1978, another inborn error of vitamin D metabolism of vitamin D (and calcidiol) that may take months to be
was isolated in which a clinical picture of pseudo-vitamin reduced.
D deficiency developed despite high circulating concen¬ The treatment of choice is replacement therapy with
trations of endogenously produced calcitriol (Marx et al. calcitriol. If rickets is florid, the initial dose may be
1978). In some pedigrees, the phenotype was com¬ between 2 and 3 µg/day until bone is healed, thereafter
pounded with complete alopecia (Rosen et al. 1979). This the maintenance dose will vary between 0-25 and 1 µg/
second variety of pseudo-deficiency, also referred to as day and remain remarkably stable throughout growth and
'vitamin D dependency type (VDD2) is caused by a adulthood (our oldest patient is a 50-year-old woman,
spectrum of mutations affecting the vitamin D receptor maintained for the last 15 years on a daily dose of
(VDR) in target tissues. The human VDR is a 50 kDa 0 25µg/day).
Figure 1 Vitamin D biosynthetic pathway. The precursot of vitamin D3, 7-dehydrocholesterol, is present in
the skin. Upon exposure to sunlight it is converted to pre-vitamin D3 which will thermally ¡sometize to
cholecalciferol (vitamin D3). The compound may be stored in fat and muscle tissues from which it is
released to reach the liver where it is converted by a hepatic microsomal hydroxylase to Q\>
25-hydtoxycholecalciferol (25(OH)D3). The latter is biologically inert and requires a further hydroxylation in
the kidney to 1,25-dihydroxyvitamin D3 (1a,25(OH)2D3), the hormonal form of the vitamin. Its synthesis is
triggered by a decrease in serum calcium which induces the release of parathyroid hormone. Parathyroid
hormone travels to the kidney and increases tubulat reabsorption of calcium and the excretion of
phosphate. In addition, possibly through a lowering of intracellular phosphate concentration, this peptide
hormone increases the conversion of 25-hydroxy- into 1,25-dihydroxyvitamin D. The alternate kidney
conversion (25-hydroxy- to 24,25-dihydroxyvitamin D) is currently thought to be a detoxification route, no
specific physiologic role having yet been assigned to 24,25-dihydroxyvitamin D (24,25(OH)2D3).
Placenta studies their collaboration and the one of their relatives set out to
map the PDDR locus by using DNA markers and linkage
While in non-pregnant subjects lct-OHase activity is analysis to approach the primary defect in PDDR. We
mainly located in kidney, we have demonstrated that this found that the gene responsible for the disease was linked
enzyme was also present in cells isolated from human to polymorphic restriction fragment length polymorphism
decidua (Delvin et al. 1985). Interestingly, when studying (RFLP) markers in region 14 of the long arm of chromo¬
decidual tissues from two PDDR patients, we found some 12 (12ql4). Multipoint linkage analysis and studies of
that it did not have the capacity to produce calcitriol, haplotypes and recombinants strongly suggest the localiza¬
indicating that decidua and kidney are both targets for tion of the PDDR locus between the collagen type II alpha
the PDDR mutation (Glorieux et al. 1995). Although the 1 (COL2A1) locus and a cluster of three anonymous
physiologic importance of this defect is unclear, the probes (D12S14, D12S17 and D12S6) which segregate as
finding provides us with a unique, albeit infrequent, a three-marker haplotype. Linkage disequilibrium be¬
source of mutant cells. tween PDDR and this three-marker haplotype supports
the notion of a founder effect in the studied population.
Furthermore, this localization of PDDR between flanking
Linkage analysis markers allows for the prediction of the carrier status in
individuals at risk in the analyzed families as well as early
Although quite rare, PDDR is present with unusual diagnosis in newborns. The accuracy of carrier detection
frequency in a subset of the French Canadian population. could exceed 99% in families informative for the markers
Over the years we have followed 36 patients and with used (Labuda et al. 1990).
Mapping PDDR and VDR different from that of PDDR, previously assigned to 12ql4
(Labuda et al. 1990), thus excluding it as a target of the
Recently the VDR gene has also been assigned to PDDR mutation.
chromosome 12 by Southern blot analysis of a panel of To understand the basis of the PDDR genotype it is
human—Chinese hamster cell hybrid DNAs. VDR cDNA necessary to characterize an appropriate cDNA probe
was also shown to reveal Apal polymorphism, allowing its corresponding to la-OHase activity. Preliminary results
use as an RFLP marker in linkage analysis. By in situ towards that goal have been recently obtained in this Unit,
hybridization VDR was found to map in the 12ql2-14 by St-Arnaud et al. (1996). Using a probe centered on the
region. Simultaneously, by using samples of 21 from our heme binding domain of the rat 24R-OHase cDNA we
PDDR families, we established that the VDR and PDDR have screened a cDNA library from kidneys of vitamin
loci were mapping in close proximity on 12ql2—14. It is D-deficient rats under reduced stringency conditions to
likely that the genetic distance between VDR and PDDR identify and clone the la-OHase cDNA. A 2-4 kb full-
is in the range of a few centiMorgans which in physical length clone was identified which codes for a protein of
terms can correspond to 1-10 megabases. We find, at predicted molecular mass of 55 kDa. Amino acid sequence
present, no specific reason for this proximity, but its similarity with the 24R-OHase enzyme was 78% within
functional significance might be established in the future the heme binding domain, but clear divergence was seen
(Labuda et al. 1992). outside this region for an overall sequence similarity of
26%. Transfection of embryonal carcinoma cells with the
sense la-OHase expression vector produced a vitamin
Molecular defect in PDDR D metabolite that co-migrates on HPLC with the
la25(OH)2D3 standard. The compound was also authen¬
Although it is clear that PDDR is due to a deficiency of ticated using two different radioreceptor assays. This rat
the renal mitochondrial la-OHase, leading to insufficient la-OHase cDNA probe will allow cloning of its human
synthesis of calcitriol (Fraser et al. 1973), the exact target of homolog, which will then be mapped to its genomic
the mutation is not known. It could affect either modula¬ location and the hypothesis that it will co-localize with the
tors of the enzyme activity or the structure of the enzyme PDDR locus tested. If it did not, then the PDDR
itself. The la-OHase is made of three distinct moieties: mutation is probably not affecting the integrity of
cytochrome P450, ferredoxin and ferredoxin reductase la-OHase but rather a controlling factor of its activity.
(reviewed in Chazarían 1990). A priori, each of these could Finally, heterogeneity (in the sense of different loci
be affected by the mutation. However, it has recently been being involved) cannot be excluded in PDDR. We are
shown that the two components of the electron-transfer currently gathering specimens from PDDR pedigrees
system which are shared with other cytochrome P450s are from other ethnic groups to ascertain whether the PDDR
encoded by single copy genes expressed in all steroido¬ locus maps there at the same genomic location as in the
genic tissues and located on chromosome 11 for the French Canadian families so far studied. These studies
ferrodoxin (Morel et al. 1988), and on chromosome 17 for should shed an interesting light on the control of vitamin
the ferredoxin reductase (Solish et al. 1988). Since PDDR D bioactivation.
patients express no evident deficiencies in the activity of
other cytochrome P450 systems, the P450 specific com¬
ponent of la-OHase remains as the most likely target for References
the PDDR mutation.
The fact that 24R-OHase activity is normal in PDDR
Albright F, Butler AM Sc Bloomberg E 1937 Rickets resistant to
does not eliminate the possibility that the cytochromes vitamin D therapy. American Journal of Disabled Children 54 529—547.
P4501a and P45024R, which possess similar amino acid Balsan S 1991 Hereditary pseudo-deficiency rickets. In Rickets,
Nutrition Workshop Series, vol 21, pp 155-165. Ed FH Glorieux.
composition, are under the control of one single gene New York: Raven Press.
(Ghazanan 1990). In this hypothesis, P4501a is a 57 kDa Balsan S, Garabedian M, Sorgniard R, Holick MF & DeLuca HF
protein which contains a 2 kDa C-terminal fragment 1975 1,25-dihydroxyvitamin D3 and 1 ,alpha-hydroxyvitamin D3 in
which is cleaved to yield the 55 kDa P45024R expressing children: biologic and therapeutic effects in nutritional rickets and
24R-OHase activity. There is however no evidence, at different types of vitamin D resistance. Pediatrie Research 9 593-599.
Barsony J, McKoy W, Smith C, DeMarco L, Kohidai L, Liberman
present, that the one enzyme hypothesis proposed for the UA & Marx SJ 1990 Heterogeneous coupling mechanisms between
avian system can be readily extrapolated to the organiza¬ vitamin D receptors and their target actions. In Calcium Regulation
tion of mammalian hydroxylases. In support of this later and Bone Metabolism, pp 174-178. Eds DV Cohn, FH Glorieux &
view we have recently cloned part of the human P45024R TJ Martin. Amsterdam, The Netherlands: Elsevier.
cDNA. By Southern blot hybridization of DNA from Delvin EE, Glorieux FH, Marie PJ & Pettifor JM 1981 Vitamin
human-hamster somatic cell hybrids and by in situ D-dependency: replacement therapy with calcitriol. Journal of
Pediatrics 99 26-34.
immunofluorescence hybridization, we mapped it to Delvin EE, Arabian A, Glorieux FH & Mamer OA 1985 In vitro
chromosome 20ql3.1 (Labuda et al. 1993). This location is metabolism of 25-hydroxycholecalciferol by isolated cells from
human decidua. Journal of Clinical Endocrinology and Metabolism 601 decrease in sensitivity to 1,25-dihydroxyvitamin D. Journal of
880-885. Clinical Endocrinology and Metabolism 47 1303—1310.
Evans RM 1988 The steroid and thyroid hormone receptor Morel Y, Picado-Leonard J, Wu DA, Chang CY, Mohandas TK,
superfamily. Science 240 889-895. Chung BC & Miller WL 1988 Assignment of the functional gene
Fanconi A & Prader A 1969 Die hereditäre mangelrachitis. Helvetica for human adrenodoxin to chromosome llql3—»ter and of
Pediatrica Acta 24 423-447. adrenodoxin pseudogenes to chromosome 20cen—»13.1. American
Fraser D, Kooh SW, Kind HP, Hollick MF, Tanaka Y Sc DeLuca HF Journal of Human Genetics 43 52-59.
1973 Pathogenesis of hereditary vitamin D dependent rickets. An Prader A, lllig R & Heierli E 1961 Eine besondere form des primäre
inborn error of vitamin D metabolism involving defective vitamin D-resistenten rachitis mit hypocalcämie und
conversion of 25-hydroxyvitamin D to la,25-dihydroxyvitamin D. autosomal-dominanten Erbgang: die hereditäre
New EnglandJournal of Medicine 289 817-822. PseudoMangelrachitis. Helvetica Pediatrica Acta 16 452-468.
Ghazarian JG 1990 The renal mitochondrial hydroxylases of the Reade TM, Scriver CR, Glorieux FH, Nogrady B, Delvin E, Poirier
vitamin D3 endocrine complex: How are they regulated at the R, Holick MF & DeLuca HF 1975 Response to crystalline
molecular level? Journal of Bone and Mineral Research 5 897-902. 1 -hydroxyvitamin D3 in vitamin D dependency. Pediatrie Research
Glorieux FH, Arabian A & Delvin EE 1995 Pseudo-vitamin D 9 586-593.
deficiency: Absence of 25-hydroxyvitamin D 1 a-hydroxylase Rosen J F, Fleischman AR, Finberg L, Hamstra AJ & DeLuca 1979
activity in human placenta decidual cells. Journal of Clinical Rickets with alopecia: an inborn error of vitamin D metabolism.
Endocrinology and Metabolism 80 2255-2258. Journal of Pediatrics 94 729-735.
Labuda M, Morgan & Glorieux FH 1990 Mapping autosomal
Scriver CR 1970 Vitamin D dependency. Pediatrics 45 361-363.
recessive vitamin D dependency type I to chromosome 12ql4 by
Scriver CR, Reade TM, Hamstra AJ & DeLuca HF 1978 Serum
linkage analysis. American Journal of Human Genetics 47 28—36.
Labuda M, Fujiwara TM, Ross MV, Morgan K, Garcia-Heras J, 1,25-dihydroxyvitamin D levels in normal subjects and in patients
Ledbetter DH, Hughes MR & Glorieux FH 1992 Two hereditary with hereditary rickets or bone disease. New England Journal of
defects related to vitamin D metabolism map to the same region of Medicine 299 976-979.
human chromosome-12ql3—14. Journal of Bone and Mineral Research Solish SB, Picado-Leonard J, Morel Y, Kuhn RW, Mohandas TK,
7 1447-1453. Hanukogly I & Miller WL 1988 Human adrenodoxin reductase:
Labuda M, Lemieux N, Tihy F, Prinster C & Glorieux FH 1993 The two mRNAs encoded by a single gene on chromosome 17cen—»25
human 25(OH)D-24 hydroxylase cytochrome P450 subunit maps to are expressed in steroidogenic tissues. Proceedings of the National
a different chromosomal location than that of pseudovitamin D Academy of Sciences of the USA 85 7104-7108.
deficient rickets. Journal of Bone and Mineral Research 8 1397-1406. St-Arnaud R, Moir JM, Messerhan M & Glorieux FH 1996
Marx SJ, Spiegel AM, Brown EM, Gardner DG, Downs RW, Attie Molecular cloning and characterization of a cDNA for vitamin D
M, Hamstra AJ & DeLuca HF 1978 A familial syndrome of la-hydroxylase. Journal of Bone and Mineral Research 11 S124.