Positional cloning of the HYP gene: A review

Kidney Intemationa4 Vol. 49 (1996), pp.

1033—1037

Positional cloning of the HYP gene: A review MICHAEL J. ECONS Department of Medicine and The Sarah W Stedman Center For Nutritional Studies, Duke University Medical Center, Durham, North Carolina, USA

X-linked hypophosphatemic rickets (HYP) is the most common linkage (100:1) [12]. LOD scores are calculated over a range of form of familial hypophosphatemic rickets. Patients variably recombination fractions (6), the genetic distance between two prcsent with lower extremity deformities, rickets, short stature, linked markers. The best estimate for the reeombination fraction bone pain, dental abscesses, enthesopathy, and osteomalacia [1]. (6) is that value at which the LOD score is maximized [Z(6)]. It is an X-linked dominant disorder [21 characterized by decreased Additionally, investigators can determine whether a particular renal tubular phosphate reabsorption and consequent hypophos- genetic marker is centromerie or telomeric to a disease gene by phatemia. Observations in the Hyp mouse, one of the murine using multilocus analysis and/or analyzing individual reeombihomologues of the human diseasc, indicate that the disorder is not due to an intrinsic renal defect [3—6]. However, the HYP gene may directly or indirectly regulate the expression of the sodium dependent phosphate cotransporter (NPT2) in the renal proximal

nants.

Linkage studies in X-linked hypophosphatemic rickets Initial studies by Machler et al [13] and Read et al [14] localized

tubule [7]. Despite extensive study, the pathophysiology of the disorder has not been elucidated. To gain better understanding of the HYP gene to Xp22 using restriction fragment length polymorphisms (RFLP). Subsequently, Thakker et al [15] determined that this disorder we have used the positional cloning approach to DXS41 and DXS43 are flanking markers for the HYP gene. locate and clone the HYP gene. However, the flanking markers were much too far apart to enable investigators to isolate and clone the HYP gene. To refine the The positional cloning approach

genetic map around the HYP locus we tested several RFLP

The positional cloning approach has been used to identify genes markers in the Xp22 region for linkage to HYP in 5 large kindreds that are responsible for a variety of diseases [8]. The advantage of this approach is that knowledge of gene function and tissue [16]. We determined that HYP is located between DXS2S7 on the telomeric side and DXS41 on the centromeric side. The distance expression are not necessary to locate the disease gene. Thus, one between these two flanking markers is approximately 3.5cM. does not need to make any assumptions about the gene. In these Unfortunately, although the marker DXS365 was tightly linked to studies investigators use linkage analysis to determine the chroHYP [Z(6) = 13.98 at 6 = 0.0], we were unable to locate DXS365

mosomal location of the disease gene and to obtain genetic

with respect to HYP since there were no reeombination events

markers that flank the gene. Subsequently, they attempt to obtain between DX5365 and HYP. Indeed the RFLP for DX5365 was "contig" maps, which are overlapping pieces of human DNA in not informative in the one mating that demonstrated a recombicosmid or yeast artificial chromosome vectors, that span the nation event between DX5257 and HYP [16]. To be an informaregion between the flanking markers. They locate the genes that tive mating the affected parent of the recombinant individual must are contained within the contig and test them for mutation in be heterozygous for the marker. Additional linkage data obtained affected individuals [reviewed in 9—11]. The first step in positional cloning is linkage analysis. Linkage analysis takes advantage of the fact that reeombination is a fairly frequent event in meiosis. When

by Rowe et al[17] determined that DXS274 was linked to the HYP gene [Z(0) = 4.2 at 6 = 0.0], but they were unable to locate it with respect to HYP since there were no reeombination events precursor cells undergo meiosis, homologous chromosomes re- between the RFLP DXS274 and HYP. Since tightly linked combine and exchange segments of equal length. This results in flanking markers are necessary prerequisites to obtain the HYP the separation of loci on a chromosome, a process referred to as crossing over. As a result, the germ cells contain chromosomes gene by positional cloning techniques, we extended our linkage studies to include an additional large HYP kindred to enable us to that are a patchwork of segments from the two parental chromodetermine the positions of DXS365 and DX5274 relative to the somes. The closer two loci lie on the same chromosome the less often they are separated during recombination. If two loci do not freely recombine they are said to be linked. To quantify linkage between two loci investigators use LOD scores. The LOD score is equal to the log10 of the odds of linkage at a particular genetic distance between two loci. A LOD (Z) score of 3 is evidence of

linkage (lOt)0:1) and a LOD score of -2 is evidence of non-

HYP gene [18]. With the expanded family resources we found only one recombination event between DX5365 and HYP [Z(6) = 17.43 at 6 = 0.012] and observed in this mating a recombination event between HYP and the previously determined telomeric marker DX5257.

In contrast, in the same mating there was no recombination between HYP and the more centromeric markers DX5274 and DXS4I. In addition, we discovered two matings that demon-

strated reeombination events between HYP and DX5274 © 1996 by the International Society of Nephrology

[Z(O) = 1033

22.15

at 6 =

0.018].

Both of these matings placed

1034

Econs: Positional cloning of the HYP gene

DXS274 on the centromeric side of the HYP gene. These tively short and since cosmids have several advantages over YACs, observations established that DXS365 is telomeric to the HYP

we constructed a cosmid contig [28] across the region (Fig. 1). The

gene, while DXS274 is centromeric to the HYP gene [18]. Simultaneously, Rowe et al [19] made a microsattelite repeat marker for DXS274. Microsattelite repeat markers are multiallele markers [201 that are generally more polymorphic than

cosmid contig allowed us to change our approach to mutation detection. In addition to trying to isolate cDNAs from the HYP region and test them for mutation in our 20 large kindreds, we used the cosmids to screen large numbers of affected individuals for deletions. These affected individuals were either members of small kindreds that were not suitable for linkage studies or were isolated cases. Through the collaborative efforts of the five laboratories that make up the HYP consortium [28] we obtained

RFLPs. As a result they tend to have far fewer noninformative matings. Combining their results for the microsattelite repeat for DXS274 and the RFLP for DX5274, Rowe et al determined that DXS274 lies centromeric to the HYP gene and telomeric to DXS41 [19]. Thus the locus order supported by the above studies DNA samples from approximately 150 unrelated affected individwas Xtel-DXS43-(DXS257/DXS365)-HYP-DXS274-DXS41-Xcen, uals. We looked for deletions by hybridizing whole cosmids to where a "1' between loci indicates that the order between them Southern blots of restriction enzyme digested genomic DNA from could not be determined. To determine the relative location of these individuals. Although deletions have never been found in HYP patients by DXS257 and DXS365, we combined HYP kindreds from two groups of investigators [18, 19] and analyzed recombination cytogenetic techniques, we thought that there might be small events in these families with a newly available microsattelite deletions that would be detected by Southern blotting. Detection repeat for DXS365 [211. Our data demonstrated that DXS365 was of a deletion with a cosmid from the contig would be a strong centromeric to DXS257 and was the closest telomeric marker to indication that the cosmid contained the HYP gene. We found three affected individuals who demonstrated deleHYP [22]. The new order was now Xtel-DXS43-DXS257DX5365-HYP-DXS274-DXS41-Xcen.

Physical mapping and the YAC contig map

The flanking markers, DXS365 on the telomeric side and DXS274 on the centromeric side, were now close enough to try to

tions in DNA contained within cosmid 611. Two of these deletions

are approximately 6 and 17 kb in size. The third deletion is over 55 kb and extends on the telomeric side of cosmid 611 to include cosmid 177. We found a forth affected individual who demonstrated a small (approximately 1 kb) deletion within cosmid 1005,

bridge the distance between them with yeast artificial chromo- which lies immediately centromeric to cosmid 611 [28]. This latter somes (YAC). YACs are yeast vectors that contain large pieces of deletion does not overlap with the other three deletions. Thus, we human DNA (up to 1 MB) and can be propagated in yeast [23]. focused our efforts on these cosmids. DX5365 and DXS274, and/or cosmids that contained these Cloning the HYP gene probes, were used to screen YAC libraries [24]. On the centroTo clone the gene the members of the HYP consortium meric side DXS274 identified four nonchimeric YACs. Two of these also contained the more centromeric marker DXS41. The employed two complementary approaches. We used the cosmids ends of YAC 83B05, which contains both DXS274 and DXS41, that contained the deletions to screen eDNA libraries. Additionwere isolated and used to screen the four YACs that were ally, we performed automated sequencing of the cosmids and identified with DXS274. The trp end of YAC 83B05 hybridized to looked for exons within the genomic sequence. Both lines of YAC 80G04, which does not contain the probe DXS41. Thus, the investigation revealed the presence of only one gene in the trp end of YAC 83B05 was the telomeric end and we used a approximately 100 kb region. Analysis of the available partial cosmid that contained this trp end to rescreen YAC library filters. sequence of this gene indicates that it contains significant homolThis cosmid identified several new YACs. On the telomeric side ogy at the peptide level to a family of endopeptidase genes, which of the HYP gene DXS365 identified two nonchimeric YACs. includes neutral endopeptidase, endothelin-converting enzyme-i Alu-PCR [25] was used to amplify sequences from one of these and the Kell antigen. Further analysis revealed that in all four YACs and these PCR products were used to rescreen the YAC patients who had deletions, the deletions involved at least one library. Five YACs were identified by this hybridization and three of these had already been identified by the trp end of YAC 83B05.

Thus, as illustrated in Figure 1, we were able to span the HYP

region with three YACs and the physical distance between flanking markers was approximately 1.5 Mb [24]. To further delineate the HYP region we derived new microsatellite markers, DXS 1683 and DXS7474, from two of the YACs

exon from this gene [28]. We labeled this gene "PEX" for phosphate regulating gene with homologies to endopeptidases on the X chromosome. Although computer searches of the genomic sequence failed to identify any other gene in the region, it was possible (although unlikely) that the PEX gene was not the HYP gene, but simply

another gene in the region that was also deleted in our four

[26, 27]. Both of these markers were physically mapped to lie between the flanking markers DXS365 and DXS274 [24]. We tested these markers in 20 large HYP kindreds. Two recombinants were seen between DXS 1683 and HYP. Both of these matings placed DXS 1683 on the centromeric side of the HYP gene [22]. Similarly, there were two recombination events between DXS7474 and HYP which placed DXS7474 telomeric to HYP [27]. The physical distance between these new flanking markers is approximately 350 kb and they both are contained on

affected individuals. To obtain more definitive evidence that the PEX gene is the HYP gene, we looked for point mutations in our patients. We determined the intron/exon boundaries for two PEX exons and designed primers to amplify across the splice sites. PCR products from our HYP families as well as controls were analyzed by single stranded conformational polymorphism (SSCP). This technique takes advantage of the fact that under nondenaturing conditions single stranded DNA has a folded conformation that is

one YAC [27].

hence electrophoretic mobility of the molecule is dependent on its

Since the distance between the new flanking markers is rela-

stabilized by intrastrand interactions. The conformation and sequence. Thus, single base pair changes alter the secondary

1035

Econs: Positional cloning of the 1-JYP gene

YAC/COSMID CONTIG MAP

DXS7474

DXS365

1/

O0E01 138 (560kb)

N

cb)

I

r,i

9O0OJ59 (6601W)

:iiii

N NI

83005 (660kb) 173A01 (370kb)

I

27C06 (400kb

I

80004 (440kb)

I

I

DXS274 DXS41

cosmid H10174

DXS1683

I

135011 (460kb)

'(AC 900A0472

rrrsssrssssrsrssssssrsssa 900E01 138(560kb)

DXS7474

DXS1 683

HYP

cos362

cos2O9l 1

con 47

cos0262

coslOOS

cosi 857

Co 635

cos65B'

cos56lO I

cos42l

I

cos6ll

cos437' cosl 77

Fig. 1. Schematic representation of YAC and cosmid contigs. Genetic markers are positioned at the top of the Figure from telomeric (DXS365) to centromcric (DXS41). YACs 900EOl138, 900A0472 and 83B05 form the minimal YAC contig. YAC 900E01138 has a small deletion (not shown) in the region of overlap with YAC 900A0472. A more detailed view of the HYP region is presented in the lower half of the figure, which contains the cosmid contig. Adapted from references [24] and [281.

structure of the DNA and result in band shifting on nondenatur- It is possible that, in the normophosphatemic individual, the PEX ing gels [29]. SSCP analysis of the two exons revealed aberrant gene is expressed at levels that are insufficient to allow detection bands in three affected individuals. Sequencing these bands by Northern blotting. Alternatively, we may not have tested the demonstrated a frameshift mutation in one patient caused by loss appropriate tissue yet. More detailed studies of tissue expression of a TC dinucleotide in the middle of one of the exons. Sequenc- are currently underway. ing of the PCR product from the other two affected individuals Mechanism revealed two point mutations in the invariant dinucleotides of the splice acceptor site of the other exon. Both of these latter Any model that proposes to explain the role of the PEX gene in mutations would potentially lead to exon skipping and this was normal phosphate homeostasis and how mutations in the PEX confirmed in one of the patients by RT-PCR experiments [281. gene cause the HYP phenotype must take into account several These point mutations establish PEX as the HYP gene and we are observations. First, HYP is an X-linked dominant disorder [2]. HYP could be a haploinsufficiency disorder where having half the currently defining additional mutations. Tissue expression

normal complement of the normal PEX gene could lead to disease. Alternatively, mutation of the PEX gene could result in a

Although RT-PCR has been successfully performed using dominant negative effect. Second, the two known sodium depen-

dent phosphate cotransporters are on chromosome 5q35 [30] and t5p [311. Data presented on the Hyp [7] and Gy [32] mice, murine cDNAs have been used to scrccn Northern blots from multiple homologues of the human disease, indicate that there is approxhuman and mouse tissues. So far, no signals have been detected. imately a 50% reduction in NPT2 mRNA in mutant mice, lymphocyte and fetal brain RNA, it is not likely that these are the most physiologically relevant sites of tissue expression. PEX gene

1036

Econs: Positional cloning of the HYP gene

indicating that murine Hyp/Gy gene has a role in the regulation of

Reprint requests to Michael .1. Econs, M.D., Box 3298, Duke University

the osteoblast or the abnormal environment in the Hyp mouse has

1. EcoNs MJ, DREZNER MK: Bone disease resulting from inherited disorders of renal tubule transport and vitamin D metabolism, in

NPT2 expression. Third, when osteoblasts from Hyp mice are Medical Center, Durham, North Carolina 27710, USA. e-mail: transplanted into the normal mouse there is a persistent miner- EconsOOl @mc.duke.edu alization defect despite the normal hormonal and biochemical References environment [33, 341. Thus, either the HYP gene is expressed in

prolonged effects on the osteoblast. Fourth, studies done by Nesbitt and colleagues [5] demonstrate that when kidneys from Hyp mice are transplanted into nephrectomized normal mice the kidneys conserve phosphorous normally. When kidneys from normal mice are transplanted into nephrectomized Hyp mice the kidneys waste phosphorus. Thus, the disease is neither transferred nor corrected by renal cross-transplantation and the kidney is not the major organ of PEX expression. Fifth, when Hyp and normal mice are joined together in parabiosis experiments the normal mouse becomes hypophosphatemic. Parabiosis does not correct the Hyp phenotype. This implies that the Hyp phenotype is caused

by the elaboration of a humoral factor that is produced in excessive amounts [3, 4]. Indeed, there are tumors that produce isolated renal phosphate wasting, which resolves with resection of the tumor. Unfortunately, this factor(s), which we have referred to as phosphatonin has only been partially purified [35, 361. There are several possible mechanisms of action of the normal PEX gene product. Since the deletions and frameshift mutations that we observed in the PEX gene in DNA from several affected individuals are likely to cause loss of function of the PEX gene product, it is extremely unlikely that these mutations are activating mutations. Thus, PEX is not the phosphatonin gene. It is more likely that the PEX gene has other roles in phosphate homeostasis. The PEX protein could normally function to degrade phos-

phatonin. Thus, mutations in the PEX gene would result in inability to degrade phosphatonin and lead to renal phosphate wasting. However, if this were the case, parabiosis of Hyp and normal mice would have led to rescue of the Hyp phenotype rather than renal phosphate wasting in the normal mouse, since one would expect that the normal mouse would be able to degrade

the excess phosphatonin. This assumes that the PEX protein would come into contact with phosphatonin before phosphatonin could cause renal phosphate wasting. Another possibility is that

the PEX protein normally functions to activate a phosphate conserving hormone. This possibility becomes more plausible when one considers data presented at this meeting that human Stanniocalcin stimulates phosphate reabsorption when administered to rats [371. However, this model would also predict rescue of the Hyp phenotype in the parabiosis experiments. A fourth possibility that fits in with the currently available data are that the PEX gene functions to inhibit the expression of phosphatonin indirectly. Thus, mutations of the PEX gene would indirectly

Disorders of Bone and Mineral Metabolism, edited by FAvus MJ, C0E FL, New York, Raven Press, Ltd., 1992, pp 935—950 2. WINTERS RW, GRAHAM JB, WILLIAMS TF, MCFALLS VW, BURNEYF

CH: A genetic study of familial hypophosphatemia and vitamin D resistant rickets with a review of the literature. Medicine 37:97—142, 1958

3. MEYER RA JR, MEYER MH, GRAY RW: Parabiosis suggests a humoral factor is involved in X-linked hypophosphatemia in mice. J Bone Miner Res 4:493—500, 1989 4. MEYER RA JR, TENENHOUSE HS, MEYER MH, KLUGERMAN AH: The

renal phosphate transport defect in normal mice parabiosed to X-linked hypophosphatemic mice persists after parathyroidectomy. J Bone Miner Res 4:523—532, 1989

5. NESBITr T, COFFMAN TM, GRIFFITHS R, DREZNER MK: Crosstransplantation of kidneys in normal and Hyp mice: Evidence that the Hyp phenotype is unrelated to an intrinsic renal defect. J Clin Invest 89:1453—1459, 1992

6. NESBITr T, EcoNs MJ, BYUN JK, MARTEL J, TENENHOUSE HS, DREZNER MK: Phosphate transport in immortalized cell cultures from the renal proximal tubule of normal and hyp-mice: Evidence that the 1-JYP gene locus product is an extrarenal factor. J Bone Miner Res 10:1327—1333, 1995

7. TENENHOUSE HS, WERNER A, BIBER J, MA S. MARTEL J, ROY S,

MURER H: Renal Na-phosphate cotransport in murine X-linked hypophosphatemic rickets: Molecular characterization. J Clin Invest 93:671—676, 1994

8. COLLINS F: Positional cloning moves from perditional to traditional. Nature Genet 9:347—350, 1995 9. WHITE R, LALOUEL JM: Chromosome mapping with DNA markers. Sci Am 258:40—48, 1988 10. FARRER LA: Gene localization by linkage analysis. Otolatyngologic Clin NAm 25:907—922, 1992 11. BOUGHMAN JA, STICK MJ, PETERSON DA, COHEN MM: Linkage analysis and predicting genetic disease. Clin Lab Med 12:449—461, 1992 12. MORGAN NE: Sequential tests for the detection of linkage.AmJHum Genet 7:277—318, 1955 13. MACHLER M, FREY D, GAL A, ORTH U, WIENKER TF, FANCONI A,

SCHMID W: X-linked dominant hypophosphatemia is closely linked to DNA markers DXS41 and DXS43 at Xp22. Hum Genet 73:271—275, 1986 14. READ AP, THAKKER RV, DAVIES KE, MOUNTFORD RC, BRENTON DP, DAVIES M, GLORIEUX F, HARRIS R, HENDY GN, KING A, MCGLADE

S, PEACOCK CJ, SMITH R, O'RIORDAN JLH: Mapping of human X-linked hypophosphatemic rickets by multilocus linkage analysis. Hum Genet 73:267—270, 1986 15. TFIAKKER RV, READ AP, DAVIES KE, WI-IYTE MP, WEKSBERO R, GLORIEUX F, DAVIES M, MOUNTFORD RC, HARRIS R, KING A, KIM

GS, FRASER D, KOOH SW, O'RIORDAN JLH: Bridging markers defining the map position of X-linked hypophosphatemic rickets. J

result in over expression of phosphatonin and lead to renal

Med Genet 24:756—760, 1987 16. ECONS MJ, BARKER DF, SPEER MC, PERICAK-VANCE MA, FAIN PR,

phosphate wasting. While it is clear that much work needs to be done, the cloning

DREZNER MK: Multilocus mapping of the X-linked hypophos-

of the PEX/HYP gene will provide us with new insights into X-linked hypophosphatemic rickets and normal phosphate homeostasis. Data presented at this conference indicate that the control of serum phosphate is a fascinating, but complex process. Acknowledgments Work done in the author's laboratory was supported by grants MOlRR-30, AR27032, and AR42228 from the National Institutes of Health.

phatemic rickets gene. J Clin Endocrinol Metab 75:201—206, 1992 17. ROWE PSN, READ AP, MOUNTFORD R, BENHAM F, KRUSE TA, CAMERINO G, DAVIES KE, O'RIoRDAN LH: Three DNA markers for hypophosphatemic rickets. Hum Genet 89:539—542, 1992 18. ECONS MJ, FAIN PR, NORMAN M, SPEER MC, PERICAK-VANCE MA,

BECKER PA, BARKER DF, TAYLOR A, DREZNER MK: Flanking markers define the X-linked hypophosphatemic rickets gene locus. J Bone Miner Res 8:1149—1152, 1993 19. ROWE PSN, GOULDING J, READ A, MOUNTFORD R, HANAUER A, OuIwr C, WHYTE MP, MEIER-EWERT 5, LEHRACH H, DAVIES KE,

O'RIoRDAN JLH: New markers for linkage analysis of hypophosphatemic rickets. Hum Genet 91:571—575, 1993

Econs: Positional cloning of the HYP gene 20. WEBER JL, MAY PE: Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction. Am J Hum Genet 44:388—396, 1989 21. BROWNE D, BARKER D, Lirr M: Dinucleotide repeat polymorphisms at the DXS365, DXS443 and DXS451 loci. Hum Molec Genet 1:213, 1992 22. ECONS MJ, ROWE PSN, FRANCIS F, BARKER DF, SPEER MC, NORMAN M, FAIN PR, WEISSENBACH J, READ A, DAVIES KB, BECKER PA, LEIIRACII H, O'RIORDAN J, DREZNER MK: Fine structure mapping of

the human X-linked hypophosphatemic rickets gene locus. J Clin EndocrinolMetab 79:1351—1354, 1994

23. BuRKE DT, CARLE GF, OisoN MF: Cloning of large segments of exogenous DNA into yeast by means of artificial chromosome vectors. Science 236:806—812, 1987 24. FRANCIS F, ROWE PSN, ECONS MJ, SEE CG, BENHAM F, O'RIORDAN

JLH, DREZNER MK, HAMVAS RMJ, LEHRACH H: A YAC contig

1037

LAB 3: ECONS MJ, NESBITr T, AND DREZNER MK; LAB 4: OUDET C AND HANAUER A; L.AB 5: STROM T, MEINDL A, LORENZ B, CAGNOLI M, MOHNIKE KL, MURKEN J, AND MEITINGER T: Positional cloning of

PEX: A gene with homologies to endopeptidases is mutated in patients with X-linked hypophosphatemic rickets. Nature Genet 11: 130—136, 1995

29. ORITA M, SUZUKI Y, SEKIYA T, HAYASHI K: Rapid and sensitive

detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 5:874—879, 1989 30. Kos CH, TIi-iY F, ECONS MJ, MURER H, LEMIEUX N, TENENHOUSE

HS: Localization of a renal sodium phosphate cotransporter gene to human chromosome 5q35. Genomics 19:176—177, 1994 31. CHONG SS, KRISTJANSSON K, ZOGHBI HY, HUGHES MR: Molecular

cloning of the eDNA encoding a human renal sodium phosphate transport protein and its assignment to chromosome 6p2l.3-p23. Genomics 18:355—359, 1993

spanning the hypophosphatemic rickets gene candidate region.

32. TENENHOUSE HS: Effect of X-linked mutations on expression of the renal specific Na/Pi cotransporter. Kidney mt (this issue)

R, GIANNELLI F, BENTLEY DR: Construction of a 2.6Mb contig in

33. ECAROT-CHARRIER B, GLORIEUX FH, TRAVERS R, DESBARATS M,

Genomics 21:229—237, 1994 25. COFFEY AJ, ROBERTS RG, GREEN ED, COLE CG, BOIlER R, ANAND

yeast artificial chromosomes spanning the human dystrophin gene using an STS-based approach. Genomics 12:474—484, 1992 26. ECONS MJ, FRANCIS F, ROWE PSN, SPEER MC, O'RIORDAN J, LEHRACH H, BECKER PA: Dinucleotide repeat polymorphism at the DXS1683 locus. Hum Molec Genet 3:680, 1994 27. ROWE PSN, GOULDING JN, FRANCIS F, OUDET C, ECONS MJ, HANAUER A, LEHRACH H, READ AP, MOUNTFORD RC, SUMMERFIELD

T, WEISSENBACH J, FRASER W, DREZNER MK, DAVIES KE, O'RIORDAN JLH: The gene for X-linked hypophosphatemic rickets maps to a 200—300 kb region in Xp22.1-Xp22.2, and is located on a

single YAC containing a putative vitamin D response element

BOUCHARO F, HINEK A: Defective bone formation by transplanted hyp-mouse bone cells into normal mice. Endocrinology 123:768—773, 1988 34. ECAROT B, GLORIEUX FH, DESBARATS M, TRAVERS R, LABELLE L:

Defective bone formation by hyp-mouse bone cells transplanted into normal mice: Evidence in favor of an intrinsic osteoblast defect. J Bone Miner Res 7:215—220, 1992

35. Ci'.i Q, HODGSON SF, K.o PC, LENNON VA, KLEE GG, ZINSMIESTER AR, KUMAR R: Inhibition of renal phosphate transport by a tumor

product in a patient with oncogenic osteomalacia. N EngI J Med 330:1645—1649, 1994

28. THE HYP CONSORTIUM: LA.B 1: FRANCIS F, HENNIG 5, KORN B, REINHARDT R, DE JONG P, POUSTKA A, AND LEHRACH H; LAB 2:

36. ECONS MJ, DREZNER MK: Tumor-induced osteomalacia: Unveiling a new hormone, N EngI J Med 330:1679—1681, 1994 37. OLSEN HS, CEJ'EDA MA, ZHANG QQ, ROSEN CA, VOZZOLO BL,

ROWE PSN, GOULDING JN, SUMMERFIELD T, MOUNTFORD R, READ AP, POPOWSKA E, PRONICKA E, DAVIES KE, AND O'RIORDAN JLH;

WAGNER GF: Human stanniocalcin, a new hormonal regulator of mineral metabolism. Proc Nati Acad Sci USA (in press)

(VDRE). Hum Genet (in press)