Hum Genet (1992) 89:539-542

9 Springer-Verlag1992

Three D N A markers for hypophosphataemic rickets Peter S. N. Rowe 1, Andrew P. Read e, Roger Mountford 2, Frances Benham 3, Torben A. Kruse 4, Giovanna Camerino 5, Kay E. Davies 6, and Jeffrey L. H. O'Riordan 1 aDepartment of Medicine, University College and Middlesex Medical School, Middlesex Hospital, Mortimer Street, London WlN 8AA, UK 2Department of Medical Genetics, University of Manchester, Manchester M13 OJH, UK 3Galton Laboratories, Wolfson Laboratory of Molecular Genetics, University College London, London NW1 2HE, UK 4Institute of Human Genetics, University of Aarhus, Aarhus, Denmark 5Department of Genetics and Microbiology, Pavia University, Pavia, Italy 6Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK Received July 25, 1991 / Revised December 20, 1991

Summary. This p a p e r presents three markers, 16D/E, p H M A I (DXS208), and CRI-L1391 (DXS274), that show close linkage for X-linked hypophosphataemic rickets (HYP). DXS274 is closely linked to H Y P (0mo~= 0.00, Zmax = 4.20), and DXS41 (99.6), (Omax= 0.00, Zma~ = 5.20). Marker 16D/E maps distal to the disease locus (Omax = 0.05, Zmax = 3.11). The p H M A I p r o b e recognises the same restriction fragment length polymorphism (RFLP) as 99.6. Multipoint analysis suggests that the most probable order of loci is Xpter-(DXS43, 16D/E)-HYP-DXS274(DXS208, DXS41)-Xcen. The location of DXS274 distal to H Y P cannot be excluded, as no recombinants were observed between DXS274 and H Y P , or between DXS274 and DXS41/DXS208. One of the families contains a large n u m b e r of recombinants, four of which are double recombinants. This most probably means that the disease in this family maps elsewhere on the X c h r o m o s o m e or on an autosome, indicating locus heterogeneity.

Introduction T h e congential disease X-linked dominant hypophosphataemic (vitamin D resistant) rickets (HYP, McKusick No. 307800, catalogue of 1990) is a commonly inherited disorder due to defective phosphate handling in the kidney. The incidence in the United Kingdom has been estim a t e d to be 1 in 20-25 000 (Davies and Stanbury 1981). Biochemical and physiological experiments have not shown conclusively whether the lesion is due to a primary defect in a sodium-dependent phosphate co-transporter in the kidney, or whether a circulating humoral factor is involved (Tenenhouse et al. 1978; Cowgill et al. 1979; Meyer et al. 1989). In an attempt to resolve this question we have used the reverse genetics approach with the aim of cloning the gene or genes responsible, and have re-

Correspondence to: P. S. N. Rowe

ported together with others the localisation of two markers, DXS41 and DXS43, flanking the H Y P locus in humans (Read et al. 1986; Machler et al. 1986; T h a k k e r et al. 1987). In addition we have also shown tight linkage of the markers DXS197 and DXS207 to the DXS43 locus (Thakker et al. 1990). In order to extend the m a r k e r f r a m e w o r k around the H Y P locus, and to narrow the gap between distal and flanking markers we used three additional markers (DXS274, DXS208, and 16DfE), thought to be localised to the region of the X chromosome containing the H Y P locus (Xp22.31 to Xp21.3), for linkage and multilocus analysis on our affected families.

Materials and methods Families Fifteen affected families were investigated in which X-linked hypophosphataemia had been inherited through two or more generations. A total of 158 individuals were studied, with 79 affected (50 females, 29 males), and 79 unaffected (23 females, and 56 males). These families and the basis for clinical diagnosis have been described in previous papers (Read et al. 1986; Thakker et al. 1987, 1990).

Southern blot analysis of families Extraction of DNA from venous blood, digestion with appropriate restriction enzymes, Southern blotting, hybridisation and autoradiography were as described previously (Read et al. 1986), except that Hybond N+ (Amersham) was used as the immobilised nylon membrane instead of Hybond N. In experiments using the polymorphic probe DXS208, the presence of repeat elements necessitated prehybridisation of blots and probe with human sonicated and denatured DNA.

Linkage analysis Three new markers were used in the linkage study, DXS208 (priMA1), 16D/E, and DXS274 (CRI-L1391). Marker DXS208

540 Table 1. Two point lod scores between markers Locus

vs

Omax

locus

Zmax

Lod score at recombination fractions 0,00

DXS43

DXS208 a DXS41 DXS274 16D/E DXS274 DXS208 a DXS41 DXS208 a DXS41

DXS43 DXS43 16D/E 16D/E DXS274

a

0.01

0.05

0.10

0.15

0,20

0.30

0.40

0.15

2.17

-w

-0.95

1.40

2.05

2.17

2.08

1.56

0.82

0.17 0.00 0.00 0.13

1.33 3.91 2.70 1.63

-~ 3.91 2.70 -2

-1.22 3.84 2.66 0.20

0.60 3.56 2.46 1.34

1.16 3.18 2.21 1.61

1.32 2.79 1.95 1.62

1.31 2.37 1.67 1.5l

1.04 1.49 1.08 1.11

0.57 0.60 0.45 0.59

0.00

5.20

5.20

5.11

4.78

4.35

3.90

3.40

2.40

1.20

Pooled DXS208 and DXS41 data

Table 2. Two point linkage data without family A Locus

DXS43 16D/E DXS274 DXS208 a DXS41 a

vs

locus

HYP HYP HYP HYP

Omax 0.14 0.05 0.00 0.04

Zm..~ 4.49 3.11 4.22 8.27

Lod score at recombination fractions 0.00

0.01

0.05

0.10

0.15

0.20

0.30

0.40

-2 -2 4.22 -~

-0.51 2.73 4.15 7.69

3.34 3.11 3.87 8.25

4.35 2.99 3.50 7.78

4.48 2.73 3.10 7.01

4.21 2.40 2.70 6.10

3.07 1.62 1.84 4.03

1.52 0.77 0.94 1.84

Pooled DXS208 and DXS41 data

(priMA1) detects a restriction fragment length polymorphism (RFLP) after digestion with PstI of 23, and 13 kb in size, respectively (Mandel et al. 1989). Probe 16D/E (G.C., unpublished results) detects an X chromosome specific RFLP after digestion of genomic DNA with PvuII (3.3, and 3 kb). Probe DXS274 (CRIL1391), was obtained from Collaborative Research (Bedford, Mass.), and detects RFLP DNA fragments of 11 and 8.5 kb after MspI digestion of genomic DNA (Donis-Keller et al. 1987). Data derived from previous analyses for DXS41, DXS43, DXS207, and DXS197 were pooled for the linkage and multilocus analyses with the new probes (Read et al. 1986; Thakker et al. 1987, 1990). For computing the linkage data we used version 5.03 of the Mlink and Linkmap programs of Lathrop and Lalouel (1984), with Haldane's mapping function. For the disease we assumed a gene frequency of 0.0001 and penetrances of 1.0 in males and 0.99 in heterozygous females. With Linkmap the marker framework used was (DXS43, 16D/E)-13 cM-DXS274-2cM-DXS41/DXS208. This was deduced from a consideration of our previously published gene markers framework (Thakker et al. 1987), and also from recent linkage studies conducted on retionschisis (RS) and CEPH families (Alitalo et al. 1991a, b).

DX: 274 DXS208 DXS41

6050-

DXS43 16D/E

1012 1010 10 8 10 6 o

40-

o 30~ = 200 ".,= 100-

10 4

10 2 -~ 1

0

10-2

" -10-20-30-

10 -4

10-6 I

-0.4

I

I

I

/

I

I

-0.2 0 O.2 0.4 Genetic distance, W (Morgans)

Fig.1. Graphical representation of five-point multilocus analysis with family A excluded. The hypophosphataemic rickets (HYP) locus was moved stepwise across the fixed framework of markers shown, and the likelihood of the pedigree data computed. The location score I(W) is twice the natural logarithm of the odds ratio. The peak indicates the most favoured gene marker order

Results T h e p r o b e p H M A I (DXS208) was f o u n d to identify the same p o l y m o r p h i s m as p r o b e DXS41 (99.6). Both probes hybridised to 22 a n d / o r 13-kb f r a g m e n t s in PstI-digested D N A a n d all individuals tested gave the same g e n o t y p e s with b o t h probes. This agrees with a linkage study of D a h l et al. (1988), o n families with RS. M o l e c u l a r studies (T. Kruse, u n p u b l i s h e d data), also suggest that the two p r o b e s m a p within 20 kb of each other. D a t a from these two p r o b e s are t h e r e f o r e c o m b i n e d in all analyses below.

T w o - p o i n t lod scores b e t w e e n pairs of m a r k e r s are s h o w n in T a b l e 1. P r o b e CRI-L1391 (DXS274) gives rec o m b i n a n t s with DXS43 b u t n o t with D X S 2 0 8 / D X S 4 1 . T h e p r o b e 16D/E o n the o t h e r h a n d shows n o r e c o m b i n ants with DXS43 or DXS274 ( m a x i m u m lod scores 3.91 a n d 2.70 respectively at zero r e c o m b i n a t i o n ) , b u t does show r e c o m b i n a n t s with D X S 2 0 8 / D X S 4 1 . W h e n r e c o m b i n a n t s b e t w e e n the m a r k e r s and H Y P were e x a m i n e d a curious p a t t e r n was o b s e r v e d in o n e family (family A).

541 Recombinants appeared to be clustered in this pedigree, and four individuals were double recombinants. This raises some doubt about the interpretation of family A. All three markers show lod scores of > 3 against H Y P , although as expected, m a x i m u m lod scores occurred at lower recombination fractions when family A was omitted (Table 2). However, regardless of whether or not family A is included in the analysis, multipoint mapping (Fig. 1) gives the most likely gene order as Xpter-(DXS43, 16D/E)-HYP-DXS274-(DXS208, DXS41)-Xcen. Although DXS274 is tightly linked to H Y P its probable location proximal to the disease locus is not completely unequivocal, due to the lack of recombinants between this marker and H Y P , and DXS41/DXS208.

Discussion The three new markers studied, DXS208, 16D/E and DXS274, are shown to be closely linked to the H Y P locus, with DXS208 indistinquishable from the DXS41 locus genotypically. Several markers have been previously identified as distal to the disease and closely linked, and include DXS43 (Read et al. 1986; Machler et al. 1986; T h a k k e r et al. 1987), DXS197 and DXS207 (Thakker et al. 1990), and G L R , the glycine receptor (Econs et al. 1990). Multilocus analysis of this group of markers has not produced evidence for their order relative to one another. The linkage analysis data presented here adds 16D/E to this cluster of distal probes. DXS274 emerges as a very useful marker. In our data (excluding family A) there were no recombinants between DXS274 and H Y P . There were also no recombinants between DXS274 and DXS41/DXS208; however, Alitalo et al. (1991a) place DXS274, 3 . 5 c M distal of DXS41. It is therefore likely that of the markers we have studied to date DXS274 is the closest proximal m a r k e r to H Y P , although our data do not allow an unequivocal proximal, or distal location for this marker. A recent study by Barker et al. (1990) has shown that an additional marker, DXS365, also shows tight linkage to H Y P with no recombinants apparent between the two loci, and this is clearly another very useful marker. In analysing the data presented here, consideration needs to be taken of one of the families studied, family A. This large four-generation family was included in all our earlier publications ( R e a d et al. 1986; T h a k k e r et al. 1987, 1990). Four individuals were double recombinants, with recombination between H Y P and both distal and proximal markers. Pascoe and Morton (1987) have argued that such close double recombinants are highly unlikely. One possible explanation is locus heterogeneity, which is one of the m o r e intractable problems in h u m a n linkage analysis. However, excluding family A from the multilocus analysis did not affect the likely gene order, but it did condense the map. If we are justified in treating family A as different, then the existing panel of markers has a lower recombination error for gene mapping than previously supposed, and offers m o r e hope of finding the H Y P gene by positional cloning.

This p a p e r presents a further refinement of the genetic m a p around the H Y P locus. Additional m a r k e r loci have been defined with their distance and order relative to H Y P and to the two flanking markers DXS41 and DXS43. The following gene m a r k e r order is proposed as the most favourable from multilocus analysis, with distances expressed as recombination fractions: 0.089 0.019 0.019 r I r q i q Xpter-(DXS43,16D/E)-HYP-DXS274-DXS208,DXS41)-Xcen These markers will help increase the resolution of the genetic map from this region, and will be of great value in future strategies aimed at cloning the gene(s) causing the abnormality in phosphate handling in X-linked HYP. F r o m our data and the other data discussed, we are able to propose the following combined gene m a r k e r order: X p t e r - ( R S , DXS207, DXS197, DXS43, G L R , 1 6 D / E ) ( H Y P , DXS274, D X S 3 6 5 ) - D X S 2 0 8 / D X S 4 1 - X c e n .

Acknowledgements. We would like to thank all the physicians who generously allowed us access to their patients. This study was funded by the Medical Research Council of the United Kingdom.

References Alitalo T, Kruse TA, Ahrens P, Albertsen HM, Chapelle A de la (1991a) Genetic mapping of 12 marker loci in the Xp22.3p21.3 region. Hum Genet 86:599-603 Alitalo T, Kruse TA, Chapelle A de la (1991b) Refined localisation of the gene causing X-linked retinoschisis. Genomics 9: 505-510 Barker DF, Speer MC, Pericak-Vance MA, Fain P, Drezner MK (1990) Multilocus mapping of the human X-linked hypophosphataemic rickets gene locus (abstract). J Bone Miner Res 5[Suppl 2] :$108 Cowgill LD, Goldfarlb S, Lau K, Slatopolsky E, Agus ES (1979) Evidence for an intrinsic renal tubular defect in mice with genetic hypophosphataemic rickets. J Clin Invest 63 : 1203-1210 Dahl N, Goonewardena P, Chotai J, Anvret M, Petterson U (1988) DNA linkage analysis of X-linked retinoschisis. Hum Genet 78 : 228-232 Davies M, Stanbury SW (1981) The rheumatic manifestations of metabolic bone disease. Clin Rheum Dis 7: 595-646 Donis-Keller H, Green P, Helms C, Cartinhour S, Weiffenbach B, Stephens K, Keith TP, Bowden DW, Smith DR, Lander ES, Botstein D, Akots G, Rediker KS, Gravius T, Brown VA, Rising MB, Parker C, Powers JA, Watt DE, Kaufmann ER, Bricker A, Phipps P, Muller-Kahle H, Fulton TR, Ng S, Schumm JW, Braman JC, Knowlton RG, Barker DF, Crooks SM, Lincoln SE, Daly MJ, Abrahamson J (1987) A genetic linkage map of the human genome. Cell 51 : 319-337 Econs MJ, Pericak-Vance MA, Betz H, Bartlett RJ, Speer MC, Drezner MK (1990) The human glycine receptor: a new probe that is linked to the hypophosphataemic rickets gene. Genomics 7 : 439-441 Lathrop GM, Lalouel JM (1984) Easy calculation of lod scores and genetic risks on small computers. Am J Hum Genet 36:460465 Machler M, Frey D, Gal D, Orth U, Weinker TF, Fanconi A, Schmid W (1986) X-linked dominant hypophosphataemia is closely linked to DNA markers DXS41 and DXS43 at Xp22. Hum Genet 73 : 271-275 Mandel JL, Willard HF, Nussbaum RL, Romeo G, Puch JM, Davies KE (1989) X-chromosome reports. Cytogenet Cell Genet 51 : 384-437 McKusick VA (1990) Mendelian inheritance in man, 9th edn, Johns Hopkins University Press, Baltimore

542 Meyer RA Jr, Meyer MH, Gray RW (1989) Parabiosis suggests a humoral factor is involved in X-linked hypophosphataemia in mice. J Bone Miner Res 4 : 493-500 Pascoe L, Morton NE (1987) The use of map functions in multipoint mapping. Am J Hum Genet 40 : 174-183 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 (1986) Mapping of human X-linked hypophosphataemic rickets by multilocus linkage analysis. Hum Genet 73 : 267-270 Tenenhouse HS, Scriver CR, McInnes RR, Glorieux FH (1978) Renal handling of phosphate in vivo and in vitro by the X-

linked hypophosphatemic male mouse; evidence for a defect in the brush border membrane. Kidney Int 14 : 236-244 Thakker RV, Read AP, Davies KE, Whyte MP, Weksberg R, Glorieux F, Davies M, Mountford R, Harris R, King A, Kim GS, Fraser D, Kooh SW, O'Riordan JLH (1987) Bridging markers defining the map position of X linked hypophosphataemic rickets. J Med Genet 24 : 756-760 Thakker RV, Davies KE, Read AP, Tippet P, Wooding C, Flint T, Woods S, Kruse TA, Whyte MP, O'Riordan JLH (1990) Linkage analysis of two cloned DNA sequences, DXS197 and DXS207, in hypophosphataemic rickets families. Genomics 8 : 189-193