Assignment of a second locus for multiple exostoses ...

2 downloads 0 Views 431KB Size Report
Hereditary multiple exostoses (EXT) is an autosomal dominant disorder of enchondral bone formation characterized by multiple bony outgrowths (exostoses),.
© 1994 Oxford University Press

Human Molecular Genetics, 1994, Vol. 3, No. 1

167-171

Assignment of a second locus for multiple exostoses to the pericentromeric region of chromosome 11 Yuan-Qing WuL*, Peter Heutinki, Bert B.A.de Vriesi, Lodewijk A.SandkuijM, Ans M.W.van den Ouweland', Martinus F.Niermeijeri, Hans GaljaanP, Edwin Reyniers3, Patrick J.Willems3 and Dicky J.J.HalleyL* 1 Department of Clinical Genetics, Erasmus University and University Hospital, Dr. Molewaterplein 50, 3015 GE Rotterdam, The Netherlands, 2Beijing Union Medical College Hospital, Beijing 100730, China and 3Department of Medical Genetics, University of Antwerp-UIA, Universiteitsplein 1, 2610 Antwerp, Belgium

Received October 4, 1993; Revised and Accepted November 17, 1993

INTRODUCTION Hereditary multiple exostoses (EXT) is an autosomal dominant disorder characterized by the presence of multiple exostoses most commonly arising from the juxtaepiphyseal region of the long bones1 ~6. Other bones can be involved including the pectoral and pelvic girdles, ribs and, less frequently, vertebrae, sternum, skull, carpal and tarsal bones2'4-6-. An additional feature of EXT is abnormal bone modelling, particularly of the long bones. This causes bowing, shortness, cortical irregularities and metaphyseal widening of the long bones, leading to striking deformities of the forearms and disproportionate short stature in severe cases1-4-6. The exostoses can also give rise to various complications, such as compression and irritation of adjacent

T o whom correspondence should be addressed

nerves, vessels, and tendons, and urinary or intestinal obstruction. The most serious complication is sarcomatous degeneration, which occurs in 0.5% to 2% of affected individuals (reviewed by Hennekam6). The prevalence of EXT is not precisely known, but is estimated at 1 in lOO.OOO6-7. Whereas EXT constitutes a separate clinical entity, multiple exostoses can also be part of other disorders, such as the Langer-Giedion syndrome (LGS), also known as tricho-rhinophalangeal syndrome type II (TRPII) 8 . Cone-shaped epiphyses, unusual facies, sparse hair and, frequently, mental retardation are the additional characteristics of LGS. Tricho-rhino-phalangeal syndrome type I (TRP I) shows phenotypic overlap with LGS (TRP II), but mental retardation is rare and exostoses are absent9. As the TRP syndromes are associated with deletions at the long arm of chromosome 8 (8q23-24) 9 ~ 13 , Buhler and Malik9 suggested that an EXT gene may be located in the same region. This was supported by the observation of a patient with multiple exostoses without additional symptoms and a balanced reciprocal translocation between 8q24.1 and llplS.S 14 . Initially, linkage analysis in 6 EXT families using probes from an 8q24.1 microdissection library of the LGS chromosomal region could not reveal an 8q24-linked EXT locus13. In contrast, Cook et al.1 recently reported evidence for linkage of EXT to markers from 8q24 in a set of 11 pedigrees, with 8 families showing a high posterior probability of segregating a chromosome 8-linked defect. Thus, locus heterogeneity for EXT is apparent with a major locus in the LGS region on chromosome 8q24 and at least one locus elsewhere in the genome. We investigated two large EXT pedigrees including 32 available affected individuals and spanning five and four generations, respectively. Linkage to markers from 8q24 was excluded and a genome search was undertaken to localize the gene defect in these families. We present evidence for an EXT locus that maps to the pericentromeric region of chromosome 11 in both families. RESULTS AND DISCUSSION Two multigenerational families with EXT Two large families with EXT were referred to us. EXT family 1 is of Dutch origin. DNA from 32 individuals including 17 affected family members from 4 generations was available for this study. EXT family 2 originates from Belgium and included IS available affected individuals from 3 generations. Pedigrees are shown in figure 1. A common ancestor of the families could not be identified. Diagnoses were based on physical and

Downloaded from http://hmg.oxfordjournals.org/ at Erasmus Universiteit Rotterdam on April 20, 2016

Hereditary multiple exostoses (EXT) is an autosomal dominant disorder of enchondral bone formation characterized by multiple bony outgrowths (exostoses), with progression to osteosarcoma in a minority of cases. The exclusive involvement of skeletal abnormalities distinguishes EXT from the clinically more complex Langer-Giedion syndrome (LGS), which is associated with deletions at chromosome 8q24. Previously, linkage analysis has revealed a locus for EXT in the LGS region on chromosome 8q24. However, locus heterogeneity was apparent with 30% of the families being unlinked to 8q24. We report on two large pedigrees segregating EXT in which linkage to the LGS region was excluded. To localize the EXT gene(s) in these families we performed a genome search including 254 microsatellite markers dispersed over all autosomes and the X chromosome. In both families evidence was obtained for linkage to markers from the proximal short and long arms of chromosome 11. Two-point analysis gave the highest lod score for D11S554 (Zmax = 7.148 at theta = 0.03). Multipoint analysis indicated a map position for the EXT gene between D11S905 and D11S916, with a peak multipoint lod score of 8.10 at 6 cM from D11S935. The assignment of a second locus for EXT to the pericentromeric region of chromosome 11 implicates an area that is particularly rich in genes responsible for developmental abnormalities and neoplasia.

168 Human Molecular Genetics, 1994, Vol. 3, No. 1

A 9

3 • 7

4 4 3 7





1

2 2 3 2 4

4 4 3 7

1 7 4

4 4 3 7

d 2 3 2

2 3 2 4

2 5 3

1 9 3

S

2

• 2 5 3

2 2 3 9 2 3

ii

i

* - li—o4

3 9 2 T 4

10

5 5 2 4 2

2

2

4

2

• 2

3 2

4

2

2 3 2 7 2

A

IS

6 3 2 2 4 7 4 2

A 2

3

3

2

Figure 1. EXT pedigrees (A: family 1; B: family 2) and haplotypes of chromosome 11 markers. Affected individuals are indicated by filled symbols. Markers were ordered according to their chromosomal localisation: pter-Dl 1S935-D11S905-D11S9O3-D11SSS4-D1 lS916-qter. Key recombination events in affected individuals between the EXT locus and chromosome 11 markers are indicated by an x. An asterisk indicates that the exact localisation of the recombination event could not be determined due to uninformativeness of die markers Dl 1S905 and D11S903 in this branch of the family.

radiological examination, as specified in Materials and Methods. One female without clinical evidence of exostoses had affected offspring. Exclusion of linkage to chromosome 8 Both families were analyzed with microsatellite markers representing the loci D8S85, and D8S199/D8S198, which have been reported7 to flank the EXT region of chromosome 8q24. Cook et al.1 deduced the most probable order centromere-

D8S85-EXT-D8S199-D8S198-telomere, when conducting multipoint analysis under the assumption of heterogeneity with 70% of families linked to chromosome 8q24. The two-point lod scores generated from our two families were strongly negative for the loci D8S8S, D8S199 and D8S198 for values of the recombination frequency up to 10% (Table 1). This excludes the EXT locus in our families from the 8q24 region. Previously, Le Merrer et al.15 failed to demonstrate linkage to that region in their families. This might be explained by linkage heterogeneity

Downloaded from http://hmg.oxfordjournals.org/ at Erasmus Universiteit Rotterdam on April 20, 2016

B

Human Molecular Genetics, 1994, Vol. 3, No. 1 169 Table 1. Pairwise lod scores between EXT and chromosome 8 markers' Locus D8S85 D8S199

Family 1 2 Total 1

2 D8S198

Total 1 2 Total

0.000

0.010

-9.731 -8.626 -18.357 -15.333 -4.967 -20.300 -4.174 -8.970 -13.144

-6.061 -5.978 -12.039 -8.431 -2.971 -11.402 -2.535 -5.118 -7.653

0.050 -3.744 -3.989 -7.733 -4.289 -1.641 -5.930 -1.633 -2.596 -4.229

Recombination fraction 0.100 0.200 -2.542 -3.021 -5.563 -2.542 -0.779 -3.321 -1.145 -1.516 -2.661

-1.249 -1.630 -2.879 -0.998 -0.132 -1.130 -0.605 -0.576 -1.181

0.300

0.400

-0.554 -0.763 -1.317 -0.324 0.034 -0.290 -0.296 -0.182 -0.478

-0.169 -0.265 -0.434 -0.035 0.029 -0.006 -0.107 -0.030 -0.137

'Markers are ordered from 8pter-qter. Table 2. Pairwise lod scores between EXT and chromosome 11 markers'

D11S914 D11S907 D11S935 DUS905 D11S903 D11S554 D11S871 D11S913 D11S916 D11S937

Family 1 2 Total 1 2 Total 1 2 Total 1 2 Total 1 2 Total 1 2 Total 1 2 Total 1 2 Total 1 2 Total 1 2 Total

0.00

0.01

-7.131 -3.017 -10.148 -4.526 -4.859 -9.385 -2.013 -3.227 -5.240 1.341 0.972 2.313 3.004 0.876 3.879 5.344 -0.442 4.902 4.123 -1.919 2.204 -6.314 -0.633 -6.947 -12.029 -1.106 -13.136 -7.146 -4.625 -11.771

-4.190 -2.068 -6.258 -2.576 -3.810 -6.386 -1.003 -1.990 -2.993 2.260 1.889 4.149 2.950 0.860 3.810 5.284 1.641 6.924 4.059 -0.021 4.038 -4.515 0.902 -3.613 -6.757 0.436 -6.321 -4.389 -1.860 -6.248

Recombination traction (0) 0.05 0.10 -2.219 -1.503 -3.722 -1.294 -2.829 -4.124 -0.276 -0.580 -0.856 2.646 2.289 4.935 2.730 0.794 3.525 5.007 2.095 7.102 3.798 1.061 4.859 -2.511 1.416 -1.095 -2.936 0.997 -2.527 -1.848 -0.114 -1.962

-1.350 -1.059 -2.409 -0.751 -2.014 -2.765 0.048 0.013 0.061 2.581 2.245 4.827 2.443 0.710 3.153 4.599 2.080 6.680 3.457 1.343 4.800 -1.587 1.484 -0.103 -1.398 1.121 -0.277 -0.795 0.500 -0.294

0.20

0.30

-0.532 -0.339 -0.871 -0.260 -0.923 -1.183 0.283 0.328 0.611 2.099 1.820 3.919 1.823 0.534 2.356 3.629 1.705 5.334 2.714 1.280 3.994 -0.703 1.282 0.579 -0.159 1.018 0.859 0.037 0.799 0.835

-0.149 -0.062 -0.211 -0.041 -0.353 -0.394 0.273 0.265 0.539 1.407 1.210 2.617 1.142 0.353 1.495 2.477 1.142 3.620 1.868 0.909 2.777 -0.267 0.893 0.626 0.255 0.719 0.974 0.280 0.658 0.938

Zmax

e

0.024 0.000 0.024 0.036 0.000 0.014 0.300 0.334 0.631 2.653 2.300 4.952 3.004 0.876 3.879 5.344 2.117 7.148 4.123 1.377 4.889 0.001 1.486 0.657 0.302 1.125 1.005 0.292 0.799 0.917

0.44 0.44 0.44 0.42 0.50 0.46 0.24 0.22 0.23 0.06 0.06 0.06 0.00 0.00 0.00 0.00 0.07 0.03 0.00 0.13 0.07 0.49 0.09 0.26 0.35 0.11 0.27 0.33 0.20 0.31

'Markers are ordered from llpter—qter

or by the limited sizes of their pedigrees combined with a lack of informativeness of their markers and the relatively large distance between the markers and the LGS region. Our results provide further support for the existence of more than one EXT locus. Genome search with microsatellite markers In order to map the EXT locus (or loci) segregating in families 1 and 2 we designed a genome search with microsatellite markers dispersed over all autosomes and the X-chromosome. When lod scores reached positivity over plus 1 in two-point linkage analyses, flanking markers were tested. In total, 254 markers were included in this study. None of the chromosomes yielded an area with indication for linkage, with the exception of chromosome 11.

Two-point analysis with markers from chromosome 11 Two-point lod scores between EXT and 24 chromosome 11 markers were calculated for each of the families at various recombination frequencies. A number of markers from the pericentromeric region showed positive lod scores (Table 2). No recombinants were observed between EXT and D11S903, but this marker was not very informative in our families. Dl 1SSS4 generated the highest lod scores in both families, with I^a, = S.344 at theta = 0 in family 1 and Z , ^ = 2.164 at theta = 0.07 in family 2. Thus, a combined Z , ^ of 7.148 was obtained at theta = 0.03 relative to D11S554, which has been mapped to the pericentromeric region of chromosome II 1 6 " 1 9 . The relative position of markers derived from the various maps that have been constructed is not always clear. The order of the markers given in Table 2 was decided upon after comparison

Downloaded from http://hmg.oxfordjournals.org/ at Erasmus Universiteit Rotterdam on April 20, 2016

Locus

170 Human Molecular Genetics, 1994, Vol. 3, No. 1

£ o

8

2

5

-30

-20

-10

0

10

20

30

40

CM from D11S935 Figure 2. Multipoint analysis for EXT versus five markers of chromosome 11. Combined analysis of families 1 and 2. The bold line indicates the 'standard'multipoint analysis, the thin line indicates the 'affected only' analysis. Relative positions of the markers are indicated by abbreviations of their D numbers (e.g. Dl 1S935 as 935). Recombination fractions were converted to centimorgans using Kosambi's map function.

of their genetic distances from markers with well established positions that appear in all or most maps. The order of the markers used for the multipoint analysis was almost exclusively based on the Genethon map (see Materials and Methods). The map position of EXT: multipoint analysis and haplotyping with markers from the pericentromeric region of chromosome 11 Since two-point linkage analysis provided evidence for linkage of an EXT locus to markers surrounding the centromere of chromosome 11, multipoint analysis was performed with the most informative markers to determine the most likely location of the EXT locus. The five markers selected for multipoint analysis were D11S935, D11S905, D11S554, D11S916, and D11S937, with male recombination frequencies of 0.040, 0.027,0.076, and 0.022 in the respective intervals. Family 1 yielded a maximum multipoint lod score of 5.610 at the marker locus Dl 1S554. The peak lod score in family 2 was 2.666 at a male genetic distance of 5.6 cM from Dl 1S935, between the positions of the marker loci Dl 1S905 and Dl 1S554. The combined maximum multipoint lod score (figure 2) equalled 8.100 at a male genetic distance of 6 cM from D11S935, in close proximity of the suggested location of marker D11S554. The strength of evidence favouring linkage may depend on the degree of penetrance. In our pedigrees nearly complete penetrance was observed. In order to eliminate the effects of possible missed diagnoses in apparently unaffected individuals, multipoint analysis was also conducted with the disease phenotype for all unaffected family members classified as unknown (figure 2). As expected, some information was lost and the peak lod score reached a lower value (6.58), but the overall pattern remained identical. To further delineate the area the approximate 95% confidence interval was constructed including all locations with a lod score exceeding Zmax — 1 in the 'affecteds only'analysis. Boundaries were on either side of D11S554, from 2.2 cM towards D11S9O3 to 2.3

The exostoses in EXT patients result from abnormal enchondral bone formation and they appear preferentially in regions of rapid bone growth. Although infrequent, malignant degeneration is a well known complication. Thus, the function of a gene involved in EXT is of relevance both for bone morphogenesis and carcinogenesis. The pericentromeric region of chromosome 11 holds a multitude of chromosomal breakpoints21 and genes22 implicated in development and neoplasia. Further studies aimed at refining the regional localization of the EXT gene on chromosome 11 are in progress and will direct the search for the chromosome 11-linked EXT locus by means of positional cloning as well as the search for candidate genes. MATERIALS AND METHODS Families and DNA Each of the families was ascertained and examined by one of the authors (BBA de V, family 1 and PJW, family 2). Two or more characteristic exostoses detected at physical examination were considered indicative for EXT. All diagnoses were radiographicaUy confirmed, except in some older patients who had both exostoses at physical examination and affected offspring, and in one female whose diagnosis was confirmed through medical records after surgical treatment. Children with obvious clinical features of EXT were included in the study only after informed consent by the parents. Apparently unaffected children below IS years of age were excluded from the study because of the lower penetrance of the disease at younger age. Cytogenetic analysis, performed in one patient from family 1, showed a normal karyotype. Blood samples were collected from 59 individuals and DNA was isolated as described by Miller et al.23. Microsatellite markers Microsatellite markers were tested in multiplex reactions essentially as decribed previously24. Eight markers from chromosome 8 were analyzed including DSS85, D8S199 and D8S198, which flank the LGS region on chromosome 8q247>25. Five markers were from the X-chromosome. The remaining markers (n = 241) were evenly distributed over all autosomes. Chromosome 11 was covered from the terminal short arm to the distal part of the long arm by 24 regionally mapped markers 1626 - 27 .

Downloaded from http://hmg.oxfordjournals.org/ at Erasmus Universiteit Rotterdam on April 20, 2016

-10 -40

cM towards D11S913. This indicated an area of 4.5 cM surrounding the D11S554 locus as the most likely position for an EXT locus on chromosome 11. Haplotype analysis (figure 1) showed recombinants in both families for Dl 1S905 and Dl 1S916, which flank Dl 1S554. One affected individual was identified (II. 7 in family 2) with an apparently recombinant allele for Dl 1S554, but without parental marker information. At present, this precludes the assignment of the chromosome 11 EXT locus in these families to either side of the D11S554 marker locus. Furthermore, the exact position of D11S554 with respect to the centromere is not yet resolved. FISH16 and somatic cell hybrid17 studies suggest a localization on the short arm in p l l —pl2. However, Dl 1S554 must also be located in close proximity to D11S28818, which might map to the proximal long arm of chromosome II 19 . Dl 1S554 was the only marker in our multipoint analysis for which the location could not be directly obtained from the existing Genethon map. Based on a separate report20, and based on evidence from our own data, we assumed a position of this marker between Dl 1S905 and D11S916. The alternative order D11S905-D11S916D11S554 was firmly rejected in our data (odds against of 7X10 5 ), but there was only little evidence in our own data against the possible order Dl 1S554-D11S905 - D l 1S916 (odds against 2.6 to 1). Further refinement of the EXT location in our families awaits a more accurate localization of the marker D11S554 with respect to surrounding markers from the pericentric region.

Human Molecular Genetics, 1994, Vol. 3, No. 1 171

ACKNOWLEDGEMENTS We thank Prof Nianhu Sun (Beijing) for her support and encouragement. During her stay in Rotterdam Y.-Q.W. worked as a fellow in the Rotterdam Foundation of Clinical Genetics. We are indebted to Drs. W.de Smet and E.J.Meijers-Heijboer for referral of family 1, Drs. J.Dumon and P.De Jonghe for referral of family 2 and to Dr. J.O.van Hemel for cytogenetic analysis of a patient from family 1. Primers were obtained through a grant from the Netherlands Organization for Scientific Research. Part of this research was funded by a grant (Concerted Action 1993-1997) of the University of Antwerp to P.J.W. We thank the MGC for financial support.

REFERENCES 1. 2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

Solomon,L. (1961) J. Bone Joint Surg.(Br) 43, 700-716. Solomon.L. (1963) / . Bone Joint Surg.(Br) 45B, 292-304. Solomon.L. (1964) Am. J. Hum. Genet. 16, 351-363. Shapiro,F., Simon,S. and Glimcher.M.J. (1979)7. Bone Joint Surg. (Am) 61A, 815-824. Jaffe,H.L. (1943) Arch. Pathol. 36, 335-357. Hennekam.R.C.M. (1991) J. Med. Genet. 28, 262-266. Cook.A., Raskind.W., Blanton.S.H., Pauli,R.M., Gregg,R.G., Francomano./C.A., Puffenberger.E., Conrad,E.U., Schmale.G., ScheUenberg,G., Wijsman.E., HechtJ.T., Wells.D. and Wagner,M.J. (1993) Am. J. Hum. Genet. 53, 7 1 - 7 9 . Langer.L.O., Krassikoff.N., Laxova.R., Scheer-WUliams.M., Lutter, L.D., Gorlin.R.J., Jennings,C.G. and Day.D.W. (1984) Am. J. Med. Genet. 19, 81-111. BOhler.E.M. and Malik.N.J. (1984) Am. J. Med. Genet. 19, 113-119. Fryns,J.P. and Van den Berghe.H. (1986) Hum. Genet. 74, 188-189. Liidecke,H.-J., Johnson.C, Wagner.M.J., WeUs.D.E., Turleau,C, Tommerup.N., Latos-Bilenska,A., Sandig,K.-R., Meinecke,P., Zabel.B. and Horsthemke,B. (1991) Am. J. Hum. Genet. 49, 1197-1206. Yamamoto.Y., Oguro.N., Miyao.M. and Yanagisawa.M. (1989)/1m. /. Med. Genet. 32, 133-135. Biihler.E.M., Biihler.U.K., Beutler.C. and Fessler.R. (1987) Clin. Genet. 31, 273-275. Ogle,R.F., Dalzell.P., Turner.G., Wass.D. and Yip,M.-Y. (1991) / Med. Genet. 28,881-883. Le Merrer.M., Othmane.K.B., Stanescu.V., Lyonnet.S., Van Maldergem.L., Royer.G., Munnich.A. and Maroteaux.P. (1992) J. Med. Genet. 29, 713-715. Hori,T., Takahashi.E.-I., Tanigami.A., Tokino.T. and Nakamura.Y. (1992) Genomes 13, 129-133. Tanigami.A., Tokino.T., Takiguchi.S., Mori.M., Glaser.T., ParkJ.W., Jones.C. and Nakamura.Y. (1992) Am. J. Hum. Genet. 50, 56-64. Phromchotikul.T., Browne.D. and Lin.M. (1992) Hum. Molec. Genet. 1, 214. Julier.C., Nakamura.Y., Lathrop.M., O'Connell.P., Leppert.M., Litt.M., Mohandas.T., Lalouel,J.-M. and White.R. (1990) Genomics 7, 335-345. Chromosome 11 Workshop, San Diego (1992).

21. Lammie.G.A. and Peters.G. (1992) Cancer Cells 3, 413-420. 22. Junien,C. and van Heyningen, V. (1991) Cytogenet. Cell Genet. 58, 459-554. 23. Miller.S.A., Dykes.D.D. and Polesky.H.F. (1988) Nucleic Adds Res. 16, 1215. 24. Weber.J.L. and May.P.E. (1989) Am. J. Hum. Genet. 44, 388-396. 25. TomfohrdeJ., Wood.S., Schenzer.M., Wells.D.E., ParrishJ., Sadler, L.A., Blanton,S.H., Daiger.S.P., Wang.Z., Wilkie,P.J. and Weber,J.L. (1992) Genomes 14, 144-152. 26. Weissenbach,J., Gyapay.G., Dib.C, Vignal.A., Morisette,J., Milasseau P., Vaysseix,G. and Lathrop.M. (1992) Nature 359, 794-801. 27. Litt,M., Kramer,P., Hauge.X.Y., Weber,J.L., Wang.Z., Wilkie.P.J., Holt, M.S., Mishra,S., Donis-Keller,H., Wamich.L., Retief.A.E., Jones.C. and Weissenbach,J. (1993) Hum. Molec. Genet. 2, 909-913. 28. Lathrop,G.M. and Lalouel.J.M. (1984) Am. J. Hum. Genet. 36,460-465. 29. Lathrop.G.M., Lalouel,J.M., Julier.C. and Ott,J. (1984) Proc. Natl. Acad. Sd. U.S.A. 81, 3443-3446. 30. Cooperative Human Linkage Center (CHLC), Fox Chase Cancer Center, Philadelphia.

Downloaded from http://hmg.oxfordjournals.org/ at Erasmus Universiteit Rotterdam on April 20, 2016

Linkage analysis Pairwise lod scores between EXT and each of the marker loci were calculated using the MLINK program of the LINKAGE package version S.I28 assuming an autosomal dominant gene with a frequency of 0.0002, an equal female and male recombination rate and equal marker allele frequencies. For all markers presented here, lod score results were not substantially affected by modification of allele frequencies at the loci. Mutation rate was set at zero. Penetrance was estimated to be about 90% in our families. The frequency of the disease in nongene carriers was set at 0.001. Multipoint analysis was performed between the EXT locus and 5 loci mapping to thepericentromeric region of chromosome 11 (D11S935, D11S905, D11S554, D11S916 and D11S937) using the LINKMAP program 29 with male recombination frequencies of 0.040, 0.027, 0.076, and 0.022 in the respective intervals. These markers were selected because more than half of the transmitting patients in our families were heterozygous at these loci. Relative positions of these loci were determined by combining previouly published linkage maps based on CEPH panel genotypings16-26-27. For this region a male vs. female recombination rate of 3:1 was assumed30 and marker allele frequencies were deduced from the families. To test the robustness of the linkage analysis an 'affecteds only' analysis was performed by phenotyping all asymptomatic individuals as 'unknown' at the disease locus.