Chromosomal rearrangement segregating with adrenoleukodystrophy ...

2 downloads 0 Views 1MB Size Report
8. Drummond-Borg, M., Deeb, S. S. & Motulsky, A. G. (1989). Proc. Natl. Acad. Sci. USA 86, 983-987. 9. Alpern, M., Sack, G. H., Jr., Krantz, D., Jenness, J., Zhang,.
Proc. Natl. Acad. Sci. USA

Vol. 90, pp. 9489-9493, October 1993 Genetics

Chromosomal rearrangement segregating with adrenoleukodystrophy: A molecular analysis (X chromosome/mutation/linkage/genetic disease/neurodegeneration)

GEORGE H. SACK, JR.*tt, MATHEW ALPERN§, THOMAS WEBSTER1, ROBERT P. FEILII, JAMES C. MORRELL*, GRACE CHEN*,**, WINSTON CHEN*,**, C. THOMAS CASKEY¶, AND HUGO W. MOSER*"** *The Kennedy-Krieger Institute, Departments of tMedicine, Biological Chemistry, Pediatrics, and **Neurology, The Johns Hopkins University School of Medicine, Baltimore, MD 21287; §Department of Ophthalmology, University of Michigan, Ann Arbor, MI 48105; $Howard Hughes Medical Institute and the Baylor College of Medicine, Institute for Medical Genetics, Houston, TX 77030; and the IILaboratoire de Genetique Moleculaire des Eucaryotes du Centre National de la Recherche Scientifique, Institut National de la Sante et de la Recherche Medicale Unite Associee 184, Institut de Chimie Biologique, Faculte de Medecine, 67085 Strasbourg Cedex, France

Contributed by Mathew Alpern, July 6, 1993

the manifestations in this kindred. Our companion paper (9) presents detailed studies of color vision in this important

ABSTRACT The relationship between X chromosomelinked adrenoleukodystrophy and the red/green color pigment gene cluster on Xq28 was investigated in a large kindred. The DNA in a hemizygous male showed altered restriction fragment sizes compatible with at least a deletion extending from the 5' end of the color pigment genes. Segregation analysis using a DNA probe within the color pigment gene cluster showed significant linkage with adrenoleukodystrophy (logarithm of odds score of 3.19 at 0 = 0.0). These data demonstrate linkage, rather than association, between a unique molecular rearrangement in the color pigment gene cluster and adrenoleukodystrophy. The DNA changes in this region are thus likely to be helpful for determining the location and identity of the responsible gene.

kindred.

MATERIALS AND METHODS Patients. The extended kindred (identified as kindred 0 in ref. 5) was ascertained by an ALD proband (now deceased). After informed consent, blood was obtained from 19 kindred members. Clinical status was determined by VLCFA analy-

sis (1). DNA. DNA was prepared from lymphoblast cell lines. Genomic DNA plugs (10) were stored in 0.5 M EDTA (pH 8) at 4°C. For enzyme digestion, 50-,ul plugs were dialyzed against 10 mM Tris-HCl/1 mM EDTA for 1 h at 23°C, equilibrated for 15 min in enzyme buffer, and then incubated with 10-20 units of enzyme for 4 h at 37°C. Transverse Alternating Field Electrophoresis (TAFE) and Blot Hybridization. Samples were separated in 0.8% agarose gels using TAFE (ref. 11; Beckman) in 5 mM Tris HCl/0.5 mM EDTA at 250 V for 18 h and 45- to 60-s switch times at 15°C; sizes were estimated by comparison with Saccharomyces cerevisiae chromosomes (12). After electrophoresis, the gels were stained with ethidium bromide, treated with 0.25 M HCl for 10 min and then 0.5 M NaOH/1.5 M NaCl for 40 min, blotted onto GeneScreenPlus (New England Nuclear) in 1Ox standard saline citrate (SSC) for 18 h, and then baked for 1 h at 80°C. Blots were hybridized to labeled probes (1-5 x 106 cpm/ml) in 1.0 M NaCl/1% SDS/10% (wt/vol) dextran sulfate at 65°C for 16 h. Filters were washed twice in 2x SSC at 23°C for 30 min, twice in 2 x SSC/1% SDS at 65°C for 15 min, and once in 0.1x SSC at 65°C for 25 min. Autoradiography was performed using Kodak XAR-5 film with an intensifying screen at -70°C. For reprobing, filters were stripped by boiling in 0.1 x SSC/1% SDS for 20 min. DNA Probes. Fig. 1 shows the normal probe positions. hs7 is a 1.2-kb EcoRI fragment of the red pigment gene cDNA; 9A&6 is a 0.4-kb genomic fragment from the 3' end of intron 4 (15). JHN60 is a fragment of the transcriptional control region 5' to the red pigment gene (17). Fr26 is a 1.9-kb HindIII-Sac I fragment located 9 kb 3' of the sixth exon of each color vision pigment gene within the pigment gene cluster (18). Fr6 and Fr7 are located 8 and 15 kb 5' to the red pigment gene, respectively (13). DX13 (also known as DXS15) is an anonymous Xq28 probe beyond the limits of Fig. 1 (19). Probes

Adrenoleukodystrophy (ALD), a devastating X chromosome-linked neurodegenerative disorder affecting young boys, is associated with a diagnostic accumulation of very long chain fatty acids (VLCFAs; ref. 1). ALD kindreds also include men with adrenomyeloneuropathy (AMN) who have similarly elevated VLCFA levels along with spasticity and gait disturbances but normal cognition. ALD and AMN are different manifestations of the same genetic defect but the precise change is unknown. For simplicity in this report, we will refer to both conditions as ALD unless distinction is necessary. After finding close genetic linkage between the polymorphic Xq28 probe DXS52 and ALD (2), we investigated the association between ALD and defective color vision. Although boys with ALD cannot be tested for their color vision because of rapid neurodegeneration, men with AMN are candidates for such analysis, and we found defective color vision in 42% (12 of 27 men) by using the FarnsworthMunsell 100-hue test (3, 4). This finding led us to examine the red/green color pigment (R/GCP) gene cluster, which showed many reorganizations in ALD (5, 6) that differed from each other and from most common types of R/GCP gene changes in otherwise normal individuals (7, 8). We proposed that this association reflects proximity of the ALD gene and the R/GCP gene cluster but we have not yet established this by formal genetic linkage analysis. Thus we have chosen to study potentially informative kindreds with a combination of molecular biologic and linkage analysis as well as psychophysical studies of color vision. We report here our studies on kindred "O" (5). Molecular analysis of the R/GCP gene region and formal segregation studies support the notion that a common genetic change underlies all

Abbreviations: ALD, adrenoleukodystrophy; AMN, adrenomyeloneuropathy; VLCFA, very long chain fatty acid; R/GCP, red/green color pigment; TAFE, transverse alternating field electrophoresis. tTo whom reprint requests should be addressed at: The Division of Medical Genetics, Blalock 1008, The Johns Hopkins Hospital, Baltimore, MD 21287.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

9489

Genetics: Sack et al.

9490

normal

Proc. Natl. Acad. Sci. USA 90 (1993)

_ centromere

JHN60

Hk kb

300

FIG. 1. Maps of the Xq28 region showing DNA probes and restriction sites. Solid arrowheads represent DNA probes where numbers correspond to "Fr" probes as described (13, 14); open arrowheads represent enzyme cleavage sites used in TAFE gel studies (M, Mlu I; N, Not I). The R/GCP gene cluster is shown by established conventions (15, 16) as repeating units containing one red and one green pigment gene, respectively. Each unit contains DNA hybridizing to probes hs7, 9A&6 (15), and Fr26 as shown. The upper map shows the normal organization and the lower map presents an arrangement compatible with the findings in the 0 kindred; the probe with the star shows an anomalous DNA size fragment in this kindred (see text and Figs. 2 and 3). The hatched box represents the intact 40-kb chromosomal domain surrounding each red or green visual pigment gene. were

labeled with 32p (20) to

a

specific activity of 1-2

x

109

cpm/,g. Segregation Analysis. Xba I digests of the DNAs were separated on a 0.6% agarose gel at 40 V for 40 h. The transfer was hybridized to probe Fr26 and segregation was analyzed using LIPED (21).

RESULTS Previous studies suggested a chromosomal deletion in the 0 kindred (5, 14) and we used TAFE gels to investigate the extent of the change. Transfers of Not I and/or Mlu I digests of DNA from a normal male with one red and one green pigment gene were hybridized to probe hs7; a single 120-kb Not I band appeared (Fig. 2A, lane 1), consistent with earlier observations (16) and a single 470-kb Mlu I band was found (lane 3). The Not I/Mlu I digest (lane 2) showed only a 120-kb band, compatible with having the Not I fragment flanked by Mlu I sites and consistent with (13) showing that a Not I fragment encompassed the R/GCP genes with a Mlu I site beyond the 3' Not I site (Fig. 1, normal map). Fig. 2B shows the Not I data for four members of the 0 kindred. Carrier females (I-2 and 11-5) showed two prominent

A

B Not

M

NNM ....

1.

=

I

MIu =

W.

.:

kb 470-

470 120-

250 120

=

1200-

FIG. 2. Hybridization of hs7 probe DNA to transverse alternating field gels prepared with digests of normal male DNA (A) and DNA from members of the 0 kindred (B and C). Lanes in B and C correspond to kindred members as shown in Fig. 4. (A) Lanes: N, Not I; MN, Mlu I/Not I; M, Mlu I. (B and C) Lanes: 1-2 and 11-5, carrier females; 1-0, normal male; 11-3, affected male.

bands (120 and 250 kb) and another less-intense larger band (-300 kb), a normal male father (1-0) had a single 120-kb band, and the hemizygous affected male (II-3) showed a single 250-kb band. The multiple bands in the samples from carrier females likely reflect sex-specific methylation differences and the presence of two X chromosomes rather than partial digestion since all DNAs were digested under the same conditions [a situation reported by others (22)]. Disappearance of the 120-kb Not I fragment likely reflects loss of the recognition site at the 5' end of the R/GCP gene cluster because (i) at least one green pigment gene of apparently normal size is present on the affected X chromosome (see below and also ref. 5) and (ii) the DNA corresponding to other probes in this region also is deleted as shown below. Mlu I digestions for the same kindred members (Fig. 2C) showed 120- and 470-kb bands in carrier females (1-2 and 11-5). The unaffected male (I-0) had a 470-kb band and the affected male (II-3) had a 120-kb band. The association of the smaller Mlu I band with the carriers and the hemizygote is consistent with the Not I observations and compatible with a change at the 5' end of the R/GCP gene cluster involving the appearance of a Mlu I site. The specificity of these changes for the R/GCP gene region was tested by stripping the blots and hybridizing them to the Xq28 probe DX13. All subjects showed identical patterns (data not shown) indicating that II-3 and the female carriers do not have total reorganization of the Xq28 region. However, they do carry Mlu I restriction fragments 350 kb smaller than normal near the R/GCP genes. These changes are compatible with a simple deletion and we used conventional gels for more details. Earlier studies (5, 14) showed absence of DNA hybridizing to several probes in 11-3 of the 0 kindred but did not establish the extent of the change. Fig. 3 shows hybridization patterns using six DNA probes and DNA from five members of this kindred. As shown by the hs7 and 9A6 hybridizations, the X chromosome carrying the ALD gene lacks the red color pigment gene but has sequences corresponding in size to one green pigment gene. In contrast, the normal male (II-1; Fig. 3, lanes 2), who must have received the other X chromosome, has both red and green pigment genes. No DNA in II-3 hybridizes to probes JHN60, Fr6, and Fr7 at the 5' end of the red gene. Thus, there is a deletion of sequences beginning 3' to the red pigment gene and extending at least 15 kb 5' from the pigment gene cluster (see Fig. 1). These observations are consistent with the TAFE gel results indicating loss of the Not I site between probes Fr6 and JHN60. The change in size of the Mlu I fragment is not as simply explained although one interpretation is that a 350-kb deletion has occurred and this is consistent with the data. The probe Fr26 permitted better localization of the DNA change as well as segregation analysis. Although normal Fr26 shows an 18.5-kb band in normal individuals, a novel 15-kb band is seen in this kindred (see Fig. 3). Fig. 4 shows that this 15-kb band cosegregates with ALD status throughout the kindred. Note the different intensities between the two Fr26 bands in all patients except the hemizygous male. Other patients (and particularly females) have different numbers of normal pigment genes and thus have multiple copies of the normal Fr26 band. Since the 15-kb band revealed by Fr26 segregates consistently with ALD in this kindred (Fig. 4), we performed linkage studies using LIPED to estimate a maximum likelihood value for the logarithm of odds score of 3.19 at 6 = 0.0. Thus, the X chromosomal DNA change in this kindred has removed the red pigment gene and juxtaposed a 15-kb DNA sequence to the remaining pigment gene. This 15-kb sequence contains an Xba I site, which establishes a different 5'

Proc. Natl. Acad. Sci. USA 90 (1993)

Genetics: Sack et al.

FR7

JHN60

FR6

1 2 3 4 5

'I2345Z l.

HS7 2 3 4 5

12 3 4 5

2 34 5

9A6

FR26

12 3 4 5

12 3 4 5

9491

....

"II:',,,b

_m

.

'*4e

__0,fo,40

-8.3

-2.2

1-

-5.4 Bg

4w .i

iO-

4W

'

-3.7 Cr *

-0.6

Taq I

_4w0F-.odS

EcoRI

18.5

-4.4 Br

rd"W

Hindl

*.

-O.82 Dg

0.8 Dr

XtoI

-1.9 Cg

Rso I

EcoFRI-BamHj

FIG. 3. Conventional Southern blot hybridization of six DNA probes using DNA from five members of the 0 kindred shown in Fig. 4 as follows. Lanes: 1, I-1; 2, II-1; 3, 11-2; 4, I-2; 5, 11-3. Note the absence of hybridization of the DNA of the hemizygote (II-3) to probes Fr7, Fr6, and JHN60 and the bands corresponding to the red pigment gene (R), the 15-kb band hybridizing to probe Fr26, and the dosage effects in the other subjects as described in the text (G, green pigment gene). All separations were performed on 1% agarose gels using the enzymes noted except for Rsa I/956 and Xba I/Fr26 for which 1.4% and 0.6% gels were used, respectively.

boundary for the fragment recognized by Fr26. The DNA lost in this rearrangement also includes the sequences recognized by JHN60 and essential for normal R/GCP gene transcription (17) but we cannot currently assess transcription from this region in retinal cone cells.

DISCUSSION Our first studies of this kindred were compatible with aberrant R/GCP genes and changes at the 5' or red end of the cluster (5). These data extend our observations using adjacent recombinant DNA probes. An important conclusion from these data is that there is an intimate relationship between a specific change in the R/GCP gene cluster (with its psychophysical consequences as described in ref. 9) and a structural change in the Xq28 DNA that is linked to ALD. This is a demonstration of linkage

between ALD and a DNA rearrangement in the R/GCP The logarithm of odds score of 3.19 at = 0.0 and the concordant segregation of Fr26 hybridizing patterns (see Fig. 4) strengthen our working hypothesis that ALD is very close to the R/GCP genes and improve the likelihood of locating the ALD gene by examining regions 5' of the R/GCP genes. Aubourg et al. (14) found no changes at the 3' end of the R/GCP genes in ALD patients but noted that DNA from patient II-3 in the 0 kindred did not hybridize to probes Fr6 and Fr7 (see figure 4 of ref. 14). Our data begin to clarify the rearrangement in this kindred. Clearly, sequences are deleted at the 5' end of the R/GCP gene cluster-shown specifically by the loss of red color pigment gene and sequences up to 15 kb 5' to the cluster JHN60, Fr6, and Fr7. However, the TAFE gel data indicate that as much as 350 kb of DNA may be missing. Reconciling these extremes will be aided by studying more regional markers and by identifying the segenes.

7

'1

-

4

9

8

°5

kb

*AAWI

-18.5 -15

:

:

FIG. 4. Segregation in a 0.6% agarose gel of normal and aberrant patterns of Fr26 hybridization with Xba I digests of DNA in extended 0 kindred. Clinical status was determined by VLCFA analysis (1).

9492

Genetics: Sack et al.

quences surrounding the Not I and Mlu I cleavage sites in the patient's genes. Another important genetic and physiologic consideration is that Nathans et al. (17) identified a region immediately 5' to the red pigment gene that is important for controlling the expression of all genes in the cluster. A probe for this region (JHN60) fails to recognize sequences in the 0 kindred (see Fig. 3). Thus, transcriptional control of the remaining visual pigment gene in the hemizygote is not understood and could be aberrant due to the presence of novel sequences. We have based our strategy for isolating the ALD gene on these observations and have begun to scrutinize the region 5' to the R/GCP gene cluster in other ALD kindreds. It may be that even relatively small deletions or other changes in this region cause significant clinical problems since, in their studies of patients with blue-cone monochromacy, Nathans et al. (17) found no individual with a deletion extending further than 15 kb 5' from the red pigment gene. Large and small X chromosome deletions have been described in factor VIII deficiency (23, 24), Duchenne muscular dystrophy (2527), and other disorders. Furthermore, contiguous gene syndromes have been reported at several sites on the X chromosome (28-31). Studying additional ALD kindreds should reveal additional molecular lesions and improve our mapping efforts. Isolating the ALD gene will be very important both for understanding the pathophysiology of this important neurodegenerative disorder and for devising appropriate treatment. In addition, however, further study of this chromosomal region may be particularly valuable for clarifying the nature of the aberrant reorganizations in pigment genes such as we have found (5, 6). There may be local features that promote aberrant recombination and deletion formation in this region. For example, there may be local reiterations (the R/GCP genes themselves present a highly homologous contiguous gene family; ref. 15) or unique repeated sequences that underlie the DNA changes we have reported. In addition to the DNA changes in this kindred, we have found remarkable concomitant changes in color vision in patient II-3. As described in our companion paper (9), these studies extend our earlier initial characterization (4) and show that, when he was first examined, although his color vision closely resembled that of blue cone monochromats, there were important differences compatible with the expression of at least some sort of color pigment gene on Xq28 (presumably, the residual "green" gene). Thus, while we had proposed that the abnormal pigment gene sequences in patient 0 might produce a nonfunctional opsin or not even be transcribed (5), there might be transcription of a mutant pigment whose action spectrum may not be merely an esoteric point of interest to visual physiologists but, instead, fundamental to clarifying the molecular genetics and evolution of the structure of R/GCP proteins. Recently, certain absorption spectra have been related to specific amino acid substitions in the pigment proteins (32, 33) and similar studies may be useful in our kindreds as well. Interestingly, when the color vision of patient II-3 was reexamined 5 years later, the contribution of the additional pigment had disappeared, likely reflecting the subtle neurodegeneration characteristic of AMN. We are grateful for the support of Mr. Daniel M. Kelly and Mr. and Mrs. William M. Griffin and to Dr. Jeremy Nathans for his gift of probes and helpful discussions. We are also grateful to Mrs. Mary A. Mix for secretarial assistance. These studies have been performed with support from the National Foundation, March of Dimes (5-650), the National Institutes of Health (HD-10981), the Mental Retardation Research Center (HD24061), the Commission of the European Communities and the United Leukodystrophy Foundation. M.A. was assisted by National Eye Institute Research Grant EY00197 and

Proc. Natl. Acad. Sci. USA 90 (1993) by a Senior Scientific Investigator Award from Research to Prevent Blindness. 1. Moser, H. W., Moser, A. B., Singh, I. & O'Neill, B. P. (1984) Ann. Neurol. 16, 628-641. 2. Aubourg, P. R., Sack, G. H., Jr., Meyers, D. A., Lease, J. J. & Moser, H. W. (1987) Ann. Neurol. 21, 349-352. 3. Farnsworth, D. (1943) J. Opt. Soc. Am. 33, 568-578. 4. Sack, G. H., Jr., Raven, M. B. & Moser, H. W. (1989) Am. J. Hum. Genet. 44, 794-798. 5. Aubourg, P. R., Sack, G. H., Jr., & Moser, H. W. (1988) Am. J. Hum. Genet. 42, 408-413. 6. Sack, G. H., Jr., & Morrell, J. C. (1993) Invest. Ophthalmol. Vis. Sci. 34, 2634-2637. 7. Nathans, J., Piantanida, T. P., Eddy, R. L., Shows, T. B. & Hogness, D. S. (1986) Science 232, 203-210. 8. Drummond-Borg, M., Deeb, S. S. & Motulsky, A. G. (1989) Proc. Natl. Acad. Sci. USA 86, 983-987. 9. Alpern, M., Sack, G. H., Jr., Krantz, D., Jenness, J., Zhang, H. & Moser, H. W. (1993) Proc. Natl. Acad. Sci. USA 90, 9494-9498. 10. Smith, C. L. & Cantor, C. R. (1987) Trends Biochem. Sci. 12, 284-287. 11. Gardiner, K., Lass, W. & Patterson, D. (1986) Somat. Cell. Mol. Genet. 12, 185-195. 12. Carle, G. & Olson, M. V. (1985) Proc. Natl. Acad. Sci. USA 82, 3756-3760. 13. Feil, R., Aubourg, P., Heilig, R. & Mandel, J.-L. (1990) Genomics 6, 367-373. 14. Aubourg, P., Feil, R., Guidoux, S., Moser, H., Kaplan, J.-C., Kahn, A. & Mandel, J.-L. (1990) Am. J. Hum. Genet. 46, 459-469. 15. Nathans, J., Thomas, D. & Hogness, D. S. (1986) Science 232, 193-202. 16. Vollrath, D., Nathans, J. & Davis, R. W. (1988) Science 240, 1669-1672. 17. Nathans, J., Davenport, C. M., Maumenee, I. H., Lewis, R. A., Hejtmancik, J. F., Litt, M., Lovrien, E., Weleber, R., Bachynski, B., Zwas, F., Klingaman, R. & Fishman, G. (1989) Science 245, 831-838. 18. Feil, R., Aubourg, P., Mosser, J., Douard, A.-M., LePaslier, D., Philippe, C. & Mandel, J.-L. (1991) Am. J. Hum. Genet. 49, 1361-1371. 19. Drayna, D., Davies, K., Hartley, D., Mandel, J.-L., Camerino, G., Williamson, R. & White, R. (1984) Proc. Natl. Acad. Sci. USA 81, 2836-2839. 20. Feinberg, A. P. & Vogelstein, B. (1983) Anal. Biochem. 132, 6-13. 21. Ott, J. (1976) Am. J. Hum. Genet. 26, 528-529. 22. Arveiler, B., Vincent, A. & Mandel, J.-L. (1989) Genomics 4, 460-471. 23. Antonarakis, S. E., Waber, P. G., Kittur, S. D., Patel, A. S., Kazazian, H. H., Jr., Mellis, M. A., Counts, R. B., Stamatoyannopoulos, G., Bowie, E. J. W., Fass, D. N., Pittman, D. D., Wozney, J. M. & Toole, J. J. (1985) N. Engl. J. Med. 313,

842-848. 24. Youssoufian, H., Antonarakis, S. E., Aronis, S., Tsiftis, G., Phillips, D. G. & Kazazian, H. H., Jr. (1987) Proc. Natl. Acad. Sci. USA 84, 3772-3776. 25. Kunkel, L. M. & co-authors (1986) Nature (London) 322, 73-77. 26. Gillard, E. F., Chamberlin, J. S., Murphy, E. B., Duff, C. L., Smith, B., Burghes, A. H. M., Thompson, M. W., Sutherland, J., Oss, I., Bodrug, S. E., Klamut, H. J., Ray, P. N. & Worton, R. G. (1989) Am. J. Hum. Genet. 45, 507-520. 27. Lindlof, M., Kiuru, A., Kaariainen, H., Kalimo, H., Lang, H.,

Pihko, H., Rapola, J., Somer, H., Somer, M., Savontaus, M.-L. & De La Chapelle, A. (1989) Am. J. Hum. Genet. 44, 4%-503. 28. Francke, U., Ochs, H. D., deMartinville, B., Giacalone, J., Lindgren, V., Disteche, C., Pagon, R. A., Hofker, M. H., VanOmmen, G. B., Pearson, P. L. & Wedgwood, R. J. (1985) Am. J. Hum. Genet. 37, 250-267. 29. Emanuel, B. S. (1988) Am. J. Hum. Genet. 43, 575-578. 30. McCabe, E. R. B., Towbin, J., Chamberlain, J., Baumbach, I., Witkowski, J., Van Ommen, G. & Koenig, M. (1989) J. Clin. Invest. 83, 95-99.

Genetics: Sack et al. 31. Ballabio, A., Bardoni, B., Carrozzo, R., Andria, G., Bick, D., Campbell, L., Hamel, B., Ferguson-Smith, M. A., Gimelli, G., Fraccaro, M., Maraschio, P., Buffardi, O., Guioli, S. & Camerino, G. (1989) Proc. Natl. Acad. Sci. USA 86, 10001-10005.

Proc. Natl. Acad. Sci. USA 90 (1993)

9493

32. Merbs, S. L. & Nathans, J. (1992) Science 258, 464-466. 33. Deeb, S. S., Lindsey, D. T., Hibiya, Y., Sanocki, E., Winderickx, J., Teller, D. Y. & Motulsky, A. G. (1992) Am. J. Hum. Genet. 51, 687-700.