Two large Israeli Bedouin extended families from the Negev desert, presenting with autosomal recessive congenital bilateral nuclear cataracts, were studied.
Figure 1depicts the two pedigrees. At the time of study, 14 patients existed, and linkage to 32 candidate genes was sought
(Table 1) . Twenty-one unaffected family members were also tested to generate haplotypes with two polymorphic markers flanking each gene. Linkage was ruled out for all loci tested, except for the
CRYBB1 locus (results not shown). Studies using polymorphic markers at the
CRYBB1 locus, shown in
Figures 1 and 2 , demonstrated that all patients in family 1
(Fig. 1A)were homozygous for a haplotype of three microsatellite markers flanking
CRYBB1 and that all their nonaffected parents were heterozygous carriers of that haplotype. In family 2
(Fig. 1B) , there were several crossover events at that locus in the various branches of this larger kindred, with several individuals marrying more distant relatives. Yet, in affected individuals of kindred 2, there was consistent homozygosity for the same allele of
D22S1167 (that is immediately adjacent to
CRYBB1), as in affected individuals of kindred 1
(Fig. 1A) . A significant lod score of 6.57 at θ = 0 was obtained for
D22S1167, a polymorphic marker residing within 0.1 cM of two β-crystallin genes,
CRYBB1 and
CRYBA4, known to encode proteins expressed in lens tissue. No mutations were found in
CRYBA4. However, as shown in
Figure 3A , an identical homozygous delG168 mutation in exon 2 of
CRYBB1 was demonstrated in affected individuals of both families, generating a frameshift (and missense protein sequence) as of amino acid 57 and a stop codon causing truncation after amino acid 107. dHPLC studies
(Fig. 3B)showed that all affected individuals of both families were homozygous for the mutation, all parents of affected individuals were heterozygous for the mutation, whereas none of 100 unrelated Bedouin individuals from southern Israel carried the mutation.