Twenty-two molecular diagnostic laboratories from 14 countries participated in a consortium study to estimate the impact of Factor VIII gene inversions in severe hemophilia A. A total of 2,093 patients with severe hemophilia A were studied; of those, 740 (35%) had a type 1 (distal) factor VIII inversion, and 140 (7%) showed a type 2 (proximal) inversion. In 25 cases, the molecular analysis showed additional abnormal or polymorphic patterns. Ninety-eight percent of 532 mothers of patients with inversions were carriers of the abnormal factor VIII gene; when only mothers of nonfamilial cases were studied, 9 de novo inversions in maternal germ cells were observed among 225 cases (approximately 1 de novo maternal origin of the inversion in 25 mothers of sporadic cases). When the maternal grandparental origin was examined, the inversions occurred de novo in male germ cells in 69 cases and female germ cells in 1 case. The presence of factor VIII inversions is not a major predisposing factor for the development of factor VIII inhibitors; however, slightly more patients with severe hemophilia A and factor VIII inversions develop inhibitors (130 of 642 [20%]) than patients with severe hemophilia A without inversions (131 of 821 [16%]).

The factor VIII gene, which Is defective In hemophilia A, Is located In the last megabase of the long arm of the X chromosome. Inversions due to intrachromosomal homologous recombination between mispaired copies of gene A located within intron 22 of the gene and about 500 kb telomerlc to it account for nearly half of all cases of severe hemophilia A. We hypothesized that pairing of Xq with its homolog inhibits the Inversion process, and that, therefore, the event originates predominantly in male germ cells. In all 20 informative cases In which the inversion originated in a maternal grandparent, DNA polymorphism analysis determined that it occurred in the male germline. In addition, all but one of 50 mothers of sporadic cases due to an inversion were carriers. Thus, these data support the hypothesis and Indicate that factor VIII gene inversions leading to severe hemophilia A occur almost exclusively In male germ cells.

In order to contribute to the development of the transcriptional map of human chromosome 21 (HC21) we have used exon trapping to identify portions of HC21 genes. Using pools of random HC21-specific cosmids from the LL21NC02-Q library and cosmids from 21q22.3 we have identified five different coding regions with strong homology to the lanosterol synthase genes of rat and yeast. This enzyme catalyzes the cyclization of squalene-2,3-epoxide lanosterol, which is the parental compound of all steroids in mammals. Using somatic cell hybrids and HC21 yeast artificial chromosomes (YACS) and cosmids, we mapped the human lanosterol synthase cDNA gene to 2lq22.3 between markers D21S25 and 21qter. Cosmid Q7G8 from the LL21NC02-Q library and YAC 145D8 from the CEPH HC21 contig contain this human gene. We cloned a portion of the human lanosterol synthase cDNA (almost 85% of the coding region) from a brain cDNA library and determined its nucleotide sequence. The predicted human protein shows 83% identity to its rat and 40% to its yeast homolog. No obvious candidate human disease exists for lanosterol synthase deficiency and the role (if any) of triplication of this gene in the various phenotypes of trisomy 21 is unknown.

In order to contribute to the development of the transcriptional map of human chromosome 21 (HC21) we have used exon trapping to identify portions of HC21 genes. Using pools of random HC21-specific cosmids from the LL21NC02-Q library and cosmids from 21q22.3 we have identified five different coding regions with strong homology to the lanosterol synthase genes of rat and yeast. This enzyme catalyzes the cyclization of squalene-2,3-epoxide lanosterol, which is the parental compound of all steroids in mammals. Using somatic cell hybrids and HC21 yeast artificial chromosomes (YACS) and cosmids, we mapped the human lanosterol synthase cDNA gene to 2lq22.3 between markers D21S25 and 21qter. Cosmid Q7G8 from the LL21NC02-Q library and YAC 145D8 from the CEPH HC21 contig contain this human gene. We cloned a portion of the human lanosterol synthase cDNA (almost 85% of the coding region) from a brain cDNA library and determined its nucleotide sequence. The predicted human protein shows 83% identity to its rat and 40% to its yeast homolog. No obvious candidate human disease exists for lanosterol synthase deficiency and the role (if any) of triplication of this gene in the various phenotypes of trisomy 21 is unknown.