Genomic structure of the human plasma prekallikrein gene, identification of allelic variants, and analysis in end-stage renal disease

Kallikreins are serine proteases that catalyze the release of kinins and ot her vasoactive peptides. Previously, we have studied one tissue-specific (H . Yu et at, 1996, J. Am. Sec. Nephrol 7: 2559-2564) and one plasma-specific (H. Yu et al., 1998, Hypertension 31: 906-911) human kallikrein gene in en d-stage renal disease (ESRD). Short sequence repeat polymorphisms for the h uman plasma kallikrein gene (KLKB1; previously known as KLK3) on chromosome 4 were associated with ESRD in an African American study population. This study of KLKB1 in ESRD has been extended by determining the genomic structu re of KLKB1 and searching for allelic variants that may be associated with ESRD. Exon-spanning PCR primer sets were identified by serial testing of pr imer pairs designed from KLKB1 cDNA sequence and DNA sequencing of PCR prod ucts. Like the rat plasma kallikrein gene and the closely related human fac tor XI gene, the human KLKB1 gene contains 15 exons and 14 introns. The lon gest intron, F, is almost 12 kb long. The total length of the gene is appro ximately 30 kb. Sequence of the 5'-proximal promoter region of KLKB1 was ob tained by shotgun cloning of genomic fragments from a bacterial artificial clone containing the KLKB1 gene, followed by screening of the clones using exon 1-specific probes. Primers flanking the exons and 5'-proximal promoter region were used to screen for allelic variants in the genomic DNA from ES RD patients and controls using the single-strand conformation polymorphism technique. We identified 12 allelic variants in the 5'-proximal promoter an d 7 exons. Of note were a common polymorphism (30% of the population) at po sition 521 of KLKB1 cDNA, which leads to the replacement of asparagine with a serine at position 124 in the heavy chain of the A2 domain of the protei n. In addition, an A716C polymorphism in exon 7 resulting in the amino acid change H189P in the A3 domain of the heavy chain was observed in 5 patient s belonging to 3 ESRD families. A third polymorphism in the coding sequence was a C699A shift that caused an, amino acid change, H183Q. This allele wa s observed in 8 cases from 6 ESRD families but was not found in any control DNAs. Individually or combined, the allelic variants observed are not stat istically associated with ESRD, though in several cases (e.g., H183Q) the s mall number of people in the population carrying these alleles limits our a bility to statistically test for significant association with ESRD. Two new CA/GT repeat polymorphic markers, designated KLK3f and KLK3g, that have he terozygosities of 0.65 and 0.84, respectively, were identified within intro ns M and N. Analysis using the relative predispositional effect technique i ndicated that the frequencies of alleles 4 and 8 of KLK3f and allele 8 of K LK3g were significantly different between controls and ESRD cases. They acc ounted for 0.226, 0.096, and 0.313, respectively, in the probands of 166 ES RD families compared to 0.172, 0.066, and 0.244 in 139 healthy race-matched controls (allele P and total P < 0.05 for all three alleles). Therefore, a lthough polymorphisms in the coding and 5'-proximal promoter of KLKB1 show no statistically significant association with ESRD in African Americans, th ere is still evidence for association of this part of chromosome 4 with ESR D. This observation suggests that other sequences within or near KLKB1, or another gene nearby, may contribute to ESRD susceptibility. (C) 2000 Academ ic Press.