POSTER NO: 483

Systematic analysis of candidate genes for cystic fibrosis modifier 1 (CFM1) on human chromosome 19q13

¹J. Zielenski, ¹D. Markiewicz, ¹X. Yuan, ¹M. Corey, ³R. Rozmahel, ¹M. Patel, ¹P. Durie, ¹L-C. Tsui, ¹The CF Modifier Collaborative Group
¹The Hospital for Sick Children, Toronto, ON, Canada, ²University of Toronto, Toronto, ON, Canada, ³University of Alabama, Birmingham, Alabama, USA

Cystic fibrosis (CF) is an autosomal recessive, lethal disease most common among Caucasians. In the classic form, it is characterized by progressive obstructive lung disease, pancreatic dysfunction, elevated sweat electrolytes and bilateral absence of the vas deference. A number of other less common clinical manifestations (meconium ileus, liver disease, diabetes, pancreatitis) may also be present. While some of the clinical variations may be explained by the type of mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene, the CFTR genotype alone is not sufficient to predict the severity of clinical disease in CF. It is thought that secondary genetic factors (modifiers) and environment contribute to the severity of CF phenotype. Previous CF mouse studies from our laboratory provided strong evidence for the presence of a cystic fibrosis modifier gene locus (cfm1) in the proximal region of mouse chromosome 7. We also demonstrated the existence of a similar genetic modifier (CFM1) in the homologous region of human chromosome 19 (region q13). The result of multipoint linkage analysis based on identity by descent (IBD) of CF sibpairs showed strong linkage for meconium ileus (MI) but not pulmonary function. To refine the CFM region, we performed association studies with additional markers using both case-control and transmission disequilibrium tests (TDT) using CF MI and non-MI patients and families with at least one affected child (trios). Of the first 18 markers tested, the strongest association with MI was detected for a microsatellite in intron 1 of the KCNN4 gene. Unfortunately, direct sequencing of the whole KCNN4 gene including 2kb of the 5' upstream region and transcript analysis of KCNN4 have not provided further support for its involvement in CF. Therefore, we have extended our analysis of two genes in its immediate proximity. The first gene (NT_011266) immediately distal to KCNN4 consists of 14 exons and spans 23kb of genomic sequence. It is partially confirmed by EST data. Sequencing analysis of its coding region in MI-discordant sibpairs revealed 3 non-coding sequence variants. We have studied one of these variants (648+46T/G) and an intragenic, complex microsatellite marker in a panel of MI (N=47) and non-MI (N=48) CF families. No significant allelic association could be detected with these two markers but studies will continue with the remaining markers. The second gene (XM_059020), proximal to KCNN4, consists of 5 exons (5.1kb). Genotyping has been performed for an intragenic tetranucleotide (GATA)n in this gene. TDT analysis with MI (N=198) and non-MI (N=87) CF families shows an excess (60%) of transmission of allele 5 (GATA₁₀) to MI patients and similar rate of non-transmission of the same allele to non-MI patients (chi²= 5.427; df=1, p=0.02). In addition, there is a slight excess of non- transmission of allele 6 (GATA₁₁) to MI children and excess of transmission of this allele to non-MI patients (p=0.065). Sequencing analysis of this gene for additional sequence polymorphisms is under way.