Analysis of variation in expression of autosomal dominant osteopetrosis type 2: Searching for modifier genes

Introduction: Autosomal Dominant Osteopetrosis type II (ADO2) is a heritable osteosclerotic disorder that results from heterozygous mutations in the ClCN7 gene. Analysis of ADO2 in our pedigrees indicates that the penetrance is 66%, with a highly variable phenotype. Methods: To identify genes that modify disease status, we performed a 10 cM genome-wide scan using 400 microsatellite markers in 112 subjects from our 8 largest ADO2 families with mutations in the ClCN7 gene. Results were analyzed by parametric linkage analysis using autosomal dominant and recessive models for affects on disease status. Follow-up genotyping with additional microsatellite markers was performed for regions with LOD scores over 1.5. In addition, we compared the frequency of two nonsynonymous SNPs, rs12926089 (V418M) and rs11559208 (K691E), and one promoter SNP rs960467 in the normal ClCN7 allele between a sample of unaffected gene carriers and clinically affected subjects to test the hypothesis that genetic variation in the non-disease allele within the ClCN7 gene might influence disease expression. Results: We found potential evidence of linkage for a modifier gene(s) on 9q21–22 with a LOD score of 1.89, which is not statistically significant, but interesting. We also found that, for SNP V418M on the non-disease allele with the wild-type ClCN7 sequence, 94.92% (56/59) of clinically affected subjects and 78.13% (25/32) of unaffected gene carriers had a valine while 5.08% (3/59) of the affected subjects and 21.88% (7/32) of unaffected gene carriers had a methionine (P < 0.03). Unfortunately, SNP K691E was not informative in our families. For SNP rs960467, on the non-disease allele with the wild-type ClCN7 gene, 87.93% (51/58) of clinically affected subjects and 62.50% (20/32) of unaffected gene carriers had a C allele while 12.07% (7/58) of the clinically affected subjects and 37.50% (12/32) of unaffected gene carriers had a T allele (P < 0.007). As expected, the polymorphisms on the disease allele were not associated with disease status. Conclusions: Chromosome 9q21–22 may harbor a modifier gene(s) that affect(s) ADO2 disease status and severity. Additionally, we find the associations between the polymorphisms on the non-disease allele and unaffected gene carrier status.