Identifying genetic determinants of osteoporosis in Malta

Abstract

Introduction & Aim: Osteoporosis is a bone disease with a strong genetic background. The aim of the study was to identify the underlying genetic cause of osteoporosis in a Maltese pedigree. Methodology: A 3-generation family of 9 relatives aged 21-76 years was recruited. Osteoporosis was defined by spine and hip T-scores or Z-scores derived from BMD measurements by DXA. The proband, a 52-year-old female, was diagnosed with osteoporosis at the hip (T-score FN -3.4; TH -2.7). Whole genome sequencing was performed on 6 relatives followed by sequential variant filtering. The shortlisted variants were tested using Competitive Allelic Specific PCR in the Malta Osteoporotic Fracture Study (MOFS) consisting of a case-control collection of 1,045 Maltese postmenopausal women and other independent families. Results: Two zygosity filtering scenarios following a dominant inheritance pattern identified 6 conserved missense variants, including STAT4 c.1309C>T, CXCR1 c.938G>A, TMEM151B c.545A>G, SLC5A10 c.287A>T, CYP1B1 c.182G>A and SUCLG1 c.236G>A, and 2 nonsense variants PCDHA7 c.2257C>T and FAM227B c.621G>A. The novel variants STAT4 c.1309C>T and TMEM151B c.545A>G were monomorphic in the MOFS. Heterozygosity for the PCDHA7 c.2257C>T was linked to a 3-fold increased risk of osteoporosis at the FN (Adj OR: 3.0 [95% CI 0.9-9.5], p=0.080) in line with the phenotype observed in affected relatives. However, the association did not reach statistical significance possibly due to the small sample size. The alternative allele frequency of the SLC5A10 c.287A>T (T=4%) and SUCLG1 c.236G>A (A=3%) variants was much higher compared to the European non-Finnish population in gnomAD suggesting the possibility of a founder effect. These two variants were detected in affected relatives from another independent 3-generation Maltese family, with heterozygosity for the SUCLG1 c.236G>A variant associated with a protective effect in the case-control collection (Adj OR: 0.5 [95% CI 0.2-10], p=0.034), possibly as a result of a small sample size and genotyping bias. Trends were observed with low BMD for the remaining variants (CXCR1 c.938G>A, CYP1B1 c.182G>A, FAM227B c.621G>A), yet none reached statistical significance especially in view of the low alternative allele frequency (<1%). Conclusion: The combination of the 4 variants from each scenario may possibly be the underlying causal determinants of familial osteoporosis in this 3-generation family. Some trends observed are suggestive of the variants’ role in the genetic architecture of bone disease, possibly via signalling pathways known to be involved in bone homeostasis, such as the WNT/β-catenin pathway and JAK-STAT pathway.