Population variability of induced gene expression: an approach for functional variant discovery in human primary cells

¹Elin Grundberg, ³Helena Brändström, ¹Kevin Lam, ¹Bing Ge, ¹Scott Gurd, ⁴Hans Mallmin, ³Östen Ljunggren, ⁴Olle Nilsson, ^1,2Tomi Pastinen
¹McGill University and Genome Quebec Innovation Centre, 740 Dr Penfield Avenue, Montreal, H3A 1A4, Canada, ²McGill University, Department of Human Genetics, 1205 Dr Penfield Avenue, N5/13, Montreal, H3A 1B1, Canada, ³Uppsala University, Department of Medical Sciences, Uppsala University Hospital, 75185 Uppsala, Sweden, ⁴Uppsala University, Department of Surgical Sciences, Uppsala University Hospital, 75185 Uppsala, Sweden

Traditional approaches have remained largely unsuccessful in identifying the underlying genetic etiology of common forms of osteoporosis. In osteoporosis the anabolic effects are mediated by human osteoblasts and multiple pharmacologic and physiologic stimuli have been demonstrated to alter phenotypes of cultured osteoblasts, many of which are believed to be relevant in treatment and pathogenesis of osteoporosis. It is surprising that despite the multiple common clinical phenotypes involving alterations in osteoblast function and proven high heritability of some of these phenotypes the natural variation of in osteoblast responses derived from unrelated individuals have not been systematically explored. Genetic variants underlying interindividual variability of mRNA levels have been demonstrated in immortalized human cell lines, but variants underlying induced gene expression in primary cells as described here have not been studied to date. Herein, we used primary human osteoblasts derived from unrelated donors undergoing total hip replacement to examine natural variation in induced gene expression. Three independent cell cultures from each individual were used and included in subsequent analyses. We have now progressed to identification of interindividually variable induction phenotypes upon multiple trophic stimuli. For example, after stimulation with bone morphogenic protein 2 (BMP2) - a protein involved in bone devlopment and osteoblast differentiation - multiple signaling pathways known to stimulate osteoblastogenesis were rapidly activated, which was assessed by expression profiling (Affymetrix HGU133). In order to study interindividual variability of the induced gene expression, quantitative RT-PCR on BMP2 responded genes within the biochemical pathways was performed. The genes studied included SMAD6, SMAD7, CXCL2, ID1, ID2, HES1, FGF18, and DLX2. In seven out of the eight selected genes, the squared difference in expression levels between individuals was higher than the squared difference of expression levels within. The fold changes in terms of difference between samples vs. within samples of the seven genes ranged from 1.4 to 6.9 with a median fold change of 2.6. These findings may indicate that the variation of induced gene expression in human primary osteblasts is genetically determined. The next steps will be to map variation in human osteoblast responses by candidate gene and whole genome association studies. This work is supported by Genome Quebec and Genome Canada