Abstract
Osteoporosis (OP) is a multifactorial disease influenced by genetic factors in more than half of the cases. In spite of the efforts to clarify the relationship among genetic factors and susceptibility to develop OP, many genetic associations need to be further functionally validated. Besides, some limitations as the choice of stably expressed reference genes (RG) should be overcome to ensure the quality and reproducibility of gene expression assays. To our knowledge, a validation study for RG in OP is still missing. We compared the expression levels, using polymerase chain reaction quantitative real time (qPCR) of 10 RG (G6PD, B2M, GUSB, HSP90, EF1A, RPLP0, GAPDH, ACTB, 18 S and HPRT1) to assess their suitability in OP analysis by using GeNorm, Normfinder, BestKeeper and RefFinder programs. A minimal number of two RG was recommended by GeNorm to obtain a reliable normalization. RPLP0 and B2M were identified as the most stable genes in OP studies while ACTB, 18 S and HPRT1 were inadequate for normalization in our data set. Moreover, we showed the dramatic effects of suboptimal RG choice on the quantification of a target gene, highlighting the importance in the identification of the most appropriate reference gene to specific diseases. We suggest the use of RPLP0 and B2M as the most stable reference genes while we do not recommend the use of the least stable reference genes HPRT1, 18 S and ACTB in OP expression assays using PBMC as biological source. Additionally, we emphasize the importance of individualized and careful choice in software and reference genes selection.