179 INTERNAL CONTROL GENES FOR QUANTITATIVE PCR OF PORCINE MESENCHYMAL STEM CELLS DURING ADIPOGENIC AND OSTEOGENIC DIFFERENTIATION IN VITRO

Abstract

Uncovering transcriptomic adaptation of porcine adult stem cells during differentiation in vitro towards a target tissue can provide important information for human adult stem cell therapeutic applications. High-throughput microarrays allow the parallel analysis of thousands of genes simultaneously. However, quantitative RT-PCR (qPCR) remains the chosen method for high-precision mRNA abundance analysis and microarray data verification. Essential for qPCR reliability is data normalization using appropriate internal control genes (ICG). The objective of this study was to find reliable ICG for normalization of qPCR data for porcine adult mesenchymal stem cells induced to differentiate toward adipogenic and osteogenic lineages. Mesenchymal stem cells were harvested from porcine adipose tissue and bone marrow and cultured in vitro with specific differentiation media for up to 3 weeks. The experiment was analyzed by a porcine 13 000-oligo microarray, and data were mined to uncover highly stable genes. Statistical analysis was performed using PROC MIXED of SAS (SAS Institute Inc., Cary, NC, USA). The model included fixed effect of time, cell type, differentiation, and all interaction between them. Pig (n = 3) was considered a random variable. Initial microarray analysis revealed 27 genes with high stability across all samples (sample/reference = 1 ± 0.2). Gene network analysis identified 20 genes without known co-regulation (i.e. common up-stream regulators). Among those genes, we could design high-quality primers (i.e. absence of primer-dimer, single amplicon) only for 10 of them (BANF1, DAK, DPH3, GTF2H3, PRR3, NSUN5, NUBP, SSU72, TIMM17B, and VPS4A), and qPCR using a standard curve was run. Stability of those genes was assessed using pairwise comparison of expression ratios. All genes examined were highly stable with TIMM17B, NSUN5, and VPS4A as the most stable. All the potential ICG tested had significant time, tissue × differentiation, and tissue × time effects. For the 3 most stable genes, we did not observe additional effects, while other ICG were significantly affected by differentiation. The analysis also indicated calculation of the normalization factor using the 3 most stable genes (NF3) as highly reliable; however, the use of 7 genes (NF7) would provide the best reliability. To assess the effect of normalization, we ran qPCR for DBI and COL1A1, genes specific of adipogenic and osteogenic differentiation, respectively. The effect on qPCR data normalization was highly apparent for the adipogenic differentiation and less apparent for the osteogenic differentiation. No differences were observed when qPCR data were normalized by NF3 or NF7. The combination of microarray data and pairwise analysis uncovered novel and high reliable ICG for qPCR normalization in adult porcine stem cells induced into adipogenic and osteogenic lineages.

This work was supported by the Illinois Regenerative Medicine Institute.