196 1 711 19 907 32 261 12 354 7,53 UBC 21 665 0 163 1 422 19 475

196 1.711 19.907 32.261 12.354 7,53 UBC 21.665 0.163 1.422 19.475 30.387 10.912 6,60 YWHAZ 24.720 0.193 1.685 22.733 32.853 10.120 6,86 Note. S.e.m, standard error of mean; CtCV%, Coefficients of variations of candidate reference genes. Results of validation programs In order to determine the stability of genes and thus find the best endogenous controls, the data were analysed by geNorm and NormFinder. In these analyses, medians were used to replace missing values because they occurred due to inconsistencies between replicates rather than from low expression. The ranking of the gene expression stability

values (M) of the tested endogenous control genes using geNorm is illustrated in Figure 1.A. The genes with the highest M, i.e. the least stable genes, gets stepwise excluded until the most stable genes remain. The Staurosporine clinical trial best two genes are ranked without distinguishing between them. HPRT1 and PPIA were identified as the most stable pair of genes, followed by PGK1 as the third most stable gene. Furthermore, pairwise variation were also calculated using geNorm in order to determine the optimal BIBW2992 concentration number of genes required for normalization, Figure 1.B. The analysis showed that HPRT1 and PPIA may be sufficient

for calculation of the normalization ACY-1215 factor and normalization to genes of interest, since the V2/3 value is in this analysis equal to the cut-off value of 0.15 [19]. However, there is a gradual decrease in the pairwise variability plot and thereby an improvement to the normalization factor Mannose-binding protein-associated serine protease by adding additional genes to the calculation. Nevertheless, two or three genes would be satisfactory for normalization according to the cut-off value of 0.15. While geNorm uses a pairwise comparison approach, NormFinder first estimates the intra-group and then the inter-group variability of expression of a control gene [17]. In contrast to the geNorm results, NormFinder ranked RPLP0 as the most stable gene, with TBP and GUSB closely behind as second and third, respectively (Figure 2). However, using this algorithm the combination of IPO8 and PPIA turned out to have a lower stability score than the most stable single gene. Thus

this combination is more suitable for normalizing qPCR. There was considerably closer agreement between the geNorm and Normfinder results on the least stable genes, with the order of 4 out of 5 worst ranking genes being identical; ACTB, 18S, B2M and TFRC. These genes had a stability value more than twice so high (geNorm) and more than 3 times so high (NormFinder) as the best ranking genes. Figure 1 GeNorm analysis of the candidate reference genes. (A) Average expression stability values of reference genes. Genes are presented in an increasing order of stability from left to right with ACTB being the least stable gene and HPRT1 and PPIA the most stable genes. (B) Determination of optimal number of control genes for normalization.

Comments are closed.