Housekeeping gene expression variability in differentiating and non-differentiating 3T3-L1 cells

Abstract

Normalization is a crucial step in gene expression analysis to avoid misinterpretation. Reverse transcription-quantitative polymerase chain reaction was used to measure the expression of 10 candidate housekeeping genes in non-differentiated (ND) and differentiated (DI) 3T3-L1 cells on days 5 and 10. We used geNorm, NormFinder, BestKeeper, RefFinder, and the ∆Ct method to evaluate expression stability. The findings revealed that (1) the expression levels of the reference genes changed over time, even in non-differentiating cells, and (2) peptidylprolyl isomerase A (Ppia) and TATA box-binding protein (Tbp) were stable reference genes for 10 days in both undifferentiated and differentiated 3T3-L1 cells. Notably, the expression of known reference genes in non-differentiating cells was altered throughout the experiment.

Keywords: adipocytes; differentiation; gene expression; normalization; reference gene.

Figure 1.

Oil Red O (ORO) staining and quantification of accumulated lipid droplets in non-differentiated (ND) and differentiated (DI) 3T3-L1 cells on days 0, 5, and 10. (a) Accumulation of lipid droplets was observed in the DI groups by staining with ORO, at days 5 and 10 after differentiation induction. Scale bar = 100 µm. (b) Quantification of lipid droplet accumulation in 3T3-L1 cells. The absorbance of the eluted ORO obtained from the stained oil droplets was measured using a microplate reader at 492 nm. *P < 0.05 and **P < 0.01 using Bonferroni test. (c) Relative Lep mRNA expression per 50 ng of RNA in all five groups. Lep expression was higher in the DI groups than in the ND groups, indicating the stage of mature adipocytes in the DI groups. *P < 0.05 and **P < 0.01 using Bonferroni test. Data are presented as the mean ± standard deviation (SD) of triplicate samples.

Figure 2.

Distribution of cycle threshold (Ct) values of each candidate reference gene across all sample groups. A line within each box indicates the median Ct values, whereas the lower and upper boundaries of each box indicate the first and third quartiles of the data, respectively. Minimum and maximum values are indicated by the lower and upper ends of the whiskers, respectively.

Figure 3.

Heatmap of the gene expression levels of 10 reference genes and the gene expression stability levels evaluated using five algorithms. (a) Heatmap image for each gene expression level in the control, non-differentiated (ND), and differentiated (DI) groups. The colours in the heatmap show the Z-scores of gene expression levels in the triplicate samples. Gene expression levels higher than the mean are indicated in red and those lower than the mean are indicated in green. (b) The expression levels of 10 reference genes were analysed using geNorm, BestKeeper, NormFinder, RefFinder, and the ΔCt method in the control, ND, and DI groups. A lower value (right side of the horizontal axis) indicates a reference gene with a more stable expression.

Figure 4.

Selection of reference genes affects the expression pattern of the target genes. (a) Expression pattern of representative stable and unstable reference genes in non-differentiating and differentiating 3T3-L1 cells. The non-normalized expression levels of Ppia, Rn18s, Hmbs, and Tfrc show different patterns indicating that the gene expression levels differed among the groups. Differences were considered statistically significant at *P < 0.05 and **P < 0.01 compared to the Control group using Dunnett’s test. (b) Comparative gene expression levels of Pparg in non-differentiated (ND) and differentiated (DI) groups against the representative gene expression levels. *P < 0.05 and **P < 0.01 using Bonferroni test. (c) Cebpa expression levels in non-differentiated (ND) and differentiated (DI) groups normalized to the selected reference genes. *P < 0.05 and **P < 0.01 using Bonferroni test. Data for all graphs are presented as the mean ± standard deviation (SD) of triplicate samples.