Evaluating and Validating Sunflower Reference Genes for Q-PCR Studies Under High Temperature Condition

Abstract

Background: Q-PCR is the method of choice for PCR- based transcriptomics and validating microarray-based and RNA-seq results. Proper application of this technology requires proper normalization to correct as much as possible errors propagating during RNA extraction and cDNA synthesis.

Objectives: The investigation was performed to find stable reference genes in sunflower under shifting in ambient temperature.

Materials and methods: Sequences of five well-known reference genes of Arabidopsis (Actin, Ubiquitin, Elongation factor-1, GAPDH, and SAND) and one well-known reference gene inhuman, Importin, were subjected to BLASTX against sunflower databases and the relevant genes were subjected to primer designing for q-PCR. Two sunflower inbred lines were cultivated at two dates so that anthesis occurred at nearly 30 °C and 40 °C (heat stress). The experiment was repeated for two years. Q-PCR was run on samples taken for two planting date separately at the beginning of anthesis for each genotype from leaf, taproots, receptacle base, immature and mature disc flowers and on pooled samples comprising of the tissues for each genotype, planting dates and also all tissues for both genotypes and both planting dates. Basic statistical properties of each candidate gene across all the samples were calculated. Furthermore, gene expression stability analysis was done for six candidate reference genes on Cq mean of two years using three independent algorithms, geNorm, Bestkeeper, and Refinder.

Results: Designed primers for Actin2, SAND, GAPDH, Ubiquitin, EF-1a, and Importin yielded a single peak in melting curve analysis indicating specificity of the PCR reaction. Basic statistical analysis showed that Actin2 and EF-1a had the highest and lowest expression levels across all the samples, respectively. Actin2 appeared to be the most stable reference gene across all the samples based on the three used algorithms. Pairwise variation analysis revealed that for samples taken under ambient temperature of 30 °C, Actin2, EF-1a, SAND and for those taken under ambient temperature of 40 °C, Actin2, EF-1a, Importin and SAND have to be used for normalization in q-PCR studies. Moreover, it is suggested that normalization to be based on Actin2, SAND and EF-1a for vegetative tissues and Actin2, EF-1a, SAND and Importin for reproductive tissues.

Conclusions: In the present research, proper reference genes for normalization of gene expression studies under heat stress conditions were introduced. Moreover, the presence of genotype-by- planting date interaction effects and tissue specific gene expression pattern on the behavior of the most three stable reference genes was indicated.

Keywords: Heat stress; Reference genes; Sunflower.

Figure 1

Expression levels of six candidate reference genes in different samples. Pooled sample is comprised of all tissues, planting dates of all the genotypes. Lines crossing the boxes depict the medians. Crossed circles indicate means. Stars indicate outlier data; Boxes indicate the interquartile range. Whiskers represent 95% confidence intervals.

Figure 2

Stability analysis using GeNorm algorithm. A) The analysis was performed for pooled samples, two vegetative tissues: B)leaf a nd C) root, two reproductive tissues: D) receptacle base and E) flower, two planting dates: F) date 1 and G) date 2, two sunflower inbred lines: H) B-line 1221 and I) B-line 19. Vertical axis shows average expression stability values (M) and the genes are arranged on the horizontal axis in each panel from the most stable (left side) to the least stable (right side) ones.

Figure 3

The effect of planting date ×genotype interaction on behavior stability of six candidate reference genes according to geNorm algorithm. The geNorm stability index of the second most stable A) and the most unstable B) reference genes were plotted for BF1221 and B-line19. Crossing of the line connected planting dates indicate interactions between the genotypes and the planting dates.

Figure 4

Pairwise variation ($V_n/V_{\mathrm{n+1}}+1$ ) was analyzed to find optimal number of reference genes for accurate gene expression normalization for q-PCR data. The analysis was performed on 6 candidate reference genes by geNorm software. Dashed line shows the cut of value so that V_n/V_n+1<0.15 means that the addition of more reference gene would have no significant contribution to normalization in q-PCR data analysis. D1 and D2 stands for the first and second planting date, respectively. All data refers to pooling all tissues from the two genotypes and two planting dates into a unit sample.

Figure 5

Validation of selected reference genes under heat stress condition. Relative expression of A) CD850746 and B) CD849228 in the sunflower leave at heat stress condition compared to normal ambient temperature was calculated after normalization based on the most stable, the two most stable and the most unstable reference genes. Under normal temperature Actin2, Actin2 + EF-1a, and Ubiquitin were the most stable, the two most stable and the most unstable reference genes, respectively. Under heat stress condition, Actin2, Actin2+ EF-1a, and GAPDG were the most stable, the two most stable and the most unstable reference genes, respectively. Relative gene expression analysis was performed according to the ……Bars indicate the standard errors (n=3).