Identification of suitable reference genes in buffalo grass for accurate transcript normalization under various abiotic stress conditions

• We screened appropriate reference genes in Buchloe dactyloides by qRT-PCR.
• GAPDH, β-ACTIN, DNAJ, and PP2A were the most reliable reference genes in all samples.
• Expression level of reference genes varies in different experimental conditions.
• The expression of DREB2 confirmed that the appropriate reference genes are crucial.

Quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR) is a sensitive technique for normalization of the gene expression level of target genes. Buffalograss (Buchloe dactyloides), a warm-season turfgrass with strong abiotic stress resistance, is widely used in North China. Up to now, no work was performed to evaluate the reference genes in buffalograss. In this study, the expression profiles of ten potential reference genes were examined by qRT-PCR in 24 buffalograss samples, which were subjected to a different treatment (salt, osmotic, cold and heat). Three qRT-PCR analysis methods (GeNorm, NormFinder, and Bestkeeper) were used to evaluate the stability of gene expression. The results indicated that DNAJ and β-ACTIN were the optimal reference genes for salt-treated leaves, and the combination of PP2A and GAPDH was better reference genes for PEG-treated leaves. Under cold stress, DNAJ and β-ACTIN showed less variety of expression level in leaves. DNAJ and GAPDH exhibited the most stable expression in heat-treated samples. To sum up, glyceral-dehyde-3-phosphate dehydrogenase (GAPDH), β-ACTIN, DNAJ-like protein (DNAJ) and protein phosphatase 2A (PP2A) were selected as the most stable reference gene among all tested samples. To further validate the suitability of these reference genes, the expression levels of DREB2 (homologs of AtDREB2) were analyzed in parallel. Our results show that the best reference genes differed across different experimental conditions, and these results should enable better normalization and quantification of transcript levels in buffalograss in the future.