Selection of Reference Genes for Normalization of Gene Expression in Thermobia domestica (Insecta: Zygentoma: Lepismatidae)

Abstract

Zygentoma occupies a key evolutionary position for understanding the evolution of insect metamorphosis but has received little attention in terms of genetic analysis. To develop functional genomic studies in this insect, we evaluated five candidate internal reference genes for quantitative RT-PCR (qPCR) studies from Thermobia domestica, a representative species of Zygentoma, including Actin 5C (Actin5C), Elongation factor-1 alpha (EF1A), Ribosome protein S26 (RPS26), Ribosome protein L32 (RPL32), and Superoxide dismutase 2 (SOD2), at different developmental stages, in various body parts, and under dsRNA microinjection and starvation stresses, using four algorithms (delta Ct, geNorm, NormFinder and BestKeeper) and a comparative algorithm (RefFinder). Specific suitable reference genes were recommended across specific experimental conditions, and the combination of RPS26 and RPL32 was appropriate for all tested samples. Employing our selected reference gene combination, we investigated the gene expression pattern of Myoglianin (Myo), a crucial gene-regulating insect metamorphosis, in ametabolous T. domestica, and demonstrated the efficiency of RNA interference (RNAi) in firebrat nymphs. This study provides a basis for reliable quantitative studies of genes and greatly benefits evolutionary and functional genomics studies in Zygentoma.

Keywords: RNA interference; Thermobia domestica; expression stability; quantitative real-time PCR; reference genes.

Figure 1

Ct value distribution of five reference genes in all analyzed samples of Thermobia domestica. The whiskers of the boxes are the minimum and maximum Ct values, and the horizontal lines inside the boxes are the median values.

Figure 2

Stable analysis of five candidate genes in different tissues of Thermobia domestica (head, prothorax, mesothorax/metathorax, abdomen, and whole body) at the developmental stages of nymphs N6 (A), N7(B), N8 (C), and N9 (D), and adults A1 (E), A3 (F), and A4 (G) based on four algorithms.

Figure 3

Stable analysis of five candidate genes at different developmental stages from nymphs (N6–N9) and adults (A1, A3, and A4) of Thermobia domestica in different tissues: head (A), prothorax (B), mesothorax/metathorax (C), abdomen (D), and whole body (E) based on four algorithms.

Figure 4

Stable analysis of five candidate genes under RNAi-Myo (A) or starvation stresses (B) based on four algorithms.

Figure 5

Evaluation of the optimal number of reference genes for normalization in Thermobia domestica. The pairwise variations (V _n/n+1, n is the number of reference genes) was calculated between two sequential normalization factors (NF_n and NF_n+1) by geNorm with a cut-off value of 0.15 [12] using data: (A) from different tissues at the developmental stages of N6–N9, A1, A3, and A4; (B) at different developmental stages in the head, prothorax, mesothorax/metathorax, and abdomen; (C) in whole insects at the 8th nymphal instar under RNAi and starvation treatments.

Figure 6

Relative expression levels of Tdmyo under various experimental conditions normalized with RPS26 and RPL32. Error bars show the mean standard error calculated from three biological replicates. (A) In the head of T. domestica from the middle nymphs to adults. (B) In four body parts from individuals at the 8th nymphal instar. (C) In whole insects at the 8th nymphal instar under RNAi treatment. (D) In whole insects at the 8th nymphal instar under feeding or starvation conditions. Different letters (a, b) in (A,B) indicate significant differences of values (ANOVA, LSD, p < 0.05). The statistical level in (C,D) was assessed with ** p < 0.01.