Multiple miRNAs jointly regulate the biosynthesis of ecdysteroid in the holometabolous insects, Chilo suppressalis

Abstract

The accurate rise and fall of active hormones is important for insect development. The ecdysteroids must be cleared in a timely manner. However, the mechanism of suppressing the ecdysteroid biosynthesis at the right time remains unclear. Here, we sequenced a small RNA library of Chilo suppressalis and identified 300 miRNAs in this notorious rice insect pest. Microarray analysis yielded 54 differentially expressed miRNAs during metamorphosis development. Target prediction and in vitro dual-luciferase assays confirmed that seven miRNAs (two conserved and five novel miRNAs) jointly targeted three Halloween genes in the ecdysteroid biosynthesis pathway. Overexpression of these seven miRNAs reduced the titer of 20-hydroxyecdysone (20E), induced mortality, and retarded development, which could be rescued by treatment with 20E. Comparative analysis indicated that the miRNA regulation of metamorphosis development is a conserved process but that the miRNAs involved are highly divergent. In all, we present evidence that both conserved and lineage-specific miRNAs have crucial roles in regulating development in insects by controlling ecdysteroid biosynthesis, which is important for ensuring developmental convergence and evolutionary diversity.

Keywords: development; ecdysteroid; insect; metamorphosis; miRNA.

FIGURE 1.

Small RNA library sequencing and annotation. (A) Length distribution of small RNA reads. The reads have peaks at 22–24 nt and 26–29 nt, accounting for 36.68% and 24.49% of the total reads, respectively. These two peaks were in accordance with the length characterization of miRNA and piRNA, respectively. (B) Annotation of small RNAs. Among the total small RNAs, rRNA accounts for 2.56%. The percentages of each type of sRNAs are indicated in brackets.

FIGURE 2.

Differently expressed miRNAs during pupation, pupal development, and eclosion in C. suppressalis by microarray analysis. (A) Heat map for 54 differentially expressed miRNAs with significant differences (P < 0.05). The selected seven development points were aging larval, prepupal, early pupal, compound eye formation, pretarsus formation, pupal elongation, and adult stages. The microarray data were clustered after normalization by LOWESS (locally weighted regression). The statistical analysis was conducted with ANOVA t-test. (B) Six miRNA expression patterns in C. suppressalis. Highly expressed miRNAs with signal value more than 500 at more than one time point are shown. The miRNA abundance was normalized to the highest value.

FIGURE 3.

qRT-PCR validation of 10 randomly selected miRNAs. Three biological replicates and three technical replicates were performed. The values are the average ± SD. The microarray signal values are also shown to compare with qPCR results, suggesting the high reliability of microarray analysis.

FIGURE 4.

The miRNA target prediction and enrichment analysis. (A) KEGG pathway analysis of 1351 target genes of 54 differentially expressed miRNAs, showing that pathways of signal transduction were specifically enriched. (B) Abundant genes enriched in some signal transduction pathways such as insect hormone biosynthesis. (C) Ten miRNAs were predicted to target the three Halloween genes. Five software packages (miRanda, TargetScan, RNAhybrid, Microtar, and PITA) were used to predict miRNA target. The genes predicted by at least four of the software packages were kept for further analysis.

FIGURE 5.

In vitro dual luciferase reporter assays of miRNA–mRNA interactions in C. suppressalis. The mean ± SEM of the relative luciferase expression ratio (firefly luciferase/Renilla luciferase, Luc/R-luc) was calculated for three biological replicates, and compared with the negative control (NC), miRNA mimics treatment. All data were analyzed with Dunnett's multiple comparison after an ANOVA ([***] P < 0.001). (A) Dual luciferase reporter assays of Csu-miR-9b and Csu-novel-239, showing that Csu-miR-9b can target at Csu-nvd. (B) Dual luciferase reporter assays of Csu-novel-260 confirmed its miRNA–mRNA interaction relationship between Csu-novel-260 and CsuDib. (C) Dual luciferase reporter assays of Csu-Bantam, Csu-miR-8, Csu-novel-80, Csu-novel-89, Csu-novel-124, Csu-novel-154, and Csu-novel-257, showing that five miRNAs except for Csu-miR-8 and Csu-novel-124 can efficiently target CsuSpo. (D) The miRNAs target sites predicted in the CsuNvd, CsuSpo, and CsuDib of C. suppressalis. Five miRNAs were confirmed to interact with CsuSpo.

FIGURE 6.

The overexpression of seven miRNAs shows that these miRNAs control the ecdysteroid biosynthesis. (A) The abundance of the seven miRNAs was significantly elevated 24 h after injecting the agomir mimics on day 4 of the sixth-instar larvae. (B) A Western blot analysis confirmed that the protein level of CsuSPOOK was significantly reduced at 24 h after injection with miRNA mimics. A slightly weak band with a molecular weight of 61.3 kDa was detected with a polypeptide antibody, and actin was used as the control. (C) The determination of the 20E titer at 24 and 48 h after injection with the miRNA mimics, showing that the 20E titer was significantly reduced in the presence of the miRNA mimics. (D) The miRNA mimic treatment induced high mortality compared with that of the control. The rescue experiments treated with 0.25 ng 20E successfully reduced the mortality, suggesting that the high mortalities were caused by reduced ecdysteroids. (E) The pupation rates of miRNA-mimics-treated larvae were significantly reduced, which can be rescued by 20E.

FIGURE 7.

Morphological traits of agomir-treated individuals at moulting and pupation at 144 h after injection of mimics. (A) Mimics-treated groups showed abnormal development. (B) Rescue groups of Csu-novel-260 and five miRNA mixture that were treated with 0.25 ng 20E showed normal development. However, the malformations in the Csu-miR-9b group were not well rescued by 20E. (C) Abnormal prepupae of C. suppressalis after treatment with Csu-miR-9b mimics. (D) The abnormal prepupae with an abdomen of dehydration but a “larvae” head as indicated by a black arrow. (E) The negative control treated with agomir with a random shuffled sequence.

FIGURE 8.

The mortality and phenotype observation of tested individuals treated with antagomir of miRNAs. (A) Mortality rate of C. suppressalis larvae after injection with 100 pmol antagomir-9b on L5D3, suggesting that the knockdown of Csu-miR-9b induced a slightly high mortality. (B) Mortality rate after treatment with a mixture of five antagomir inhibitors on L6D2. The joint knockdown of five miRNAs did not induce a high mortality compared with that of the control. (C) The eclosion rate of C. suppressalis pupae after injection with 100 pmol antagomir-260 inhibitor on the second day of pupation. The eclosion rate was slightly reduced by the antagomir. (D) The development phenotype of antagomir-treated individuals, showing that knockdown of miRNAs did not induce apparent developmental defects compared with the control.

FIGURE 9.

The expression patterns of seven miRNAs as estimated by microarray and qRT-PCR. (A) The expression of miRNAs and their target genes determined by microarray, showing that the miRNAs were generally coexpressed with their target genes. (B) The expression profiles of seven miRNAs at different developmental stages from every day of the fifth-instar larvae, the sixth-instar larvae, and pupae in C. suppressalis. The values are expressed as the mean ± SEM determined by three biological replicates. All data were analyzed with Dunnett's multiple comparison with (*) at 0.05 and (**) at 0.01 significance level. The results indicated that these seven miRNAs have different expression patterns.

FIGURE 10.

The model of multiple miRNA-mediated regulation of ecdysteroid biosynthesis, which controls the moulting and pupation in striped rice stem borer. Csu-miR-9b might block the cholesterol dehydrogenation by targeting CsuNvd, but Csu-novel-260 controls the hydroxyl group addition by targeting CsuDib, and five miRNAs control the conversion of dehydrocholesterol to diketol by jointly targeting CsuSpo.