Cytosolic and mitochondrial ribosomal proteins mediate the locust phase transition via divergence of translational profiles

Significance Outbreaks of locust plagues are largely attributed to the phase transition from solitary to gregarious locusts. Many studies have demonstrated transcriptional and posttranslational regulation in phase change. However, the translational regulation in the locust phase transition is unclear. Here, we found plasticity in polysome profiles between solitary and gregarious locusts. A divergence with ribosomal proteins from cytoplasm and mitochondria modulates behavioral features of gregarious and solitary locusts. The findings reveal that the population density of locusts, as an environmental signal, can initiate translational regulation for the phenotypic plasticity of insects. Important clues in searching for targets to control pests are also provided. Insights into energy metabolism regulation at the translational level in eukaryotes are presented as well.


This PDF file includes:
Supporting text Figures S1 to S7 Tables S1 to S3

mRNA-seq
Total RNA was extracted using a TRIzol reagent kit (Invitrogen, USA). RNA quality was assessed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) and checked by RNase-free agarose gel electrophoresis. mRNA was enriched from total RNA with Oligo (dT) beads, and rRNA was removed by a Ribo-Zero™ Magnetic Kit (Epicenter, Madison, USA). The mRNA was fragmented into short segments and reverse transcribed into cDNA with random primers. Then, the cDNA fragments were purified by a QiaQuick PCR extraction kit (Qiagen, The Netherlands). End repair, poly (A) addition, and Illumina sequencing adapter ligation were performed for sequencing. The ligation products were selected on the basis of size using agarose gel and sequenced through Illumina HiSeq2500 by Gene Denovo Biotechnology Co. (Guangzhou, China).

RNA-binding protein immunoprecipitation assay
A RIP Kit (BersinBio, Bes5101) was used in this experiment. Total protein was extracted by TRIzol and separated into three parts for input, IgG, and RPL10A incubation. According to the manuals, 5 μg of RPL10A antibody was used in each reaction. After immunoprecipitation, the mRNAs in each reaction were collected by RNA extraction solution (phenol:chloroform:isoamyl alcohol=25:24:1, pH<5) and pelleted with ethanol. The RNA was dissolved in nuclease-free water and reverse transcribed by M-MLV Reverse Transcriptase (Promega, M1705). qPCR was performed with SsoFast EvaGreen Supermix (BIORAD, 1725202). The primers for qPCR are listed in the supplemental materials Table S3.

Bioinformatics analysis for mRNA-seq and Ribo-seq
We initially removed low-quality reads with the default parameters of the HTQC package (v-1.92.1) to compare gene expression between gregarious and solitary locusts (1), and then the remaining reads were mapped to the Locust genome (2) by HISAT2 (v-2.2.1) (3). The expression of genes was normalized by fragments per kilobase million based on unique mapping reads, and clustering analysis was performed using R. Differentially and transcriptionally expressed genes were detected with fold change ≥ 1.4 and P value < 0.05 as cutoffs, in which the P value and FDR were calculated using the DEGseq package (4-6).
The remaining Ribo-seq reads were aligned to the reference genome (2) using HISAT2 (v-2.2.1) (3) after removing the rRNA, tRNA, snRNA, and snoRNA. The unique mapping reads were used to calculate the length distribution of reads protected by ribosomes; the frequency distribution in the intergenic region of the genome; and the 5′ UTR, CDS, 3′ UTR, and intron region of the annotated gene. The unbiased metagene profiles were generated by counting 18-40-nt unique mapping reads at each position. Each read was represented by the first nucleotide mapping in the genome and normalized per million mapped reads (RPM). The TE was calculated by RPM in the CDS of Ribo-seq divided by RPM in the CDS of mRNA-seq (5,6). The significantly and differentially expressed genes at the translational level were defined using the same parameter by mRNA-seq. Clustering analysis of translational datasets was performed on TE.
We used the nine-quadrant diagram to identify genome-wide transcriptional and translational differences in gregarious and solitary locusts. There were five expression patterns, such as transcriptional changes alone, translational changes alone, opposite changes, homodirectional changes, and no changes. The opposite changes represent one gene with opposite changes at transcriptional and translational levels. The homodirectional changes represent one gene with consistent changes at transcriptional and translational levels.

UTRs annotated based on third-generation sequencing
Previous PacBio and Nanopore read (7) datasets were corrected and mapped to the locust genome by Minimap2 (8)

Fig. S1. (A)
Gene ontology enrichment of differentially expressed genes (DEGs) between solitary and gregarious locusts were analyzed in six different developmental stages from egg to adults, including egg, first to fifth instars, and adults. Red bar denotes the gene ontology enriched for the genes upregulated in gregarious locust, while blue bar denotes those downregulated in gregarious locust. X axis denotes the minus log-transformed P values (base 10) of the adjusted p value derived from the enrichment analysis (Fisher's exact test or 2-test). (B) Gene ontology enrichment of DEGs after solitarization (isolating gregarious locusts (IG)) and gregarization (crowding solitarious locusts (CS)) compared with controls. The IG and CS processes in both tissues brain (B) and thoracic (T) tissues were denoted as CS, IG, B_CS, B_IG, T_CS, and T_IG, respectively. The DEGs common to all the B_CS, B_IG, T_CS, and T_IG were denoted as "Common". The controls were denoted as "Brain and Thoratcic". X axis denotes the minus logtransformed P values (base 10) of the adjusted P value derived from the enrichment analysis (Fisher's exact test or 2-test).