Nucleases as a barrier to gene silencing in the cotton boll weevil, Anthonomus grandis

Abstract

RNA interference (RNAi) approaches have been applied as a biotechnological tool for controlling plant insect pests via selective gene down regulation. However, the inefficiency of RNAi mechanism in insects is associated with several barriers, including dsRNA delivery and uptake by the cell, dsRNA interaction with the cellular membrane receptor and dsRNA exposure to insect gut nucleases during feeding. The cotton boll weevil (Anthonomus grandis) is a coleopteran in which RNAi-mediated gene silencing does not function efficiently through dsRNA feeding, and the factors involved in the mechanism remain unknown. Herein, we identified three nucleases in the cotton boll weevil transcriptome denoted AgraNuc1, AgraNuc2, and AgraNuc3, and the influences of these nucleases on the gene silencing of A. grandis chitin synthase II (AgraChSII) were evaluated through oral dsRNA feeding trials. A phylogenetic analysis showed that all three nucleases share high similarity with the DNA/RNA non-specific endonuclease family of other insects. These nucleases were found to be mainly expressed in the posterior midgut region of the insect. Two days after nuclease RNAi-mediated gene silencing, dsRNA degradation by the gut juice was substantially reduced. Notably, after nucleases gene silencing, the orally delivered dsRNA against the AgraChSII gene resulted in improved gene silencing efficiency when compared to the control (non-silenced nucleases). The data presented here demonstrates that A. grandis midgut nucleases are effectively one of the main barriers to dsRNA delivery and emphasize the need to develop novel RNAi delivery strategies focusing on protecting the dsRNA from gut nucleases and enhancing its oral delivery and uptake to crop insect pests.

Fig 1. In silico analysis of candidate nucleases identified within the CBW gut.

(A) Amino acid alignment of CBW nucleases with those from other insects and human sugar non-specific nucleases (D. melanogaster, B. mori, S. gregaria, Mus musculus, B. taurus, H. sapiens, and S. cerevisiae). The asterisks indicate the conserved motif Hx₂₂Nx₄Qx₅Nx₈E, which is involved in metal binding and catalysis. Identical and similar amino acids are highlighted in green and yellow, respectively. (B) CBW nucleases showing a C-terminus DNA/RNA sugar non-specific endonuclease domain and an N-terminus signal peptide. The black arrows and black cylinders represent the predicted β-strand and α-helix, respectively. “SP” is the predicted secretory peptide, and “TM” is a predicted transmembrane domain. The accession numbers refer to the UniProt protein data bank. (C) Maximum-likelihood phylogenetic tree of the non-specific endonuclease family divided by different animal species. Groups I and II are included in the red and blue areas, respectively. The red disks indicate the prediction of a secretory peptide (>0.7) at the N-terminal section of the nucleases. The accession numbers correspond to the UniProt or NCBI databases.

Fig 2. Biochemical characterization of CBW gut juice.

(A) CBW gut juice (GJ), which is able to degrade both dsRNA, ~ 200bp, and dsDNA, > 5000 bp (as observed), has non-specific nuclease activity. MM: Molecular Marker 1-Kb Plus DNA ladder (Invitrogen); GJ: Gut Juice. Samples were incubated with GJ for 30 minutes at 37°C. (B) The optimal pH for nuclease activity ranges from 5.5 to 6.5, indicating that the nucleases function best at acidic pH.

Fig 3. RT-qPCR analysis of CBW nuclease expression at different developmental stages.

(A and B) CBW was dissected to obtain the gut and carcass, and nuclease expression was then measured in these samples. The bar chart shows that AgraNuc1 expression is similar in the gut and carcass of the adult (A) and larvae (B), whereas AgraNuc2 and AgraNuc3 are highly expressed in the gut only. (C and D) The insect gut was sectioned into the anterior midgut (AMG), posterior midgut (PMG) and posterior gut (PG), and the expression levels of the nucleases in these sections were evaluated. Higher expression of AgraNuc2 and AgraNuc3 was observed in the PMG of both adults (C) and larvae (D), whereas AgraNuc1 expression was similar in all gut sections. Agra-β-actin and Agra-β-tubulin were used as reference genes. The relative expression (UA) was calculated based on the lowest expression value that was obtained. Statistical analyses of the average transcripts expression levels were performed using Tukey’s test with a 0.05% significance level for comparisons between treatments.

Fig 4. Analysis of CBW nucleases two days after gene silencing by RT-qPCR and dsRNA digestion assay.

(A) Insect microinjection was performed with 500 ng of dsRNA against each nuclease and a mixture of all three dsRNAs (in a total of 1500 ng of dsRNA) and the analysis was performed two days after the microinjection. dsRNA against gus was used as a negative control, and Agra-β-actin and Agra-β-tubulin were used as reference genes. The relative expression (UA) was calculated based on the lowest expression value that was obtained. Statistical analyses of the average transcripts expression levels were performed using Tukey’s test with a 0.05% significance level for comparisons between treatments. The bar chart shows that the expression of the nucleases, including each individual nuclease and all three nucleases together, was silenced. (B) dsRNA (~ 200 bp) was incubated with CBW gut juice (GJ) for 30 minutes at 37°C. GJ was collected two days after RNAi nuclease gene silencing, and 1% agarose gel electrophoresis was performed to analyze dsRNA digestion. GJ was collected from uninjected insects and from injected insects with all three nucleases silenced at once. GJ: Gut Juice, KD: knocked down, WT: wild type, CBW: cotton boll weevil.

Fig 5. Analysis of CBW ChSII gene expression after nuclease gene silencing.

Two days after microinjection of the nuclease dsRNA into the CBW body cavity, which silenced the AgraNuc genes, the insect was starved for two days, and 500 ng of AgraChSII dsRNA was orally administered. The insects with silenced nucleases (fourth bar) showed a decrease in AgraChSII transcript expression compared with the control insects (first, second and third bars). RNA extraction, cDNA synthesis and RT-qPCR were performed with the whole insect. dsRNA against gus was used as a negative control, and Agra-β-actin and Agra-β-tubulin were used as reference genes. The relative expression (UA) was calculated based on the lowest expression value that was obtained the average transcripts expression levels were performed using Tukey’s test with a 0.05% significance level for comparisons between treatments.