Alternative splicing (AS) promotes phenotypic plasticity to adverse conditions by altering the transcripts of genes directly involved in adaptation to those conditions. However, whether changes in AS occur in a few, select genes or at a larger scale is unclear. Agricultural insect pests are regularly exposed to xenobiotic stress from insecticide applications and thus develop resistance. Here, we used a genome-wide approach to show that the fruit pest Drosophila suzukii, resistant to multiple insecticides, exhibits aberrant splicing in many classes of genes, including those involved in general stress response and processes previously implicated in insecticide resistance, such as metabolic and cuticular pathways. Our results suggest that resistance is not simply a consequence of broad, transcriptome-wide increases in AS events. Instead, the specific classes of genes that are differentially spliced are better determinants of insecticide resistance. Furthermore, we observe that sublethal insecticide exposure promotes AS events even in the absence of substantial differential gene expression. Our proposed model underlying the development of insecticide resistance in D. suzukii aligns with the genetic accommodation framework, in which repeated insecticide exposure alters the regulation of environmentally induced AS events, ultimately leading to genetic assimilation. This leads to elevated AS of specific transcripts, a phenomenon normally induced by insecticide exposure, becoming genetically encoded and constitutively expressed, even in the absence of insecticide exposure. In summary, this study provides insights into the role of AS in enabling insects to diversify genome function to survive acute insecticide treatment and to develop xenobiotic resistance upon prolonged exposure.

RNA extraction and Library Preparation for Illumina short-read (PE150) sequencing and PacBio long-read sequencing

Lines used in this study:

Entrust_Resis1 & Entrust_Resis2 = Isofemale Lines Resistant to Entrust
Entrust_Sus1 & Entrust_Sus2 = Isofemale Lines Susceptible to Entrust
MustangMaxx_Resis1 & MustangMaxx_Resis2 = Isofemale Lines Resistant to Mustang Max
MustangMaxx_Sus1 & MustangMaxx_Sus2 = Isofemale Lines Susceptible to Mustang Max
Malathion_Resis1 & Malathion_Resis2 = Isofemale Lines Resistant to Malathion
Malathion_Sus1 & Malathion_Sus2 = Isofemale Lines Susceptible to Malathion

RNA was extracted from female flies that were entrained at 25°C in 12-hour light:12-hour dark cycles for two full days. On day 3, flies were collected on dry ice at ZT16, which is 16 hours after lights-on. This time point was selected because D. suzukii exhibits a low level of cytochrome P450 expression at this time, enabling any overexpression to be more readily observed. Fly bodies were separated from heads using frozen metal sieves (Newark Wire Cloth Company, Clifton, NJ, USA). Eight to ten female bodies were homogenized using a Kimble® pellet pestle® Cordless Motor (DWK Life Sciences, Rockwood, TN, USA) and Kimble™ Kontes™ Pellet Pestle™ (DWK Life Sciences) in 300 mL of TRI reagent (Sigma, St. Louis, MO). 60 mL of 100% chloroform (Sigma) was added to the homogenized tissue and incubated at room temperature for 10 minutes. The upper aqueous layer was recovered after spinning down and transferred to a new microcentrifuge tube. RNA was precipitated with an equal volume of 100% isopropanol at −20°C overnight. After spinning down, the RNA pellet was washed with 70% ethanol once and allowed to air dry. The pellet was then resuspended in 20 μL 1X Turbo DNA-free kit buffer (Thermo Fisher Scientific, Waltham, MA, USA) and treated with Turbo DNase per manufacturer’s instructions. RNA quality was assessed with the Agilent 2100 Bioanalyzer system (Agilent Technologies, Santa Clara, CA, USA) and the Qubit RNA IQ kit (Invitrogen, Waltham, MA, USA) on a Qubit 4 Fluorometer (Invitrogen). RNA purity was measured on a Nanodrop 1000 (Thermo Fisher Scientific), and RNA quantity was measured with the Qubit RNA HS (high sensitivity) assay kit (Invitrogen).

Illumina short-read sequencing libraries were prepared with 1 μg of high-quality RNA and the TruSeq Stranded mRNA Library Prep Kit (Illumina, San Diego, CA, USA) according to the manufacturer’s protocol. A total of twenty-four libraries were prepared: three biological replicates of two Malathion-resistant lines and two susceptible lines treated with either the acetone vehicle or 20ppm Malathion. Library insert size and quality were measured with the Agilent 2100 Bioanalyzer System, and library concentration was measured with the Qubit 4 Fluorometer. All libraries generated from D. suzukii lines treated with the acetone vehicle were pooled together, and all libraries generated from D. suzukii lines treated with the sublethal dosage of Malathion were pooled together, such that there were twelve libraries per pool. Pooled samples were sent to Novogene (Sacramento, CA, USA) for sequencing on the HiSeq X Ten platform (Illumina) using PE150. Entrust and malathion sequencing data is form Tabuloc et al. 2024 (SciRep).

For PacBio Iso-Seq, high-quality RNA was sent to the DNA Technologies and Expression Analysis Core Laboratory at UC Davis for library preparation and sequencing. A total of 48 libraries were sequenced on five 8M SMRT cells on a PacBio Sequel system (PacBio, Menlo Park, CA, USA). Each of the five SMRT cells consisted of one of the following pooled libraries: (1) all Entrust resistant and susceptible lines, (2) all Mustang Maxx resistant and susceptible lines, and (3-5) 1 of 3 biological replicates of vehicle- and Malathion-treated D. suzukii that are either resistant or susceptible to Malathion.

Alternative splicing analysis

Using Salmon (v. 1.10.1), cleaned Illumina RNA-Seq reads were mapped to the Drosophila suzukii genome assembly (RefSeq: GCF_013340165.1) and gene expression was quantified using default parameters. Differential expression of gene isoforms and splicing events were determined using SUPPA2 (v.2.3). Specifically, splicing events on the updated annotation were determined with generateEvents. Using the option ‘-e {SE,SS,MX,RI,FL}’, we opted to call exon skipping (SE), alternative splice sites (SS), mutually exclusive exons (MX), intron retention (RI), and first/last exons (FL). Relative abundance of each splicing event (psi) was determined with psiPerEvent. Finally, differential isoform expression was assessed with diffsplice using classical mode, using the Susceptible line as reference for Resistant vs. Susceptible and Sus_1 as reference for Susceptible vs. Susceptible comparions. Unless otherwise stated, default parameters were used for all analyses done with SUPPA2.

Features with FDR < 0.05 were considered differentially expressed. The number of significant differentially spliced events for each group (either Resis1 versus pooled susceptible lines, Resis2 versus pooled susceptible lines, or Sus_1 versus Sus_2) was divided by all detected events and multiplied by 100 to get the percentage of significant differentially spliced events that occurred. Similarly, the number of genes that have at least one significant splicing event was divided by all detected genes and multiplied by 100 to get the percentage of significant differentially spliced genes that are present. The values between each resistant line and the pooled susceptible group for each insecticide were compared using Fisher’s Exact Test followed by Benjamini and Hochberg’s multiple comparison test on R. Both the values of the Fisher’s Exact Test and the corrected q value are provided throughout the manuscript. Outlier values were determined with ROUT (Q=0.1%) on GraphPad Prism (v11.0.2).

Functional enrichment analysis

Gene Ontology (GO) enrichment of genes was performed using ShinyGO 0.76.3. GO terms and processes were considered enriched if the false discovery rate (FDR) < 0.05. GO categories for the enrichment performed on differential splicing data were divided into categories only enriched in the resistant line comparisons (either Resis1 versus pooled susceptible lines or Resis2 versus pooled susceptible lines), only in the susceptible line comparison (Sus1 versus Sus_2), or overlap between the two comparisons.

Differential expression gene (DEG) analysis

Differential expression gene analysis for the Malathion-resistant and -susceptible lines was performed using sequencing reads derived from Illumina short-read sequencing. First, rRNA reads were removed using SortMeRNA (v. 2.1). Adapters (ILLUMINACLIP parameters 2:30:10) and low-quality ends (LEADING: 10, TRAILING:10, MINLEN:36) were trimmed using fastp (v. 0.24.0). Cleaned reads were aligned to our Drosophila suzukii genome reannotation based on the LBDM_Dsuz_2.1.pri assembly (accession no. GCF_013340165.1) using STAR (v. 2.7.11b). Count data from STAR (--quantMode GeneCounts) served as input in the DESeq2 package in R to perform differential expression analysis on each resistant line vs both pooled susceptible lines. Each Malathion-resistant line was compared to pooled susceptible lines separately as each resistant line might exhibit resistance due to different mechanisms.

Additional comparisons were performed to assess whether exposure to a sublethal dosage elicits a response at the gene expression level. Thus, the acetone vehicle was compared to the Malathion treatment. Both susceptible lines were pooled together in these comparisons since they were combined in the resistant to susceptible comparisons. For each comparison, genes with fold change differences between either resistant vs susceptible populations of vehicle vs insecticide treatment with a Benjamini-Hochberg adjusted p-value < 0.05 were considered differentially expressed. Correlation between biological replicates was calculated with Pearson’s correlation coefficient, which was determined with the ‘stats’ package in R (v. 4.2.1). Expression differences of key genes between the resistant and susceptible lines were calculated with two-way ANOVA followed by two-stage linear set-up procedure of Benjamini, Krieger, and Yekutieli on GraphPad Prism.

Data from: Changes in alternative splicing are associated with insecticide response and resistance

Data files

Abstract

Information for data tables

Data from: Changes in alternative splicing are associated with insecticide response and resistance

Data files

Abstract

README: Data from: Changes in alternative splicing are associated with insecticide response and resistance

Information for data tables

Methods