Reagents and Chemicals
The pure analytical grade chemicals of s-methoprene and PBO-8 were obtained from Dow Agrosciences Ltd (CPC2 Capital Park, Fulbourn, Cambridge, England, CB21 5XE). In addition, the commercial formulations of these two chemicals, such as Diacon IGRTR and PBO8-Synergist, containing 33.6% active ingredient (a.i) of s-methoprene and 91.3% of PBO, respectively, were used for the tests.
Insect strains
Two strains of R. dominica, which are susceptible (Lab-S) and resistant to s-methoprene (Met-R), respectively, were used in this study. The Lab-S was collected form a grain storage shed at Oakey in 1971, whereas, Met-R strain was collected from Roma, Queensland [26]. Since then, the insect populations of these two strains were reared on whole wheat kernels and maintained at standard room temperature and relative humidity (RH) at Queensland Department of Agriculture and Fisheries (QDAF), Australia. Adult beetles less than three weeks old were randomly selected from the culturing jars and used in the bioassays.
Laboratory bioassays
Two sets of s-methoprene concentrations were used in the bioassays. These include, 0, 0.01, 0.03, 0.1 and 0.3 mg kg-1 for Lab-S and 0, 1, 3, 10 and 30 mg kg-1 for Met-R. For PBO bioassay, the wheat grains were applied with a recommended label rate for combinations, 0.013 lt per 45.3 kg of wheat. Untreated clean and infestation-free wheat grains were used for treatments. The moisture content of the grain was adjusted to 13.5% before the initiation of the experiment. Three lots of wheat containing 2 kg each were sprayed with different dose rates of s-methoprene alone, PBO alone and the combination of s-methoprene + PBO; hence, the combinations of the formulations were: a) s-methoprene alone, b) PBO alone and c) s-methoprene with PBO, in all possible combinations (control, 0.01 mg/kg, 0.03 mg/kg, 0.1 mg/kg, 0.3 mg/kg, PBO, 0.01 mg/kg + PBO, 0.03 mg/kg + PBO, 0.1 mg/kg + PBO, 0.3 mg/kg + PBO for Lab-S and control, 1 mg/kg, 3 mg/kg, 10 mg/kg, 30 mg/kg, PBO, 1 mg/kg + PBO, 3 mg/kg + PBO, 10 mg/kg + PBO, 30 mg/kg + PBO for Met-R). The required volume of each treatment including the combinations, was applied using a specialized airbrush (Badger 100, Kyoto BD-183 K Grapho-tech, Japan). An additional Series of 2 kg wheat lots, sprayed with water in parallel to each treatment, was used as untreated control. Twenty grams (20 g) of grain samples from each treatment was selected for bioassays. This sample was placed inside cylindrical bioassay vials (3 cm in diameter, 8 cm in height). Ten adults of R. dominica were released into each vial. The vials were then placed in incubators set at 27.5oC and 75% RH. The mortality of adults was recorded after 7, 14 and 21 days of exposure. Thereafter, all parental adults were removed from the vials, and the vials with the treated grain were returned to the same incubators and maintained for 65 d more to ensure that the immatures in the treated grain will develop up to the adult stage. Then, the number of adults that emerged in treatments and control were compared and per cent reduction in progeny production was estimated. The entire experiment was repeated three times (jars) with each containing 3 subreplicates (vials).
RNA isolation, library construction sequencing and qPCR validation
Ten 2nd to 3rd instar larvae of R. dominica of Lab-S and Met-R strains were pooled respectively and preserved in RNA later, and total RNAs of each was extracted using the GeneJet RNA Purification kit (ThermoScientific), according to the manufacturer’s protocol. The extracted RNA was treated with Turbo DNase (Ambion), in order to remove any traces of genomic DNA. The purity and concentration of RNA were estimated using a Nanodrop spectrophotometer based on 260/280 and 260/230. RNA samples were sent to Macrogen (Korea) for mRNA paired-end library construction with the Illumina Truseq stranded mRNA sample preparation kit, following the manufacturer’s instructions. Each library was sequenced with the paired-end method for a read length of 100 bp. Two μg of RNA was used for cDNA synthesis using the reverse transcriptase kit from Minotech (Heraklion, Greece), according to the manufacturer’s instructions. qPCR validation was conducted for a subset of genes. The primers used are shown on Table S3. Briefly, a 5-fold dilution series of pooled cDNA was used to assess the efficiency of the qPCR reaction for each gene-specific primer pair. A no template control (NTC) was also included to detect possible contamination. The reactions consisted of 0.6 μM primers each, and Kapa SYBR FAST qPCR Master Mix (Kapa-Biosystems). Experiments were performed using 3 biological and 3 technical replicates for each gene. The levels of the validated genes were measured by Real-time qPCR (RT-qPCR) amplification on a CFXConnect (BioRad). Relative expression levels were calculated as previously described [63].
Computational analyses
RNAseq reads from both strains (total of ~688 million reads) were assembled with Trinity v2.5.1 [64], using parameters “--seqType fq --SS_lib_type RF --max_memory 350G --CPU 24”. InterProScan v5.28-67 [65] was used in order to identify conserved domains within each assembled transcript. Moreover, BLAST v2.8.0+ [66] searches were run in order to identify similarities using the Uniref50 database that is specifically built for similarity-based functional annotation [67].
Transcript abundance was estimated with Kallisto [68]. Next, the scripts bundled with Trinity were used for running the differential expression analysis with EdgeR [69] in order to find transcripts that were differentially expressed between the resistant and the susceptible strain (FDR <0.05). Custom Perl and bash scripts were used for parsing the EdgeR output and identifying genes of interest. Gene Ontology (GO) term analyses were done using gProfiler [70].
For the detection of polymorphisms in the methoprene tolerant gene we firstly mapped the raw reads to the Trinity transcripts using hisat2 [71], then generated a mpileup file with samtools [72], and searched for SNPs with VarScan v2.4.4 [73]. Finally, the identified SNPs were visually inspected across the extracted methoprene tolerant transcript using samtools and the data were loaded into the Integrative Genomics Viewer v2.6.3 [74].
In order to identify transcripts with similarity to cytochrome P450 (CYP) genes we first ran the TransDecoder program that is bundled with Trinity v2.8.5 [75] and obtained the encoded peptides in each transcript. Subsequently, putative CYP-related proteins were identified by the presence of the IPR001128 InterPro domain, in the InterProScan output file. The curated CYPs set identified in T. castaneum were obtained from [48] and used as a reference for classifying the herein identified R. dominica CYPs. Finally, the early-diverged CYP51A1 [76] from Homo sapiens was used as an outgroup. Multiple sequence alignment was performed with MAFFT v7.271 [77] with parameters “--auto --threads 8” and trimming was done with Trimal v1.2rev59 [78], with parameters “--gt 0.50”. A Maximum Likelihood phylogeny with 100 bootstrap replicates was inferred with RAxML v8.2.11 [79], with parameters “-m PROTGAMMAAUTO”. Branches with <50% bootstrap support were collapsed with TreeGraph2 [80] and the resulting Newick tree was loaded to a locally deployed instance of EvolView v2 [81] for post-processing. The vector graphics editor Inkscape v0.92 was used for the final polishing.
Bioassay data analysis
The data of progeny production were analyzed separately for each strain using ANOVA to test the treatment effects. When preliminary tests indicated that variances were not equal, the data were transformed to log (x+1) (for the susceptible strain O’Brien test: F=1.01, P=0.437; for the resistant strain O’Brien test: F=1.84, P=0.073). Means were separated by using the Tukey-Kramer HSD test at the 5% level. For each strain the Student's t-test was used to determine differences between s-methoprene alone and s-methoprene with PBO. Statistical analysis was performed by using the JMP 7 software (SAS Institute Inc., Cary, NC, USA).