The wild-type GA-1 strain of the red flour beetles, T. castaneum, were reared on organic wheat flour containing 10% yeast at 30°C. The final instar larvae were staged based on untanned white head phenotypes observed immediately after molting.

The fall armyworm, S. frugiperda, larvae were reared on artificial diet at 27 ± 1°C, 14 h:10 h (light: dark) photoperiod and 65 ± 5% relative humidity (RH). Each larva was placed in a separate cup. Pupae were sexed and kept in cages for adult eclosion. Pupae were checked daily for emergence and the moths were fed on 10% sugar water.

The gene encoding SoxC transcription factor from T. castaneum and S. frugiperda were identified by searching the non-redundant protein sequences (NR) database at NCBI using D. melanogaster SoxC gene (Flybase ID: FBgn0005612) protein sequence as a query. Candidate sequences for the SoxC gene in T. castaneum (XP_973116.1) and S. frugiperda (XP_035435714.1) were amplified by reverse transcription PCR (RT-PCR) using primers shown in Supplementary Table S1.

Total RNA was isolated from treated and control insects using TRIzol reagent (Thermo Fisher Scientific). The concentration of RNA was measured using a NanoDrop-2000 spectrophotometer (Thermo Fisher Scientific). cDNA was synthesized from 2 µg of total RNA samples using M-MLV reverse transcriptase (Invitrogen, Thermo Fisher Scientific) in a 20 µL reaction following the manufacturer’s protocol. Diluted (10×) cDNA, gene-specific primers and iTaq Universal SYBR Green Supermix (Bio-Rad, Hercules, CA) were used in reverse transcriptase quantitative PCR (RT-qPCR). All the primers used for RT-qPCR are shown in Supplementary Table S1.

To prepare the dsRNA of TcSoxC gene, a pair of specific primers with T7 polymerase promoter (5′-TAA​TAC​GAC​TCA​CTA​TAG​GG-3′) at their 5′-end were used to amplify a 437 bp fragment of the TcSoxC gene (Supplementary Table S1). The PCR products were purified using QIAquick PCR purification kit (QIAGEN, Hilden, Germany), which were then used as templates for dsRNA synthesis using the Invitrogen 5× MEGAscript T7 Transcription kit (Invitrogen, Thermo Fisher Scientific Inc., Vilnius, Lithuania). The dsRNA was treated with DNase I (Ambion, Austin, TX) and purified using a phenol/chloroform extraction followed by ethanol precipitation. Finally, the dsRNA was dissolved in nuclease-free water and checked for quality by electrophoresis on agarose gel. The concentration of dsRNA was measured using a NanoDrop-2000 spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA) to be 3–5 μg/μL. The control dsRNA targets the malE gene of Escherichia coli (Bai and Palli, 2010). All dsRNA products were stored in −20°C.

sgRNAs were designed based on the criteria: 5′-GG-(N)18-NGG-3'. sgRNA templates were produced (Supplementary Table S1) as described previously (Zhu et al., 2016; Zhu et al., 2020) using the MAXIscript T7 In Vitro Transcription Kit (Invitrogen) following the manufacturer’s instructions. The purified sgRNAs were quantified using a NanoDrop-2000 spectrophotometer (Thermo Fisher Scientific) and stored in −70°C. The Cas9 protein used in our study was purchased from Thermo Fisher Scientific (TrueCut Cas9 Protein v2, A36497, Invitrogen).

Newly molted final instar larvae were used for microinjection. 400 ng of dsRNA was injected into each larva on the dorsal side of the abdomen. The NANOJECT III (Drummond Scientific Co.) was used for dsRNA microinjection with an aspirator tube assembly fitted with 3.5″ glass capillary tube (Drummond Scientific Co.) pulled by a needle puller (Model P-1000, Sutter Instruments Co., Novato, CA). Injected larvae were allowed to recover for 1 h at room temperature and then transferred to a 30°C incubator under standard conditions. Control larvae were injected with dsRNA targeting E. coli malE gene.

The S. frugiperda adults were placed in a cage (L × W × H: 15 cm × 15 cm × 10 cm) that was covered with brown paper on the top and fed with 10% sugar water. The fresh eggs (laid within 1 h) were lined on a microscopic slide, and a mixture containing 300 ng/μL Cas9 protein (TrueCut Cas9 Protein v2, A36497, Invitrogen) and 100 ng/μL of each sgRNA was injected into less than 1 h old eggs using a microinjector (NARISHIGE IM300, Japan). To increase mutation efficiency, three sgRNAs of the same concentration were co-injected (Zhu et al., 2020). Injected eggs were kept in a sterile petri dish at 27 ± 1°C until hatching.

The mutagenesis was assessed using gene fragment sequencing and T7EI assay. Genomic DNA was isolated from individual insects using DNeasy Blood & Tissue Kit (QIAGEN) following the manufacturer’s protocol. PCR was carried out using the PrimeSTAR GXL DNA Polymerase (TaKaRa) in 50 μL reaction to amplify the target fragments from the genomic DNAs. PCR products was purified using QIAquick PCR purification kit (QIAGEN) and were cloned into TOPO PCR4 vector (Invitrogen, Thermo Fisher Scientific) following the manufacturer’s protocol. 10G competent cells were then transfected with the plasmid construct and positive clones were sequenced. Primers are shown in Supplementary Table S1.

To estimate the mutation rate, the T7EI assay was performed using the PCR products as described previously (Zhu et al., 2020). Briefly, a total 200 ng of PCR products in 1 × NEB buffer 2 were hybridized under the following cycles: 95°C for 5 min, 95–85°C at -2°C ^s−1, 85–25°C at −1°C ^s−1, and held at 4°C. Then, 10 units of T7EI (New England Biolabs) was added to the reaction and incubated at 37°C for 15 min. Finally, the reaction mixtures were run on 2% agarose gels containing Gel Red, and the gel electrophoresis results were photographed under a UV light. The image lab (Bio-lab) software was used for analyzing gel images. The percentage of mutagenesis was calculated using the method described previously (Zhu et al., 2020).

Technical grade stable ecdysone agonist (SEA), RH-102240 [N-(1,1-dimethylethyl)-N'-(2-ethyl-3-methoxybenzoyl)-3,5-dimethylbenzohydrazide] was dissolved to 100 μg/μL in acetone. Newly molted final instar larvae of T. castaneum were injected with 400 ng of TcSoxC dsRNA. When the injected individuals entered into quiescent stage (q-stage) and the compound eyes appeared, 2 µL of SEA solution was added on the dorsal side abdomen of the insects. The control insects were treated with the same volume of acetone. 12 h after the treatment, insects were sampled and total RNA was isolated. The expression of target genes was determined using RT-qPCR and the primers used are listed in Supplementary Table S1.

Phenotypes of both T. castaneum and S. frugiperda were observed during eclosion (pupal-adult metamorphosis). All the phenotypes were photographed using a Nikon stereomicroscope (SMZ745T, Nikon, Japan). Eclosion behavior of T. castaneum after TcSoxC RNAi was recorded at 28 ± 1°C using a Canon digital camera VIXIA HF G50 video recorder.

To determine if SfSoxC protein directly binds to the SfEH promoter, we performed electrophoretic mobility shift assay (EMSA) using a probe targeting the putative SoxC-binding motifs located within the SfEH promoter region. EMSA was performed using the LightShife Chemiluminescent EMSA Kit (Thermo Fisher Scientific) following the manufacturer’s protocol. All the primers used for probes are shown in Supplementary Table S1.

Using the PCR-based accurate synthesis (PAS) method, the SfSoxC gene ORF was cloned into the prokaryotic expression vector pCzn1 between the NdeI and the XbaI sites (Zoonbio Biotechnology, China). The recombinant plasmid pCzn1-SfSOXC was transformed into TOP10 E. coli strain for sequencing. After nucleotide sequence was determined, the recombinant plasmid pCzn1-SfSOXC was transformed into the Arctic-Express (DE3) E. coli strain (Zoonbio Biotechnology, China) for recombinant protein expression. Expression of the target gene was induced using IPTG (final concentration 0.5 mM). The recombinant histidine-tagged His-SfSOXC protein was detected in the supernatant and purified by affinity chromatography using a Ni-IDA-Sepharose Cl-6B Microspin column (Zoonbio Biotechnology, China). The recombinant protein was confirmed by western blot analysis.

To construct promoter-pG5luc vectors for the luciferase assay, the potential promoter sequences of SfEH, containing 20 bp homologous arms for the pG5luc vector digested with KpnI and HindIII, were amplified from genomic DNA using Prime STAR GXL DNA Polymerase (TakaRa). The amplified products were inserted into pG5luc vector using the Gibson assembly master mix (NEB) to yield the SfEH/P2574-pG5luc vector. The ORF of SfSoxC genes, containing 20 bp homologous arms for the pIEx-4 vector digested with NcoI and AscI, were amplified using Prime STAR GXL DNA Polymerase (TakaRa), and then cloned into NcoI and AscI digested pIEx-4 vector as described above, yielding pIEx-4-SfSoxC vector.

To determine whether SfSoxC activates SfEH promoter, luciferase assays were carried out in the Sf9 cells. Sf9 cells were seeded in 96-well culture plates at a density of 2 × 10⁵ cells per well and incubated at 27°C overnight. The cells were then transfected with 100 ng of constructs per well using 0.8 μL of CellFectin II reagent (Thermo Fisher Scientific) in 50 μL of Sf-900 II medium. At 4 h post-transfection, the medium was removed and replaced with 100 μL fresh medium. Luminescence was quantified using SpectraMax i3x (San Jose, CA) at 48 h post-transfection. Medium was removed and cells were washed with 100 μL of 1 × PBS, then 100 μL of ice-cold lysis buffer was added to each well. The cell plates were placed on a shaker for 10 min at room temperature. 20 and 10 μL of cell lysate were used for luciferase activity assay and protein concentration determination, respectively, as described previously (Chen et al., 2020).

Results are presented as means ± SE. Statistical analyses were conducted in SPSS version 24. Treatments were compared using unpaired t-test with variances treated separately. All tests were conducted by one-tailed test.

In the red flour beetle, after stable ecdysone agonist (SEA) treatment, TcSoxC, TcEH, TcCCAP and TcBur mRNA levels increased compared to those in control insects treated with solvent. Injection of dsSoxC before treatment with SEA reduced induction of EH, CCAP and Bur genes by SEA (Figure 1; Supplementary Figure S1). These data suggest that SoxC may be required for ecdysone regulation of the neuropeptide genes tested.

Injection of dsSoxC into newly molted last instar larvae triggered 75–91% knockdown of target gene in pupae and adults (Figures 2A,B). The dsSoxC-treated larvae progressed to the pupal stage but were unable to complete the transition from pupa to adult. Some of the pharate adults were stuck inside pupal integument. The tanning, as well as the expansion of the forewings (elytra) and hindwings, were abnormal in dsSoxC treated insects (Figure 2D) compared to those treated with control dsmalE (Figure 2C).

As expected, the expression of neuropeptide genes (EH, CCAP and Bur) decreased in dsSoxC treated insects compared to that in control insects treated with dsmalE (Supplementary Figure S2). Although the expression of ETHp gene also decreased in dsSoxC treated insects, the decrease in expression is not consistent in all stages tested. For example, the expression of ETHp increased in day 4 pupae in dsSoxC-treated larvae compared to the control insects (Supplementary Figure S2B).

Three sgRNAs targeting exon2 of SfSoxC were used to knockout SoxC gene in S. frugiperda (Figure 3A). A total of 216 eggs were injected with Cas9 protein and sgRNAs (Supplementary Table S2). 83% mutation rate was detected in 30 larvae tested. Some of the injected larvae (39) died during first instar larval stage (Supplementary Table S2). Thirty-nine larvae developed to pupal stage and died at the end of the pupal stage (Figures 3B-I,II). Forty-eight insects showed abnormal phenotypes during eclosion (Figure 3B-III; Supplementary Table S2). The pharate adults were stuck inside pupal cuticle and the expansion of wings was not complete (Figure 3B-III). To confirm the mutagenesis in SfSoxC gene induced by CRISPR/Cas9, T7EI assay was performed. PCR amplification of mutated sequences was carried out using genomic DNA isolated from eight randomly selected mutants. The mutant DNA was cut into three fragments by T7EI (Figure 3C-top) and only one band without T7EI treatment (Figure 3C-bottom). Furthermore, to determine the indels of SfSoxC gene, the PCR products were cloned and sequenced. Four different alleles displaying mutations in 15 sequenced clones were identified (Figure 3D). According to the T7EI assay, the heterozygotes develop an abnormal eclosion molt phenotype and there were no individuals for doing a backcross. Besides, some homozygotes died at 1^st instar larval and the others develop an abnormal eclosion molt phenotype (Figure 3B-III; Supplementary Table S1).

Expression of the four neuropeptide genes significantly changed after disrupting the SfSoxC gene (Supplementary Figure S3). At all four stages checked, EH and Bur expression decreased significantly (Supplementary Figures S3A,D), whereas the expression of ETHp increased significantly (Supplementary Figure S3B). The expression of CCAP gene slightly decreased at PP and PD0, and decreased significantly at PD7 and AD0 stages (Supplementary Figure S3C).

To test whether the development of insects is affected by SfSoxC gene knockout, the development time and pupa weight were recorded. In females, neither the development time nor pupal weight showed any significant difference between wild-type and mutants (Figures 4A,B,D). However, in males, significant differences in development time of 1st, 2nd, and 6th instar larvae and pupal weight were observed (Figures 4A,C,D).

To determine if SfSoxC directly binds to the conserved SoxC binding motif in the promoter of SfEH, we performed EMSA using a probe containing the SoxC-binding motif sequences found in the promoter of SfEH (Figure 5A). A specific shifted band was detected in the EMSA assay when the biotin-tagged probe was mixed with SfSoxC protein (Figure 5B, lane 4). This band was not detected when the biotin-tagged mutated probe and SfSoxC protein were used in the assay (Figure 5B, lane 8). Competition assays showed that the binding of SfSoxC to the SfEH promoter is specific (Figure 5B, lane 5 and 6). These results show direct binding of SfSoxC protein to the SoxC binding motif in SfEH gene.

The SfSoxC gene is a transcription factor. To confirm that SfSoxC directly activate the expression of SfEH, we performed the luciferase reporter assays in Sf9 cells. The activity of luciferase gene expressed under the control of SfEH promoter increased in cells overexpressing SfSoxC, suggesting that SoxC binds to its target site in the EH promoter and activates transcription (Figure 6).