Introduction
The legume pod borer, Maruca vitrata, is a major destructive insect pest of leguminous crops across tropical and subtropical regions (Srinivasan et al., Reference Srinivasan, Tamò and Malini2021). The larvae predominantly feed on flower buds, flowers, and immature pods, resulting in substantial yield losses ranging from 20% to 80% in crops such as pigeonpea, cowpea, mung bean, and soybean (Sharma and Franzmann, Reference Sharma and Franzmann2000). Conventional management strategies largely depend on the extensive application of broad-spectrum chemical insecticides. However, the increasing incidence of insecticide resistance, coupled with growing concerns regarding environmental contamination and human health risks, necessitates the development of innovative and sustainable pest management approaches (Ullah et al., Reference Ullah, Guru-pirasanna-pandi, Murtaza, Sarangi, Gul, Li, Chavarín-Gómez, Ramírez-Romero, Guedes, Desneux and Lu2025).
Recent advances in molecular approaches, particularly RNA interference (RNAi) and CRISPR/Cas9-mediated genome editing, have emerged as powerful tools for functional genomics research and sustainable insect pest management (Geerthana et al., Reference Geerthana, Suresh, Prabhakar, Rani, Thangaraj, Paramasivam, Kumar, Asokan and Karuppannasamy2025). In Maruca vitrata, RNAi-mediated silencing of critical genes such as α-amylase and tropomyosin has been shown to adversely affect larval survival and reproductive fitness, demonstrating the potential of gene-targeting strategies for pest suppression (Chatterjee et al., Reference Chatterjee, Yadav, Rathinam, Karthik, Chowdhary, Sreevathsa and Rao2022). Although RNAi technology enables transient suppression of gene expression, CRISPR/Cas9-based genome editing offers the advantage of precise and heritable genetic modifications (Boettcher and McManus, Reference Boettcher and McManus2015). Consequently, CRISPR/Cas9 has emerged as a promising next-generation genetic control platform for area-wide pest management strategies, including population suppression and precision-guided sterile insect technique (pgSIT) (Kandul et al., Reference Kandul, Liu, Buchman, Shriner, Corder, Warsinger-Pepe, Yang, Yadav, Scott, Marshall and Akbari2022).
Despite its potential, successful implementation of genome editing in non-model insect species requires rigorous validation of target genes and editing components prior to in vivo applications. For essential reproductive genes such as vitellogenin (Vg), in-vitro validation of single-guide RNAs (sgRNAs) provides a rapid and efficient non-transgenic approach to assess cleavage efficiency and functional suitability before undertaking embryo microinjection experiments. Such preliminary validation significantly reduces experimental time, minimises biosafety concerns, and circumvents regulatory challenges associated with the generation of stable transgenic insect lines. This approach is particularly relevant in countries such as India, where the field-level deployment of gene-edited insects remains under regulatory evaluation and biosafety scrutiny.
Vg gene encodes the precursor of vitellin, the major yolk protein essential for oocyte maturation, embryogenesis, and successful reproduction in insects (Tufail et al., Reference Tufail, Nagaba, Elgendy and Takeda2014). Functional disruption of Vg or its receptor has been shown to severely impair reproductive fitness in several insect species. For instance, CRISPR/Cas9-mediated knockout of the Vg gene in Plutella xylostella significantly reduced egg hatchability, highlighting the critical role of this gene in insect reproduction (Peng et al., Reference Peng, Wang, Zou, Qin, Vasseur, Chu, Zhai, Dong, Liu, He and Yang2020). Despite its biological significance, the Vg gene of M. vitrata has not been comprehensively cloned or evaluated as a target for genome editing applications.
Although several reproductive genes have been investigated as potential targets for genetic pest management, the Vg gene was selected in the present study because it plays a central role in yolk protein synthesis, oocyte maturation, embryonic development, and female fecundity. Alterations in Vg expression have been shown to adversely affect reproductive performance and population growth in insects. For example, sublethal exposure to bistrifluron significantly reduced Vg expression in Spodoptera exigua, resulting in decreased fecundity and reproductive fitness, further highlighting the importance of Vg as a promising molecular target for pest management (Hafeez et al., Reference Hafeez, Li, Yousaf, Khan, Imran, Zhang, Huang, Zhang, Shah, Wang and Fernández-Grandon2021).
Successful CRISPR/Cas9-mediated genome editing has previously been reported in Maruca vitrata, including targeted mutagenesis of the secretory phospholipase A2 (sPLA2) gene to investigate its roles in immunity, development, and reproduction, as well as deletion of insulin-like peptide genes to elucidate their functions in growth, metabolism, and reproductive development (Lee et al., Reference Lee, Jung and Kim2020; Al Baki et al., Reference Al Baki, Kyo Jung and Kim2021). However, CRISPR/Cas9-based studies targeting the Vg gene have not been reported in this species. Therefore, the present study focuses on the molecular characterisation of Vg, structural analysis of the Vg-vitellogenin receptor (VgR) interaction through molecular docking, and experimental validation of sgRNAs targeting Vg, thereby providing molecular resources for future genome-editing studies aimed at reproductive disruption and sustainable management in M. vitrata.
Materials and methods
Rearing of Maruca vitrata
The stock culture of M. vitrata was maintained at the Biopesticide Laboratory, Division of Basic Sciences, ICAR-IIHR, Bengaluru, Karnataka, India. The nucleus culture of M. vitrata was obtained from ICAR-National Bureau of Agricultural Insect Resources (NBAIR), Bengaluru, Karnataka, India, and maintained at ICAR-IIHR. A low-cost rearing protocol was adopted. The species identity was confirmed by assessing their taxonomic key, and additionally, molecular identification was carried out using the Universal primers LCO 1490 (5′-GGTCAACAAATCAAAAGATATTGG-3′) and HCO-2198 (5′ TAAACTTCAGGGTGACCAAAAAATCA-3′). A 1:1 sex ratio was maintained for mating. The insects were maintained in a separate insect-rearing room under controlled environmental conditions (Temperature: 25 ± 2°C, Relative humidity: 70 ± 10%, and Photoperiod: 14 hours of light and 10 hours of dark) on a natural host/artificial diet. M. vitrata wild-type flies were cultured in cuboidal wooden chambers, bound by a wired mesh on one side, a cloth with an opening on one side, and a fine plastic mesh on the remaining three sides, with a wooden floor. In total, 1:1 mixture of yeast and sugar and 1:5 diluted vitamin mixtures (local brand) in separate plastic cups were provided to the flies as food.
RNA extraction and cDNA preparation
The total RNA was extracted from a single female adult of M. vitrata using the TRIsureTM reagent (Bioline). In total, 50–100 mg of samples were homogenised in 1 ml of TRIsure using DEPC (Diethyl pyrocarbonate) treated mortar and pestle. For smaller quantities of tissue (1–10 mg), 800 μl of TRIsure was used. The homogenate was transferred to a 2 ml centrifugation tube, followed by centrifugation at 12000 rpm for 15 minutes at 4°C. Supernatant was collected into a fresh 1.5 ml tube. Chloroform (0.2 ml/ ml of TRIsure used) was added and shaken vigorously by hand for 15 seconds. Samples were incubated at room temperature for 3 minutes. Centrifugation at 12000 rpm for 15 minutes at 4°C formed three phases: an organic phase, an interphase, and a colourless upper aqueous phase containing RNA. The upper phase was taken without disturbing the interphase. RNA was precipitated by mixing cold isopropyl alcohol (0.5 mL/mL of TRIsure used). In total, 3M Sodium acetate solution (150 μl) was added, followed by incubation at room temperature for 10 minutes and centrifugation at 12000 rpm for 15 minutes at 4°C. The supernatant was discarded, and pellets were washed with 75% alcohol (1 ml/ ml of TRIsure). The samples were vortexed and centrifuged at 12000 rpm for 5 minutes at 4°C. Supernatant was discarded, and the pellets were air-dried. Pellets were dissolved in 20 μl of DEPC-treated water. The purity of the RNA was checked by Nanodrop (Thermo Scientific, USA), and the integrity was confirmed by denaturing agarose gel electrophoresis (1.5%) and visualised using ethidium bromide (0.5 μg/mL). The samples were temporarily and permanently stored at −80°C. A total of 1 µg total RNA was used for cDNA synthesis using the RevertAid First Strand cDNA synthesis kit (Thermo Fisher Scientific, USA) by following the manufacturer’s instructions (Incubation at 42°C for 1 hour 30 minutes, followed by 70°C for 5 minutes) and was stored at −20°C until further use. The cDNA conversion was confirmed by PCR amplifying the housekeeping gene, mtCOI, using gene-specific primers.
PCR amplification of vitellogenin cDNA fragments
Vg coding sequence (CDS) of M. vitrata was PCR amplified using 5 sets of primers (table 1) since the CDS is 5.3 kb long, and further obtained the full-length CDS sequence of Vg by generating contigs. PCR was performed in a thermal cycler (ABI-Applied Biosystems, Veriti, USA) using the amplicon-specific annealing temperatures. The total reaction volume of 25 μl contained approximately 0.2 µM of each primer, 10X LA PCR Buffer ll (Mg2 + free), 1.5 mM MgCl2, 0.25 mM of each dNTP, 50 ng of M. vitrata cDNA, and 0.25U of LA Taq DNA Polymerase (TAKARA BIO INC, USA). The PCR conditions used for Vg amplification are provided in table 2. The amplified products were resolved on 1.2% agarose gel. The PCR product was eluted using Nucleo-Spin Gel and PCR clean-up kit (Macherey-Nagel, Germany) and further ligated to pTZ57R/T cloning vector (Thermo Scientific, USA). The ligated products were cloned into Escherichia coli chemically competent cells, strain DH5α. Later, the transformants were identified using blue/white selection on a Luria–Bertani agar plate with ampicillin (100 mg/mL). Plasmids were isolated using a plasmid isolation kit (Thermo Scientific, USA), and the clones were run on 1.2% agarose gel and confirmed by differential mobility check and restriction endonuclease assay. The cloned plasmids were sequenced in triplicate by Sanger sequencing, and the sequence was confirmed by the BLAST programme of the National Centre for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/).
Primer details used for amplifying vitellogenin CDS from M. vitrata
Table 1 Long description
Columns are Sl. No., Primer Name, 5 to 3 Sequence, and Amplicon size in bp. Rows 1 and 2 list 1st fragment primers: FP sequence GAGCCACCACAAGAGACAC with amplicon size 1069, and RP sequence TGTTGTCCA AACGCTGTTG with no amplicon size shown. Rows 3 and 4 list 2nd fragment primers: FP sequence CTCCTGTACTCCATCACGCA with amplicon size 1140, and RP sequence ACGCACTTCATACTCTTCAGC with no amplicon size shown. Rows 5 and 6 list 3rd fragment primers: FP sequence AGTCTCCAAGTGCTGTCACT with amplicon size 1216, and RP sequence TGACTGGTGAGGGTGTTCAT with no amplicon size shown. Rows 7 and 8 list 4th fragment primers: FP sequence GACATCAACGGAGAGGTGC with amplicon size 1206, and RP sequence TGAGTTCACATCTGTGTACCAG with no amplicon size shown. Rows 9 and 10 list 5th fragment primers: FP sequence CGAAGAGTTGAGCCCTGC with amplicon size 1346, and RP sequence CCTCCACCATCTTAGGGTTC with no amplicon size shown.
PCR conditions for the amplification of vitellogenin CDS
Table 2 Long description
PCR conditions are organized by stage, with columns for Stage, Conditions, and Cycles. Initial denaturation is 98 degrees Celsius for 10 seconds, with 1 cycle. Denaturation is 98 degrees Celsius for 5 seconds, with no cycle count shown. Annealing is listed as 57 degrees Celsius for 30 seconds for fragment 1, 57 degrees Celsius for 30 seconds for fragment 2, 58 degrees Celsius for 30 seconds for fragment 3, 57 degrees Celsius for 30 seconds for fragment 4, and 57 degrees Celsius for 30 seconds for fragment 5, with 35 cycles. Extension is 68 degrees Celsius for 1 minute 30 seconds, with 1 cycle. Final extension is 72 degrees Celsius for 20 minutes, with no cycle count shown.
Phylogenetic analysis
The fragments of Vg CDS were assembled using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). Vg sequences from other insect species were retrieved from the NCBI database. Sequences were aligned using ClustalW with default settings in BioEdit (Version 7.2.5). The molecular phylogenetic tree was generated in MEGA 11 (Tamura et al., Reference Tamura, Stecher and Kumar2021) using the Maximum Likelihood (ML) method with complete gap deletion and resampled with 1000 bootstrap replications.
Signal peptide and protein domain prediction
SignalP 5.0 was used to predict the signal peptide (https://services.healthtech.dtu.dk/services/SignalP-5.0/). The Simple Modular Architecture Research Tool (SMART) (https://smart.embl-heidelberg.de/) was used to analyse the functional and conserved domains of the Vg protein. The obtained domains were compared with the related insect species domains. The amino acid sequence of target genes was obtained using the ExPASy translation tool (https://web.expasy.org/translate/).
Designing of sgRNA for LPD_N domain and signal peptide sequence
The exon sequences of target genes were used as an input to design the sgRNA. The sgRNA was designed in the first exon (LPD_N domain). The target sites were selected based on the criteria: 5′-GGN-18nts-NGG 3′ (Hwang et al., Reference Hwang, Fu, Reyon, Maeder, Tsai, Sander, Peterson, Yeh and Joung2013; Ashok et al., Reference Ashok, Bhargava, Asokan, Pradeep, Kennedy, Manamohan and Rai2023). The sgRNAs used in the present experiment were designed using the CRISPOR tool available at (http://crispor.tefor.net/) (Concordet and Haeussler, Reference Concordet and Haeussler2018). Based on the genome variants, sgRNAs with minimised off-target potential and high on-target scores were selected. The secondary structure of the selected single-stranded RNA was verified using the RNAfold web server (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi). The sgRNAs were designed using two primers, namely CRISPR_F and CRISPR_R. The forward primer (CRISPR_F) was initially designed with linker and T7 promoter sequence, followed by guide sequence and guide RNA scaffold. The Universal reverse primer (CRISPR_R) contains the complementary sequence of the guide RNA scaffold (table 3). These primers were used for a standard PCR reaction with Takara LA Taq to generate a DNA template. It was purified using PCR clean-up kit (Macherey-Nagel, Germany) and used as template for in-vitro transcription.
sgRNA sequences designed for LPD_N domain of vitellogenin
Table 3 Long description
Entries are listed by Sl. No, Name, Sequence, and RNA fold energy value in kilocalories per mole. Sl. No 1, Name, sgRNA_1, has Sequence, AATACGACCTGTCCACCCAC, with an RNA fold energy value of minus 0.00 kilocalories per mole. Sl. No 2, Name, sgRNA_2, has Sequence, TCTCTGAGGGCTCTAGCAGC, with an RNA fold energy value of minus 0.20 kilocalories per mole. Sl. No 3, Name, sgRNA_3, has Sequence, TGGTGGACCTGGGCGTGTTG, with an RNA fold energy value of minus 0.00 kilocalories per mole. Across the three rows, sgRNA_2 is the only entry with a nonzero negative RNA fold energy value.
PCR amplification and in-vitro transcription of sgRNA
In vitro transcription of sgRNA was carried out in a total of 50 µl reaction, containing 10 µl of 5x transcription buffer, 10 µl of 10 mM NTP mix (2 mM final concentration), 1 µg of purified template, 1.25 µL of RiboLock™ RNase Inhibitor (50 U), and 1.5 µl of T7 RNA polymerase (30 U). The mixture was incubated in a water bath at 37°C overnight. The template DNA was removed by DNase treatment using DNase I, RNase-free (Thermo Scientific, USA), as per the manufacturer’s instructions. DNase I was inactivated by phenol/chloroform extraction, and the sgRNA was precipitated by ethanol and quantified using a NanoDrop (Thermo Scientific, Waltham, MA, USA), diluted to 300 ng/μl in RNase-free water and stored at −80°C.
Validation of sgRNAs through in-vitro restriction assay
The in-vitro restriction assay was carried out to ensure the proper recognition and cleavage of the target site by the ribonucleoprotein (RNP) complex. The PCR product of the target genes containing the target sgRNA site was used as a template. All components used for in vitro restriction were obtained from NEB, USA. The cleavage reaction was performed in a total volume of 30 µL by adding the following components: 3 µL of 3.1NEBuffer, 30 nM of in vitro transcribed sgRNA cassette, and 30 nM of Cas9 enzyme (NEB), making up the reaction to 27 µL by adding ddH2O and incubating the reaction at 25°C for 15 minutes followed by 100 ng of purified PCR product added to the reaction mix, the contents were mixed thoroughly. Then, the mixtures were incubated at 37°C for 30 minutes, followed by the addition of 1 µL of Proteinase K (20 mg/mL) (NEB, USA) and incubating the reaction for 10 minutes at room temperature. The in vitro digested products were resolved by 1.5% agarose gel electrophoresis.
Three-dimensional structure prediction and visualisation of Vg and VgR
The nucleotide sequences of Maruca vitrata Vg and VgR were translated into amino acid sequences using the ExPASy Translate tool (https://web.expasy.org/translate/). The resulting protein sequences were used for homology-based three-dimensional (3D) structure prediction using the SWISS-MODEL server (https://swissmodel.expasy.org/) (Bienert et al., Reference Bienert, Waterhouse, De Beer, Tauriello, Studer, Bordoli and Schwede2017; Waterhouse et al., Reference Waterhouse, Bertoni, Bienert, Studer, Tauriello, Gumienny, Heer, De Beer, Rempfer, Bordoli and Lepore2018). Suitable template structures were automatically identified based on sequence identity, and multiple models were generated. The best-quality models were selected based on structural validation parameters including GMQE, QMEAN scores (Studer et al., Reference Studer, Rempfer, Waterhouse, Gumienny, Haas and Schwede2020), and Ramachandran plot analysis, with preference given to models showing a higher proportion of residues in favoured regions. The selected models were downloaded in Protein Data Bank (.pdb) format and visualised using UCSF Chimera (https://www.cgl.ucsf.edu/chimera/) (Pettersen et al., Reference Pettersen, Goddard, Huang, Couch, Greenblatt, Meng and Ferrin2004) to generate cartoon representations and PyMOL (https://pymol.org/) (DeLano, Reference DeLano2002) to produce surface representations of the predicted 3D structures. The quality of the predicted 3D structures of Vg and VgR was further evaluated using ERRAT and ProSA-web servers (Wiederstein and Sippl, Reference Wiederstein and Sippl2007). ERRAT (https://servicesn.mbi.ucla.edu/ERRAT/) was used to analyse the statistics of non-bonded atom–atom interactions and to assess the overall quality factor of the models. ProSA-web (https://prosa.services.came.sbg.ac.at/prosa.php) was employed to determine the overall model quality by calculating the Z-score, which reflects the degree of nativeness of the predicted structures relative to experimentally determined protein structures. The ERRAT overall quality factor and ProSA Z-scores were used as additional validation parameters to assess the reliability of the homology-modelled structures.
Molecular docking analysis of Vg and VgR
Protein–protein docking between Vg and the VgR was performed using the HDOCK server (http://hdock.phys.hust.edu.cn/) (Yan et al., Reference Yan, Wen, Wang and Huang2017) with default parameters. The homology-modelled 3D structures of Vg and the VgR were uploaded in PDB format, and docking was carried out without defining any binding site residues. The docking model exhibiting the lowest docking energy score was selected as the most favourable complex. The selected docked complex was visualised using PyMOL, and 3D surface representations were generated to examine the interaction interface between the two proteins. Identification of interacting residues was performed using LigPlot+ (https://www.ebi.ac.uk/thornton-srv/software/LigPlus/) (Laskowski and Swindells, Reference Laskowski and Swindells2011), which generated 2D interaction diagrams highlighting hydrogen bonds and hydrophobic interactions. Both 2D interaction maps and 3D surface interaction figures were generated for analysis and presentation.
Results
Cloning and sequencing
Total RNA isolated from female adults of M. vitrata displayed three distinct ribosomal RNA bands corresponding to 28S, 18S, and 5S rRNA on 2% agarose gel electrophoresis, indicating high RNA integrity and purity. Successful first-strand cDNA synthesis was validated using mtCOI-specific primers, which amplified the expected 110 bp fragment, thereby confirming the efficiency of the reverse transcription reaction. The full-length CDS of the Vg gene of M. vitrata (∼5.3 kb) was successfully amplified using gene-specific primers in overlapping fragments (Figs. S1 and S2). The amplified fragments were cloned into a pTZ57R/T vector and subjected to sequencing analysis. BLAST analysis of the assembled sequence revealed 99.04% sequence identity with the reported M. vitrata Vg CDS (GenBank accession: MG799570.1), thereby confirming the authenticity and accuracy of the cloned sequence.
Molecular phylogeny and protein domain architecture
Multiple sequence alignment followed by phylogenetic reconstruction using the ML method demonstrated that the cloned M. vitrata Vg sequence clustered within the Maruca lineage. The sequence showed close evolutionary association with Leucinodes orbonalis and Cnaphalocrocis medinalis, supported by a bootstrap value of 95% (fig. 1). The phylogenetic clustering further validated the identity and evolutionary conservation of the cloned Vg gene. In silico protein domain analysis of the predicted M. vitrata Vg protein identified three highly conserved domains characteristic of insect Vg, including the LPD_N domain (31–738 aa), DUF1943 domain (770–1049 aa), and VWD domain (1434–1614 aa) (fig. 2). Comparative exon–intron analysis revealed that the LPD_N domain was encoded within a single exon without intronic interruptions, making it a suitable target for CRISPR/Cas9-mediated genome editing. Based on this observation, three sgRNAs were designed targeting the LPD_N domain and signal peptide region of the Vg gene (fig. 3)
Molecular phylogenetic tree of M. vitrata vitellogenin with different insect vitellogenin’s based on the amino acid sequences.
Figure 1 Long description
The circular phylogenetic tree illustrates the evolutionary relationships among different insect species, categorized into six groups: Coleoptera, Hymenoptera, Hemiptera, Blattaria, Diptera and Lepidoptera. Each group is represented by a distinct color: Coleoptera in orange, Hymenoptera in blue, Hemiptera in green, Blattaria in brown, Diptera in purple and Lepidoptera in red. The tree is rooted at the center, with branches extending outward to various species. Notable species include Maruca vitrata, Cnaphalocrocis medinalis and Leucinodes orbonalis, which are closely related within the Lepidoptera group, indicated by red branches. The tree also shows a bootstrap value of 95 percent near the center, supporting the evolutionary association of certain species. Other species such as Apis mellifera and Bombus ignitus are part of the Hymenoptera group, shown in blue. The tree visually represents the genetic and evolutionary distances between these insect species, with each branch and node indicating divergence points and lineage relationships.
Domains predicted from M. vitrata vitellogenin.
The structure of single-guide RNA designed from M. vitrata vitellogenin.
Figure 3 Long description
The secondary structures of three single guide RNAs, labeled as sgRNA1, sgRNA2 and sgRNA3, are presented. Each section includes a sequence and a corresponding energy value. For sgRNA1, the sequence is AATACGACCGTGTCACCCAC, with an energy value of zero point zero zero kcal per mol. The structure is depicted with red circles forming two loops and a stem. For sgRNA2, the sequence is TCTCTGAGGGCTCTAGCAGC, with an energy value of minus zero point nine zero kcal per mol. The structure includes green and blue circles forming loops and a stem. For sgRNA3, the sequence is TGGTCGACCTGGGCTGTTG, with an energy value of zero point zero zero kcal per mol. The structure is shown with green and red circles forming loops and a stem. Each sgRNA structure is illustrated with a circular and linear representation, highlighting the nucleotide arrangement and secondary structure formation.
.
In vitro transcription and in vitro restriction assay
All three sgRNA oligonucleotides were successfully amplified through PCR (Fig. S3) and subsequently used as templates for in vitro transcription. The synthesised sgRNAs were purified and evaluated for integrity and concentration through agarose gel electrophoresis and NanoDrop quantification (Fig. S4). The results confirmed adequate yield and quality of sgRNAs for downstream CRISPR/Cas9 cleavage assays. The cleavage efficiency of the designed sgRNAs was assessed through an in vitro Cas9-mediated cleavage assay using PCR-amplified Vg fragments as target substrates. Among the three sgRNAs evaluated, sgRNA1 exhibited efficient site-specific cleavage, generating the expected fragments of 1035 and 1485 bp. Similarly, sgRNA2, designed against the signal peptide region, also demonstrated effective cleavage activity. In contrast, sgRNA3 failed to induce cleavage of the target DNA sequence. Secondary structure prediction analysis indicated that sgRNA3 lacked an accessible unpaired guide RNA region, likely impairing the formation of a stable Cas9–sgRNA RNP complex and subsequent target recognition (Fig. S5). Collectively, these findings identified sgRNA1 and sgRNA2 as efficient candidates for subsequent CRISPR/Cas9-mediated genome editing studies in M. vitrata.
Protein structure of vitellogenin and vitellogenin receptor
The 3D structure of Vg from Maruca vitrata was predicted using homology-based modelling implemented in the SWISS-MODEL server. Vg from honey bee haemolymph (PDB ID: 9ENR, chain A) was identified as the most appropriate structural template based on sequence homology and model quality indices. The generated Vg model exhibited a GMQE score of 0.58 and a QMEANDisCo global score of 0.60 ± 0.05, indicating a structurally reliable and well-supported model. The predicted structure displayed the conserved folding architecture typical of insect Vgs and showed strong structural similarity with previously characterised Vg proteins. The finalised structure was visualised using UCSF Chimera and PyMOL to generate cartoon and surface representations for detailed structural interpretation (fig. 4). The 3D structure of VgR from Maruca vitrata was predicted through homology-based modelling using the SWISS-MODEL platform. The low-density lipoprotein receptor-related protein 2 (PDB ID: 8EM7) was selected as the best-fit template based on sequence identity and structural quality metrics. The predicted VgR model showed a GMQE score of 0.51 and a QMEANDisCo global score of 0.56 ± 0.05, indicating moderate to good confidence in the generated structure. Structurally, the modelled receptor exhibited conserved features characteristic of the low-density lipoprotein receptor superfamily, consistent with the expected architecture of insect Vg receptors. The finalised model was visualised using UCSF Chimera and PyMOL to obtain high-resolution cartoon and surface representations for structural examination (fig. 5).
Three-dimensional structure of vitellogenin from M. vitrata. The three-dimensional structure of vitellogenin was predicted by homology modelling using the SWISS-MODEL server, with honey bee vitellogenin (PDB ID: 9ENR, chain A) as the template. (A) Cartoon representation of the predicted vitellogenin structure showing the overall folding and domain organisation. (B) Surface representation of the vitellogenin protein generated using PyMOL, illustrating the surface topology and potential interaction regions.
Three-dimensional structure of vitellogenin receptor from M. vitrata. The three-dimensional structure of the vitellogenin receptor was predicted using homology-based modelling in the SWISS-MODEL server, employing low-density lipoprotein receptor-related protein 2 (PDB ID: 8EM7) as the template. (A) Cartoon representation of the predicted vitellogenin receptor structure highlighting the overall receptor architecture. (B) Surface representation of the vitellogenin receptor visualised in PyMOL, showing the molecular surface features relevant for ligand interaction.
Validation of protein models
Structural validation of the predicted Vg model was performed using Ramachandran plot analysis and MolProbity assessment. The Ramachandran plot revealed that 89.38% of amino acid residues were positioned within the most favoured regions, whereas only 2.78% of residues were detected as outliers (fig. 6A; table 4). In addition, the model achieved a MolProbity score of 1.47, reflecting favourable stereochemical parameters and acceptable backbone conformations. These results collectively demonstrate the high conformational quality and stereochemical reliability of the modelled Vg structure. The stereochemical quality of the predicted VgR model was evaluated using Ramachandran plot and MolProbity analyses. Ramachandran plot assessment indicated that 86.48% of the residues were distributed within the favoured regions, supporting the conformational validity of the modelled receptor structure (fig. 6B; table 4). Furthermore, the model yielded a MolProbity score of 2.20, indicating acceptable stereochemical quality for a large multidomain receptor protein. These observations confirmed the overall structural stability and conformational appropriateness of the VgR model.
Structural validation parameters of homology-modelled vitellogenin and vitellogenin receptor from Maruca vitrata
Table 4 Long description
Structural validation parameters are listed for two proteins, vitellogenin and vitellogenin receptor, with columns for Template, PDB ID, GMQE score, QMEANDisCo Global, MolProbability score, and Ramachandran favoured. Row 1 reports vitellogenin with template PDB ID 9ENR, GMQE score 0.58, QMEANDisCo Global 0.60 plus or minus 0.05, MolProbability score 1.47, and Ramachandran favoured 89.38 percent. Row 2 reports vitellogenin receptor with template PDB ID 8EM7, GMQE score 0.51, QMEANDisCo Global 0.56 plus or minus 0.05, MolProbability score 2.20, and Ramachandran favoured 86.48 percent. Across the two rows, vitellogenin has higher GMQE and QMEANDisCo Global values, while the vitellogenin receptor has the higher MolProbability score.
Ramachandran plot validation of vitellogenin and vitellogenin receptor structures from Maruca vitrata. The Ramachandran plots illustrate the φ–ψ dihedral angle distribution of amino acid residues in the homology-modelled proteins. (A) Vitellogenin and (B) vitellogenin receptor models display the majority of residues within favoured regions, with a minimal proportion of outliers, confirming acceptable stereochemical quality of the predicted structures.
Figure 6 Long description
The scatter plot labeled (A) shows the phi versus psi dihedral angle distribution for Vitellogenin. The horizontal axis represents phi, ranging from negative 180 degrees to 180 degrees and the vertical axis represents psi, also ranging from negative 180 degrees to 180 degrees. The plot displays dense clusters of points in the negative phi and positive psi regions, indicating favored conformations. Several contour-like regions surround these clusters, suggesting areas of higher density. Outliers are scattered outside these regions. The scatter plot labeled (B) shows the phi versus psi dihedral angle distribution for the Vitellogenin receptor. The axes are the same as in plot (A), with phi on the horizontal and psi on the vertical. Similar dense clusters are observed in the negative phi and positive psi regions, with contour-like regions indicating density. Outliers are present outside these clusters. Both plots use color coding to represent different categories or density levels, though specific meanings are not labeled. The plots illustrate the stereochemical quality and favored regions of the modeled structures, with minimal outliers indicating acceptable conformational validity.
Accuracy of protein models
The quality of the predicted Vg structure was further evaluated using the ERRAT server to examine non-bonded atomic interactions. ERRAT analysis produced an overall quality factor of 86.003%, indicating a high level of structural consistency and acceptable atomic interaction patterns throughout the model (fig. 7A). The obtained score further supports the reliability of the homology-modelled Vg structure. ERRAT analysis of the VgR model generated an overall quality factor of 82.231%, demonstrating satisfactory non-bonded atomic interactions and acceptable structural reliability of the predicted receptor model (fig. 7B). The obtained quality score supports the robustness of the homology-derived VgR structure. Reliability was tested using the ProSA-Web server generated a Z-score of −10.92 for the predicted Vg model (fig. 8A). The obtained Z-score falls within the range typically observed for experimentally resolved protein structures of comparable size, confirming the overall structural plausibility and model accuracy. Collectively, the Ramachandran plot, MolProbity, ERRAT, and ProSA-Web analyses substantiated the structural integrity and reliability of the predicted Vg model. The predicted VgR structure was further validated using the ProSA-Web server, which generated a Z-score of −12.85 (fig. 8B). The obtained score was within the range commonly observed for experimentally determined protein structures with similar molecular size and folding characteristics, indicating good overall model quality. Together, the Ramachandran plot statistics, MolProbity score, ERRAT quality factor, and ProSA-Web validation confirmed the stereochemical accuracy and structural reliability of the predicted VgR model.
ERRAT analysis of homology-modelled vitellogenin and vitellogenin receptor from Maruca vitrata. ERRAT was used to assess the overall quality of the predicted three-dimensional protein structures based on non-bonded atomic interactions. (A) ERRAT plot of vitellogenin, showing a high overall quality factor, indicating acceptable structural reliability of the homology-modelled protein. (B) ERRAT plot of vitellogenin receptor, demonstrating an overall quality factor within the acceptable range for large receptor proteins, supporting the validity of the predicted model.
Figure 7 Long description
Text, Program: ERRAT2. File: vitellogenin.pdb. Chain: A. Overall quality factor asterisk: 86.003. A bar graph with the vertical axis label, Error value asterisk, with tick labels 95 percent and 99 percent. The horizontal axis label is, Residue number window center, with tick labels 40, 60, 80, 100, 120, 140, 160, 180, 200, 220, 240, 260, 280, 300, 320. Two horizontal reference lines align with 95 percent and 99 percent. Bars vary in height across the residue range, with multiple tall spikes reaching near the 99 percent line and above it. The tallest clusters of spikes appear around residue numbers near 40 to 60, near 80 to 100, near 160 to 180, near 200 to 240 and near 320. Text, Program: ERRAT2. File: vitellogenin receptor.pdb. Chain: A. Overall quality factor asterisk: 82.231. A bar graph with the vertical axis label, Error value asterisk, with tick labels 95 percent and 99 percent. The horizontal axis label is, Residue number window center, with tick labels 40, 60, 80, 100, 120, 140, 160, 180, 200, 220, 240, 260, 280, 300, 320. Two horizontal reference lines align with 95 percent and 99 percent. Bars vary in height across the residue range, with several tall clusters reaching near the 99 percent line and above it. The tallest clusters of spikes appear around residue numbers near 55 to 70, near 130 to 150, near 210 to 250 and near 290 to 320.
ProSA-web analysis of homology-modelled vitellogenin and vitellogenin receptor from Maruca vitrata. ProSA-web was used to assess the overall quality of the predicted three-dimensional protein structures by calculating the Z-score relative to experimentally determined structures. (A) ProSA-web Z-score plot of vitellogenin, showing a Z-score within the range characteristic of native protein structures of comparable size. (B) ProSA-web Z-score plot of vitellogenin receptor, indicating a Z-score consistent with reliable and structurally sound protein models.
Protein–protein docking results of vitellogenin and vitellogenin receptor complex
Protein–protein docking between Vg and VgR was carried out using the HDOCK server to evaluate the molecular interaction and binding stability of the complex. Among the generated docking conformations, the top-ranked model with the lowest docking score (−295.66) and a high confidence score (0.9485) was selected as the most reliable Vg-VgR interaction model (Table 5). 3D visualisation of the docked complex using PyMOL demonstrated a well-organised interaction interface between Vg and VgR. Surface representation analysis revealed substantial surface complementarity between the two proteins, suggesting efficient molecular recognition and stable complex formation (fig. 9). The interaction interface indicated close spatial association between the ligand and receptor molecules, supporting the biological relevance of the predicted docking model. Further characterisation of intermolecular interactions using LigPlot + identified several amino acid residues contributing to the stabilisation of the Vg-VgR complex through hydrogen bonding and hydrophobic interactions (fig. 10). Key residues from the VgR (chain B), including Arg1635, Thr925, Gln724, His975, Arg1201, Arg1164, and Asp1146, formed hydrogen bonds with corresponding Vg residues Asn1371, Asn723, Ala757, Ser1388, Glu785, Pro90, and Arg1070 of chain A, respectively (table 6). In addition to hydrogen bonding, several neighbouring residues from both proteins were involved in hydrophobic contacts that further strengthened the docking interface and contributed to overall complex stability. Collectively, the docking analysis demonstrated a strong and specific interaction between Vg and its receptor, supported by favourable docking parameters, high surface complementarity, and multiple stabilising intermolecular interactions. These findings provide structural insights into the molecular basis of Vg-VgR recognition and suggest the functional significance of this interaction in Vg transport and reproductive physiology in M. vitrata.
Docking score parameters of the vitellogenin and vitellogenin receptor complex predicted using HDOCK
Table 5 Long description
Parameters reported are docking score and confidence score for a predicted docking model. The columns are Parameter, Value, and Description. Docking score has a value of minus 295.66, described as an HDOCK energy based scoring function where lower values indicate stronger predicted binding. Confidence score has a value of 0.9485, described as the confidence level of the predicted docking model with a range from 0 to 1.
Key intermolecular interactions between vitellogenin and vitellogenin receptor identified by LigPlot+
Table 6 Long description
Columns list Sl. No, VgR residue, Vg residue, and Interaction type for intermolecular contacts. All seven entries have Interaction type, Hydrogen bond. Sl. No 1 pairs VgR residue, Arg1635 with Vg residue, Asn1371. Sl. No 2 pairs Thr925 with Asn723, and Sl. No 3 pairs Gln724 with Ala757. Sl. No 4 pairs His975 with Ser1388, Sl. No 5 pairs Arg1201 with Glu785, Sl. No 6 pairs Arg1164 with Pro90, and Sl. No 7 pairs Asp1146 with Arg1070.
Surface representation of the docked complex between (A) vitellogenin and (B) vitellogenin receptor from M. vitrata. Protein–protein docking between vitellogenin and the vitellogenin receptor was performed using the HDOCK server. The top-ranked docking model with the lowest docking score (−295.66) and high confidence score (0.9485) was selected. The docked complex was visualised using PyMOL in surface representation, highlighting the complementary interaction interface between vitellogenin and the vitellogenin receptor, indicative of stable molecular recognition.
Two-dimensional interaction map of (A) vitellogenin and (B) vitellogenin receptor generated using LigPlot+. LigPlot+ analysis of the selected docked vitellogenin and vitellogenin receptor complex showing key intermolecular interactions at the binding interface. Hydrogen bonds between interacting residues are represented by dashed lines, while hydrophobic interactions are depicted as spoked arcs. Several residues from both vitellogenin and the vitellogenin receptor contribute to stabilising the complex through hydrogen bonding and hydrophobic contacts.
Figure 10 Long description
”The image showing a two-dimensional residue interaction map of Vitellogenin and Vitellogenin receptor. Residues from both proteins are shown as labeled nodes. Hydrogen bonds between interacting residue pairs are represented by green dashed lines. Hydrophobic interactions are shown as spoked arcs around residue nodes. In this panel, residues from chain B of the receptor, including arginine 1635, threonine 925 and glutamine 724, are shown forming hydrogen bonds with chain A residues asparagine 1371, asparagine 723 and alanine 757, respectively. Additional residue nodes with spoked arcs indicate hydrophobic contacts involving neighboring residues such as tyrosine 900, methionine 908, asparagine 1101, asparagine 368, asparagine 369, alanine 176, tyrosine 044, threonine 329, glutamine 049, glutamine 097, tryptophan 900, alanine 751, tryptophan 56 and asparagine 1136. These nodes appear along the upper and lower horizontal dashed boundary lines separating the two protein.
Discussion
Vg is recognised as the major yolk protein precursor in insects and plays an indispensable role in vitellogenesis, reproduction, and embryonic development (Guo et al., Reference Guo, Zu, Liu, Yan, Yang and Xu2025). In the present investigation, the complete CDS of the Vg gene from M. vitrata was successfully cloned and molecularly characterised. The authenticity of the obtained sequence was confirmed through sequence homology analysis and phylogenetic reconstruction. The cloned sequence exhibited a high degree of similarity with the predicted Vg sequence of M. vitrata, thereby validating the accuracy of the amplification and annotation. Furthermore, phylogenetic analysis demonstrated that the M. vitrata Vg clustered closely with homologous sequences from other Lepidopteran insects, indicating a strong evolutionary relationship among members of this order. Such phylogenetic conservation of Vg genes has previously been reported in several lepidopteran species, highlighting the highly conserved nature and evolutionary importance of Vgs across insect taxa (Ashok et al., Reference Ashok, Bhargava, Venkatesh, Balasubramani, Murugan, Geethalakshmi, Manamohan, Jha and Asokan2025). The conserved architecture and sequence identity observed in the present study suggest that the Vg gene performs essential biological functions that have been maintained during insect evolution.
Detailed protein domain analysis of the predicted M. vitrata Vg protein identified the presence of three highly conserved domains, namely the lipoprotein N-terminal domain (LPD_N), domain of unknown function 1943 (DUF1943), and von Willebrand factor type D domain (VWD), which are characteristic features of Vg proteins in Lepidopteran insects. These domains are known to play critical roles in lipid transport, protein stability, and yolk formation during oocyte maturation. Among these, the LPD_N domain is considered particularly important because of its involvement in receptor-mediated uptake and transport of yolk proteins. Interestingly, the LPD_N region in the present study was encoded by a single exon, making it an attractive and accessible target for CRISPR/Cas9-mediated genome editing (Zhang et al., Reference Zhang, Wu, Li, Zhan, Wang, Song, Zhang, He, He, Xu and Deng2025). Previous studies have demonstrated that disruption of Vg genes or VgR genes severely affects oogenesis, fecundity, and embryonic development in several insect species. Functional knockout studies using CRISPR/Cas9 have shown that deletion of Vg in Leucinodes orbonalis, as well as disruption of VgR genes in Plutella xylostella and Bactrocera dorsalis, resulted in substantial reductions in egg hatchability and reproductive fitness (Peng et al., Reference Peng, Wang, Zou, Qin, Vasseur, Chu, Zhai, Dong, Liu, He and Yang2020; Ashok et al., Reference Ashok, Bhargava, Venkatesh, Balasubramani, Murugan, Geethalakshmi, Manamohan, Jha and Asokan2025; Xie et al., Reference Xie, Wang, Dou, Smagghe, Zhang and Wang2025). These findings strongly support the potential utility of targeting the Vg gene in M. vitrata as a viable molecular strategy for population suppression and pest management.
In addition to sequence characterisation, structural analysis of the Vg and VgR proteins provided further support for their predicted biological functions. The homology models exhibited satisfactory structural quality, as indicated by the GMQE values, Ramachandran plot statistics, ERRAT scores, and ProSA Z-scores, suggesting that the predicted models were suitable for downstream interaction analysis. Protein–protein docking further demonstrated a favourable interaction between Vg and VgR, consistent with the established role of VgR in mediating receptor-dependent uptake of Vg during oocyte development. Similar structural validation parameters and docking-based approaches have been successfully employed to investigate different proteins in insects, supporting the reliability of computational structural analyses for functional interpretation (Pradeep et al., Reference Pradeep, Ashok, VinayKumar, Babu, Subhaharan, Nagesha and Asokan2025).
The successful application of CRISPR/Cas9 technology largely depends on the efficient design and validation of sgRNAs, which direct the Cas9 nuclease to specific genomic loci. Although several computational and web-based platforms are available for sgRNA prediction, the actual cleavage efficiency of sgRNAs often differs considerably between in silico predictions and experimental validation under in vitro or in vivo conditions (Jasieniecka and Domingues, Reference Jasieniecka and Domingues2025). In the present study, three sgRNAs targeting the Vg gene were designed and evaluated through in vitro cleavage assays. Among these, sgRNA1 and sgRNA2 exhibited efficient cleavage activity, whereas sgRNA3 failed to produce detectable cleavage. Similar inconsistencies between computational prediction scores and experimental performance have also been reported in earlier studies (Ruitón et al., Reference Ruitón, Effio, Campos, Cornejo and Farro2025), emphasising the necessity of empirical validation prior to genome editing experiments.
The lack of activity observed in sgRNA3 may be attributed to unfavourable secondary structural conformations that interfered with target recognition and Cas9 complex formation. Efficient sgRNAs generally possess a structurally stable scaffold containing three stem-loop structures along with an accessible and unpaired guide RNA region that facilitates precise hybridisation with the target DNA sequence. Structural analysis conducted in the present study revealed that the guide region of sgRNA3 was not freely accessible because of unintended hybridisation with portions of the scaffold RNA. Such intramolecular interactions likely hindered the formation of a functional Cas9-sgRNA RNP complex and prevented efficient binding to the target site. In contrast, sgRNA1 and sgRNA2 retained the necessary structural features required for effective target recognition and cleavage, thereby demonstrating robust in vitro activity. The successful performance of these sgRNAs highlights their suitability for future functional genomics applications, particularly RNP-mediated embryo microinjection experiments in M. vitrata.
The utilisation of Cas9-sgRNA RNP complexes represents a promising non-transgenic genome editing approach that minimises concerns associated with stable transgene integration, regulatory constraints, and biosafety risks (Singh et al., Reference Singh, Rahangdale, Pandita, Saxena, Upadhyay, Mishra and Verma2022). Compared with DNA-based transformation systems, RNP-mediated editing provides transient nuclease activity, reduces off-target effects, and avoids permanent incorporation of foreign genetic material into the host genome. Such strategies form the conceptual basis of emerging technologies such as pgSIT, which employs CRISPR-induced heritable mutations in essential reproductive genes to suppress pest populations in an environmentally sustainable manner (Kandul et al., Reference Kandul, Liu, Sanchez, Wu, Marshall and Akbari2019). Therefore, the identification and validation of highly efficient sgRNAs targeting the Vg gene in M. vitrata provide a valuable foundation for the development of advanced CRISPR-based genetic control strategies aimed at sustainable management of this economically important pest species.
Despite the promising findings of the present study, several limitations should be considered before the practical application of CRISPR/Cas9-mediated genome editing in M. vitrata. Although sgRNAs were designed using CRISPOR to maximise target specificity and minimise predicted off-target effects, comprehensive genome-wide off-target assessment was beyond the scope of the present study. Future studies employing a high-quality reference genome and whole-genome sequencing will be necessary to validate off-target mutations following in vivo genome editing. Furthermore, although qualitative in vitro cleavage assays confirmed the functionality of sgRNA1 and sgRNA2, quantitative estimation of cleavage efficiency using densitometric analysis was beyond the scope of the present study. Future studies will incorporate quantitative analyses to further evaluate and compare sgRNA performance. In addition, efficient delivery of CRISPR/Cas9 components into embryos and optimisation of embryo microinjection protocols remain major technical challenges that can influence editing efficiency and embryo survival. Furthermore, the deployment of gene-edited insects for pest management requires comprehensive biosafety evaluation and compliance with national and international regulatory frameworks. Addressing these challenges will be essential for translating CRISPR/Cas9-based strategies into practical and sustainable pest management applications.
Conclusion
This study elucidates the molecular characterisation of Vg CDS in M. vitrata and confirms the efficacy of functional sgRNAs for CRISPR/Cas9 targeting. This research identifies sgRNA1 and sgRNA2 as effective candidates, laying the groundwork for subsequent functional genomics and biocontrol strategies targeting this significant legume pest. Additional research utilising embryo microinjection and heritable genome editing is necessary to validate the reproductive phenotypes linked to Vg disruption and to assess the practicality of employing genome editing for comprehensive management of M. vitrata.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0007485326101308.
Acknowledgements
The authors sincerely thank the Director of ICAR-IIHR, ICAR-NBAIR Bengaluru, for providing the necessary infrastructure and support. They also gratefully acknowledge the ICAR Network Project on Agricultural Bioinformatics and Computational Biology for funding this study.
Author contributions
C.P.: Investigation, data curation, formal analysis, validation, visualisation, writing – original draft. P.M.S.: Data analysis, validation. S.N.N.: Bioinformatics analysis, data interpretation. R.A.: Supervision, conceptualisation, resources, funding acquisition, project administration, writing – review and editing. A.K.: Conceptualisation, CRISPR/sgRNA experimental design, supervision, validation, data interpretation, visualisation, writing – review and editing.
