Overview
APHiGO aims to combat aphid damage to citrus crops through safe and eco-friendly treatments, addressing the shortcomings of existing chemical methods that result in environmental pollution. As described on the project webpage, the application of chemical pesticides often leads to unavoidable chemical residues, jeopardizing the health and balance of organisms in the treated areas and even triggering the frightening consequences of chemical resistance. To demonstrate that APHiGO is a viable new approach for managing citrus aphid infestations, we have outlined our validation logic in detail on the Experiment website. In this section, we will comprehensively present the key results from each of the validation modules.
Detection Module
To monitor changes in aphid population density and thereby predict the optimal pesticide application time, we constructed an engineered Bacillus subtilis strain capable of responding to sucrose—one of the main components of honeydew (aphid excrement). We designed a recombinant plasmid pgrac100-MS that can convert sucrose into methyl salicylate (MeSA) and successfully transformed it into Bacillus subtilis 168Δ4. After induction with sucrose, quantitative analysis via liquid chromatography-mass spectrometry (LC-MS) showed that the MeSA yield reached 1.595 ng·mL⁻¹/OD under the induction concentration of 8%. This engineered bacterium lays the foundation for both the strain and quantitative method required for the early warning of citrus aphids.
View Details →RNA interference therapy
To screen for the RNA form with the highest insecticidal efficiency and its corresponding production method, we carried out the optimization of RNA production conditions and analyzed the silencing and lethal effects. By adopting the bidirectional promoter and multi-terminator production strategy described by Juliana da Rosa、SJ Ross et al., we successfully achieved the efficient preparation of RNA molecules. At the same time, the fourth-generation RNA molecule we constructed using two natural miRNA scaffolds from aphids, api-mir-71 and api-mir-3017a, exhibited a high aphid mortality rate of up to 60% after being applied at a concentration of 800 ng/μl for 120 hours.
View Details →Delivery Module
The susceptibility of RNA molecules to degradation during delivery is a major bottleneck restricting their clinical application. To improve RNA stability, we developed a virus-like particle (VLP)-based delivery system, specifically engineering MS2 VLPs capable of specifically packaging amiRNA via genetic engineering approaches. MS2 VLPs are non-replicating nanoparticles based on the capsid protein of bacteriophage MS2, exhibiting icosahedral symmetry. Herein, we successfully validated the integrity of the capsid structure of this delivery system and its effective resistance to nucleases, and further explored the feasibility of delivering it to the shoots of trees through trunk injection.
View Details →Fungal Contact-Killing Module
To enhance the virulence of Metarhizium, we modified Metarhizium via an Agrobacterium-mediated genetic transformation method to express target RNA molecules. This method is based on the ability of Agrobacterium to transfer the T-DNA region on its Ti plasmid into host cells, thereby achieving the integration of the target gene. During the transformation process, Agrobacterium delivers T-DNA into host cells through its secretion system and integrates the target gene into the Metarhizium genome, ensuring the stable expression of the target gene in the fungus. Here, we successfully verified the ability of the transformed Metarhizium to express the target RNA, and further analyzed the virulence of Metarhizium at the phenotypic level.
View Details →
Detection module
Background Information
In March 2025, we conducted a research survey in Shanwei, Guangdong Province—a region with a high incidence
of citrus aphid infestations. It was found that citrus aphids are tiny in size and often hide on the
undersides of leaves, making them difficult to detect. This frequently causes farmers to miss
the optimal pesticide application time. Based on this observation, the team aimed to develop a biosensor
component capable of responding to the population density of citrus aphids.
Aphids excrete honeydew immediately after feeding; the amount of honeydew is positively correlated with
their population density, and one of its main components is sucrose Using this
characteristic, we designed and constructed a biosensor that can respond to sucrose signals. Taking Bacillus
subtilis—a biocontrol bacterium approved as a biopesticide—as the chassis, we implanted the
following genetic components: ① a sucrose-sensing element; ② the pchBA gene cluster (which can convert
chorismate, a metabolic product of sucrose, into salicylic acid (SA));③BSMT methyltransferase (which can
convert salicylic acid into volatile MeSA) Furthermore, the methyl salicylate
produced by this engineered bacterium can be recognized by hardware devices, enabling
real-time monitoring of pest density. This fundamentally solves the problem of aphids being
"difficult to detect" and advances the vision of precision agriculture and ecological sustainability.
Implementation Objectives
1. Construct an engineered Bacillus subtilis strain capable of expressing MeSA under sucrose induction.
2.Conduct a preliminary evaluation to determine whether the engineered strain can respond to changes in sucrose concentration, thereby expressing methyl salicylate at different concentrations.
Verification process
Plasmid construction
pgrac100 is a commonly used E. coli-Bacillus subtilis shuttle vector, containing two resistance genes (for ampicillin and chloramphenicol). When conducting protein expression in Bacillus subtilis, chloramphenicol can be used for screening.
We inserted the following genetic elements into the Multiple Cloning Site (MCS): ① sacB promoter, a sucrose-inducible promoter; ② Ribosome Binding Site (RBS); ③ pchA expression cassette: expresses isochorismate synthase, which catalyzes the conversion of chorismate to isochorismate; ④ pchB expression cassette: expresses isochorismate pyruvate lyase, which converts isochorismate into pyruvate and salicylate; ⑤ BSMT1 expression cassette: methylates salicylate to produce methyl salicylate (a volatile ester); ⑥ terminator.
Fig 2-1. Schematic diagram of the pgrac100-MS recombinant plasmid.
Plasmid transformation
After constructing and amplifying the plasmid, we prepared competent cells of Bacillus subtilis 168Δ4 using the medium-alternating culture method and transformed both the empty vector and the recombinant plasmid into the cells. Single colonies clones were randomly picked from resistant plates, and after amplification, the plasmids were extracted for PCR and double digestion.Plasmid PCR showed that the empty vector control yielded a band at 1537 bp, while the recombinant plasmid showed the expected band at 1777 bp. The double digestion results each showed two bands corresponding to the plasmid vector and the digested fragment. The empty vector digestion product yielded a band at 776 bp, while the recombinant plasmid digestion product showed the expected band at 2968 bp. These results confirmed the successful construction and transformation of the recombinant plasmid.
Fig 2-2.
Results of pgrac100-MS recombinant plasmid transformation
(a) Growth of colonies on chloramphenicol-resistant plates after plasmid transformation.
(b) Agarose gel electrophoresis results of plasmid PCR and restriction digestion. M: DL5000 Marker; 1: PCR
result of the empty vector, with the target band at 776 bp; 2: PCR result of the recombinant plasmid, with the
target band at 2968 bp; 3: digestion result of the empty vector, with the target band at 1537 bp; 4: digestion
result of the recombinant plasmid, with the target band at 1777 bp. The gel was prepared with 2% agarose and
run at 120 V for 45 minutes.
Quantification of Methyl Salicylate by Liquid Chromatography-Mass Spectrometry
We induced the engineered Bacillus subtilis with sucrose at concentrations of 0%, 4%, and 8% (w/v) for 18 hours, and quantified methyl salicylate (MeSA) using liquid chromatography-mass spectrometry (LC-MS). Based on the standard curve, we confirmed after diluting 5 times the production of MeSA and determined the yields to be 1 ng·mL⁻¹/OD, 1.379 ng·mL⁻¹/OD, and 1.595 ng·mL⁻¹/OD, respectively.
Fig 2-3.
Expression levels of methyl salicylate
(a) Standard curve of methyl salicylate.(b) Methyl salicylate concentrations induced by different sucrose
concentrations (dilute 5 times and quantify). Since sucrose induction inhibits the growth of Bacillus subtilis
to varying degrees, the expression levels were normalized using OD values to exclude the influence of cell
number. The OD₆₀₀ values for induction at 0%, 4%, and 8% were 4.102, 3.263, and 2.194, respectively.
The PDF of the original file is provided below and can be viewed by clicking the link. The sample values represent the content after a 5-fold dilution.
Discussion
In this part of the experiment, we successfully constructed and transformed the pgrac100-MS recombinant plasmid, obtaining a genetically stable engineered Bacillus subtilis 168Δ4 strain. Through liquid chromatography-mass spectrometry (LC-MS), we initially verified that the constructed engineered strain can express MeSA.
However, in this experiment, the engineered Bacillus subtilis strain under 0% sucrose induction concentration was detected to produce 1 ng·mL⁻¹/OD of methyl salicylate, indicating that there is basal leakage in the system. Therefore, in subsequent experiments, we will test the methyl salicylate expression level of the empty vector and use it as a control. To further optimize this strain, we will attempt to introduce a more stringent sucrose operon, such as the suc dual regulation system, to reduce leaky expression.
In addition, on the basis of the qualitative verification in this module, we also initially confirmed that the engineered strain can respond to changes in sucrose concentration, thereby expressing methyl salicylate at different concentrations. In the future, we will establish a strict sucrose response gradient, and combine it with the aphid population density and the detection limit of the detection hardware to achieve a closed loop from qualitative response to quantitative early warning.
RNA interference therapy
Background Information
RNAi is an eukaryotic gene silencing mechanism that cuts RNA to form siRNA, assembles the RISC complex, and degrades target mRNA, leading to organism death. Due to its high specificity and environmental friendliness, RNAi has become a core technology for precise pest control in agriculture. It addresses key issues in citrus aphid management, such as pesticide resistance, environmental pollution, and poor stability of biological control methods. This module focuses on the entire RNAi process: screening essential aphid genes to ensure lethality; optimizing RNA design (upgrading from single-target dsRNA to dual/tri-fusion dsRNA, tri-shRNA, bi-amiRNA) to enhance efficiency and stability; utilizing bidirectional promoters and multiple terminators to achieve high-level dsRNA production and address transcription "read-through" issues; quantifying silencing efficiency via RT-qPCR and validating the effect of fourth-generation RNA through lethality tests. The goal is to optimize parameters and develop a practical control solution to support the green transformation of the citrus industry.
Implementation Objectives
1. Verify the quality of plasmid linearization products by electrophoresis, confirm that multiple T7 terminators are more conducive to efficient dsRNA transcription than a single terminator, and provide a reference for high-efficiency RNA production.
2. Validate the effectiveness of procedures such as plasmid transformation, double-enzyme digestion linearization, and in vitro transcription, and optimize the transcription system to ensure the stable synthesis of target dsRNA.
3. Clarify the insecticidal activity of the four generations of RNA against aphids. Focus on verifying the synergistic insecticidal effect of fused dsRNA, and evaluate the advantages of tri-shRNA and bi-amiRNA to support product iteration.
Verification process
Optimization of a high-efficiency in vitro dsRNA production system
▼The results showed that a transcription duration of 6 hours enabled relatively high yield in a short time. By optimizing the template with 3-tandem terminators and reverse-designed 3-tandem terminators, the phenomenon of transcription read-through was reduced. Compared with the template without terminators, the yield of dsRNA synthesized via in vitro transcription using the 3-tandem terminator template was increased.
Fig 3-3. (a) Significance analysis of transcription product concentrations for NT, 1T, 2T, and 3T templates at 1-fold dilution; (b) significance analysis of transcription product concentration for NT, 1T, 2T, and 3T templates at 5-fold dilution; (c) significance analysis of transcription product concentration for NT, 1T, 2T, and 3T templates at 10-fold dilution; (d) linear regression curves showing the relationship between dilution ratio and concentration.
If you are interested, please click here to learn more about our detailed experimental procedures.
To investigate the impact of different terminators on RNA production efficiency, we constructed dsRNA
expression vectors containing a single T7 terminator and multiple T7 terminators for
in vitro
transcription. Moreover, we set restriction enzyme sites between each terminator to enable the
conversion between templates with different terminators. We observed the band patterns of
the
products via gel electrophoresis, analyzed the yield and purity of the products, and determined the
effectiveness of the terminators.
We first prepared linear templates of NT, 1T, 2T, and 3T (xT
represents
x terminators, and NT represents no terminator). Through agarose gel verification, we successfully
obtained the linear templates in the target band region.
Fig 3-1. (a)MW:marker1,2:TAKARA DL1000 DNA marker; marker3: BBI DNA Marker A (25~500 bp);1:CHS-1T(456bp);2:CHS-2T(594bp).(b)MW:marker(left):TAKARA DL1000 DNA marker; marker(right): BBI DNA Marker A (25~500 bp) ;1:CHS-NT(346bp);2:CHS-3T(716bp);xT stands for x terminators, and NT stands for no terminator.
We performed in vitro transcription on different templates separately. In Fig 3-2. (a) ,the transcription product of the 3T template showed relatively higher purity compared with the others, but the difference in yield among them was not obvious in Fig 3-2. (b).
Fig 3-2. (a) is a schematic diagram of agarose gel electrophoresis analysis for purity comparison; (b) is a schematic diagram of agarose gel electrophoresis analysis for yield comparison. MW: Marker; NT-CHS: CHS double-stranded RNA transcribed from the terminator-free template; 1T-CHS: CHS double-stranded RNA transcribed from the template with the T7UUCG terminator; 2T-CHS: CHS double-stranded RNA transcribed from the template with two tandem terminators (T7UUCG terminator and rrnBT1 terminator); 3T-CHS: CHS double-stranded RNA transcribed from the template with three tandem terminators (T7UUCG terminator, rrnBT1 terminator, and T7 terminator).
To more accurately determine the transcription product yield of templates with different terminators, we diluted the RNA samples 1-fold, 5-fold, and 10-fold with nuclease-free water respectively, then measured their concentrations using a Nanodrop. The test was repeated 3 times, with nuclease-free water serving as the blank group. We calculated the average value of the obtained concentration data, established linear regression equations between the dilution factor and the diluted concentration, and then calculated the original concentration and conducted significance analysis. The linear regression equations for NT, 1T, 2T, and 3T are as follows:YNT=6317*X-29.37,Y1T=9538*X+164.8, Y2T=9604*X-100.3,Y3T=11006*X+240.3.
where X represents the dilution ratio (C now/C original) and is dimensionless, and Y represents the concentration corresponding to the dilution ratio, with the unit of ng/μL. Fig 3-3, we found that the concentration of dsCHS transcribed using the template with tandem triple terminators (3T) was significantly higher than that of templates with other terminators.
Fig 3-3. (a) Significance analysis of transcription product concentrations for NT, 1T, 2T, and 3T templates at 1-fold dilution; (b) significance analysis of transcription product concentration for NT, 1T, 2T, and 3T templates at 5-fold dilution; (c) significance analysis of transcription product concentration for NT, 1T, 2T, and 3T templates at 10-fold dilution; (d) linear regression curves showing the relationship between dilution ratio and concentration. NT-CHS：YNT=6317*X-29.37,R2=0.9967；1T-CHS：Y1T=9538*X+164.8,R2=0.9997；2T-CHS： Y2T=9604*X-100.3,R2=0.9969；3T-CHS：Y3T=11006*X+240.3,R2=0.9971。ns：p>0.05,*：0.05,**：p <0.01,***：p<0.001,****：p<0.0001< /p>
We also investigated the effect of transcription duration on transcription efficiency. We transcribed the same batch of plasmid templates for 4 h, 6 h, and 12 h respectively, measured the concentration of the products after different dilutions, and performed significance analysis on the results (as shown in Fig 3-4).It was found that the concentration of the transcription products after 6 h and 12 h was significantly higher than that after 4 h, but there was no significant difference between 6 h and 12 h. Therefore, after comprehensive consideration, 6 h was determined to be the more efficient transcription duration.
Fig 3-4. (a) Significance analysis of product concentrations after 4 h, 6 h, and 12 h of transcription at 2-fold dilution; (b) significance analysis of product concentrations at 5-fold dilution. ns: p > 0.05, *: p < 0.05, **: p < 0.01.
The results showed that a transcription duration of 6 hours enabled relatively high yield in a short time. By optimizing the template with 3-tandem terminators and reverse-designed 3-tandem terminators, the phenomenon of transcription read-through was reduced. Compared with the template without terminators, the yield of dsRNA synthesized via in vitro transcription using the 3-tandem terminator template was increased.
Single-target dsRNA Stage
▼For the verification of dsCYP450 and dsCP, nuclease-free water was used as the control, and aphid nymphs were treated with the corresponding dsRNA solutions at 800 ng/μl. The aphid pretreatment, preparation of feeding devices, and culture conditions were all consistent with those of the CHS dsRNA experiment, ensuring the efficiency of dsRNA absorption and the comparability of data. Results calculated using the same mortality correction formula showed that at 120 hours, the corrected mortality rate of aphids in the 800 ng/μl dsCP treatment group reached 30%, which was significantly higher than the 20.48% of dsCHS at the same concentration; the corrected mortality rate of the dsCYP450 treatment group further increased to 31.4%, showing the strongest lethal activity among the three single-target dsRNAs.
Fig 3-10. (a): Cumulative mortality rates of Toxoptera citricida (brown citrus aphids) treated with 400 ng/μl dsCHS and 800 ng/μl dsCHS, compared with the control group, at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72h, 84h, 96h, 108h and 120 h post-treatment. Figure (b): Cumulative mortality rates of Toxoptera citricida (brown citrus aphids) treated with 800 ng/μl dsCHS, 800 ng/μl dsCYP450, and 800 ng/μl dsCP, compared with the control group, at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72h, 84h, 96h, 108h and 120 h post-treatment.
If you are interested, please click here to learn more about our detailed experimental procedures.
Preliminary Preparation
To ensure the smooth progress of subsequent RT-qPCR experiments, we conducted a preliminary experiment in advance to verify the primers of genes involved in qPCR. However, since the genes of the species selected in our experiment have not been fully annotated, we could not ensure the specificity of the designed primers through exon-spanning design or genomic BLAST during primer screening. Instead, we could only design primers based on the cDNA templates of the corresponding genes, and our options were limited. Therefore, we synthesized a large number of primers for further screening.Following the advice of Expert Bai Yongsheng, the length of the internal reference fragment amplified by the primers used for qPCR should be similar to that of the target gene fragment to ensure consistent amplification efficiency, thereby reducing experimental errors. He suggested that we could use cDNA reverse-transcribed from the RNA of Toxoptera citricida (brown citrus aphid) as a template for conventional PCR, and preliminarily determine the specificity of the primers based on the results of 3.5% agarose gel electrophoresis to conduct the preliminary screening of primers.
Fig 3-5. Figure (a) shows the agarose gel electrophoresis results of conventional PCR for the positive reactions of primers CHS1–8 and their negative control (NC); Figure (b) shows the electrophoresis results for the positive reactions of primers sets CHS9–12 and CP1–4, along with their negative control (NC); Figure (c) shows the results for the positive reactions of primer sets CP5–7 and CYP450 1–5, along with their negative control (NC); and Figure (d) shows the results for the positive reactions of primers sets CYP450 8–13 and their negative control (NC).
Fig 3-6. Figure (e) shows the agarose gel electrophoresis results of conventional PCR for the positive reactions of primers EF1α1–8 and their negative control (NC); Figure (f) shows the electrophoresis results for the positive reactions of primers EF1α9–16 and their negative control (NC); Figure (g) shows the supplementary electrophoresis results for abnormal lane results observed during the primer validation process.
From the preliminary screening results of conventional PCR and the corresponding negative control results, we observed that some primers exhibited non-specific amplification of short fragments. Additionally, the appearance of multiple fragments in some negative control reactions indicated potential contamination of the existing primers. Therefore, we discarded the non-specific primers and replaced the contaminated ones based on the electrophoresis results. For the primers that passed the preliminary screening, we further characterized their specificity using qPCR melting curve analysis. After the characterization, the primers we finally selected are as follows:
Fig 3-7. Figure of qPCR primer screening results
Design
Through literature review and screening in the NCBI database, we identified a set of potential targets for the brown citrus aphid. Using the BLAST tool to analyze the homology of these target genes, we set a homology threshold of 30% and retained genes candidates below this threshold to minimize off-target risks. We then selected closely related species (aphid pests) with sequence homology above 85% for alignment analysis in MEGA, identifying highly conserved regions of the gene within aphid pests to pinpoint efficient siRNA target sites. Next, we screened high-scoring siRNA fragments from multiple public databases and, based on these, prioritized dsRNA fragments that maximally covered the previously selected high-scoring siRNA sequences, ultimately generating dsRNA molecules 300–500 bp in length.
Through these screening steps, we obtained a 300 bp dsCHS, a 329 bp dsCYP450, and a 363 bp dsCP spanning the full-length sequence. All resulting sequences were validated by BLAST to ensure high specificity and controllable off-target risk.
Build
Plasmid Amplification
We performed transformation and amplification of the constructed plasmid template pUC57_CHS-tri-terminator by introducing it into competent DH5α cells (an E. coli strain suitable for efficient routine plasmid cloning). The colonies grew well, and distinct single colonies were visible. We then expanded the bacterial culture to amplify the plasmid, obtaining a large quantity of the plasmid template.
Preparation of Linear Template
We verified the obtained plasmid template by double-enzyme digestion and agarose gel electrophoresis to ensure that we had the correct plasmid containing the DNA template (target band at 722 bp). After confirming the correctness of the plasmid, we performed double-enzyme digestion to prepare the linear template for transcription.
Transcription Reaction
We carried out bidirectional promoter-mediated in vitro transcription on the above-obtained linear template to obtain the target gene dsRNA. After transcription, we verified the product by agarose gel electrophoresis and successfully detected a dsRNA band at 352 bp, with minimal signs of degradation visible on the gel. We also prepared dsCYP450 and dsCP using the same workflow.
Figure3-8: Agarose gel electrophoresis verification of dsCHS transcription products,dsCHS: 300 bp,Marker (left): TAKARA DL500 DNA Marker,Marker (right): TAKARA DL1000 DNA Marker
Test
Lethal Efficiency Verification of Single-target dsRNA
To clarify the lethal effects of the dsRNAs targeting CHS, CP, and CYP450 that we designed and produced, and to identify the optimal concentration for single-target dsRNA, we first conducted a concentration gradient test on dsCHS. Using nuclease-free water as the blank control, we prepared dsCHS solutions at concentrations of 400 ng/μl and 800 ng/μl. For each group, 30 aphid nymphs in consistent physiological condition were selected (nine groups in total), and all were subjected to a 3-hour fasting pretreatment. The aphids were then transferred to feeding devices containing the corresponding dsCHS solution or nuclease-free water; the devices were preloaded with citrus shoots sterilized with 75% ethanol, allowing the aphids to acquire and absorb dsRNA by feeding on the shoots.
All feeding devices were placed in an artificial climate chamber with conditions set to 25 ± 1°C, 70 ± 5% relative humidity, and a 16:8-hour photoperiod (light:dark). The number of dead aphids was observed and recorded every 12 hours, and fresh feeding devices were replaced every two days (to ensure continuous dsRNA supply and maintain normal aphid survival). The final mortality was counted after 120 hours.
Fig 3-9. Schematic diagram of the feeding device
The results showed that calculated using the mortality correction formula: at 120 hours, the corrected mortality rate of aphids in the 400 ng/μl dsCHS treatment group was only 11.22%; in contrast, the corrected mortality rate of the 800 ng/μl dsCHS treatment group increased significantly to 20.48% at the same time point (Fig 3-10. (a)). This result not only confirms that the CHS dsRNA we designed and produced has a clear lethal effect on aphids, but also indicates that its lethal effect is concentration-dependent—with the 800 ng/μl concentration showing better efficacy. Therefore, we uniformly adopted the 800 ng/μl dsRNA concentration in the subsequent verification experiments for dsCYP450 and dsCP.
The corrected mortality rate formula is as follows:
Fig 3-10. (a): Cumulative mortality rates of Toxoptera citricida (brown citrus aphids) treated with 400 ng/μl dsCHS and 800 ng/μl dsCHS, compared with the control group, at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h,72h,84h,96h,108h and 120 h post-treatment. Figure (b): Cumulative mortality rates of Toxoptera citricida (brown citrus aphids) treated with 800 ng/μl dsCHS, 800 ng/μl dsCYP450, and 800 ng/μl dsCP, compared with the control group, at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h,72h,84h,96h,108h and 120 h post-treatment.
Silencing Efficiency Verification of Single-target dsRNA
To quantitatively assess the silencing efficiency of single-target dsRNAs on their target genes, we collected aphid samples treated with 800 ng/μl dsCHS, dsCP, dsCYP450, and the corresponding nuclease-free water control for 120 hours after completing the lethality experiment, and analyzed the expression levels of the target genes through the following experiments.
1.Extraction and Quality Testing
We extracted total RNA from aphids in each treatment group using the phenol–chloroform method. After extraction, we precisely determined the RNA concentration and purity using a NanoDrop spectrophotometer. Meanwhile, RNA integrity was assessed by 1% agarose gel electrophoresis. The results showed that the 28S rRNA and 18S rRNA bands were clear, sharp, and bright, with the brightness of the 28S rRNA band approximately twice that of the 18S rRNA band. This indicated that the extracted RNA had excellenstrong integrity, with minimal degradation or contamination, and fully met the requirements for subsequent experiments.
Figure 3-11: Results of total RNA integrity detection in aphid samples by 1% agarose gel electrophoresis. MW represents the molecular weight marker (100–5000 bp); Lanes 1–4 correspond to total RNA from different aphid samples.
2.Reverse transcription to synthesize cDNA
From the above-qualified RNA samples, we synthesized cDNA via reverse transcription using the TransScript® One-Step gDNA Removal and cDNA Synthesis SuperMix kit. The specific reaction system and conditions referred to the protocol.
3.Silencing efficiency measurement
Using EF1α as the internal reference gene, we performed qPCR analysis using the SYBR Green method. Each cDNA sample was analyzed with three technical replicates, and the reaction mixture was prepared in a 96-well plate. Using the nuclease-free water treatment group as the calibrator sample, we calculated the relative gene expression levels using the ΔΔCt method, and then derived the silencing efficiency.
The results showed that the CHS gene silencing rate in the dsCHS treatment group was 27%, the CP gene silencing rate in the dsCP treatment group was 43%, and the CYP450 gene silencing rate in the dsCYP450 treatment group was 30% (Fig 3-12) These results indicate that the designed single-target dsRNAs can effectively silence the corresponding target genes and show significant differences compared with the control group.
Fig 3-12. The relative expression levels of CHS, CYP450, and CP19(CP) in Toxoptera citricida after treatment with 800 ng/μl dsCHS, dsCYP450, and dsCP for 120 h, respectively, compared to the control. Error bars represent the standard deviation (SD) derived from at least three biological replicates. All datasets were statistically analyzed using a Student's t-test (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).
Fused multi-target dsRNA
▼The results showed that after 5 days, the corrected mortality rate of aphids in the dsF2 treatment group reached 40.8%, which was significantly higher than that of the single-target dsRNA (maximum 31.4%); the corrected mortality rate in the dsF3 treatment group further increased to 44.9%, which was not only superior to the single-target dsRNA but also higher than the dual-fusion dsRNA (Fig 3-15. (a)). These findings fully confirm the synergistic effect of the dual-target RNAi strategy and further highlight the stronger insecticidal advantage of triple-target synergy.
Figure 3-15.(a): Cumulative mortality rates of brown citrus aphids treated with 800 ng/μl dual-fusion dsRNA (dsF2) or triple-fusion dsRNA (dsF3) at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72 h, 84 h, 96 h, 108 h, and 120 h, compared with the control. Figure (b): Relative expression levels of CHS and CYP450 in brown citrus aphids treated with 800 ng/μl dsF2 for 120 h, compared with the control.Error bars represent the standard deviation (SD) derived from at least three biological replicates. All datasets were statistically analyzed using a Student's t-test (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001). Figure (c): Relative expression levels of CHS, CYP450, and CP19(CP) in brown citrus aphids treated with 800 ng/μl dsF3 for 120 h, compared with the control.Error bars represent the standard deviation (SD) derived from at least three biological replicates. All datasets were statistically analyzed using a Student's t-test (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).
If you are interested, please click here to learn more about our detailed experimental procedures.
Design
In the first-generation design, we successfully constructed dsRNA targeting a single key gene of the brown citrus aphid. After discussions with Dr. Chen Ruoyu, we decided to construct a multi-target fusion dsRNA for the second generation, aiming to enhance the lethal effect by targeting multiple key genes of the aphid. Based on the results of the first-generation experiments, we selected short sequences (60–100 bp) with high specificity and high conservation, and fused these optimized short fragments in a direct tandem arrangement. RNAfold prediction analysis was used to avoid introducing complex structures or repeat sequences at the gene fusion sites. Ultimately, we obtained a dual-target fusion dsRNA (dsF2, targeting CHS and CYP450) and a triple-target fusion dsRNA (dsF3, targeting CHS, CYP450, and CP). The dsRNAEngineer tool was used to ensure no off-target risk.
Build
Preparation of Linear Template
We transformed the plasmids pUC57_dsF2 and pUC57_dsF3, which contain the target dsDNA templates, into competent DH5α cells for plasmid amplification, yielding large quantities of the plasmid templates harboring the target dsDNA. Double-enzyme digestion followed by agarose gel electrophoresis showed clear, correct bands without any non-specific bands, confirming that we had successfully prepared high-quality DNA templates suitable for subsequent transcription.
Transcription Reaction
We used the resulting linear templates for in vitro transcription, followed by agarose gel electrophoresis verification, which confirmed that we had obtained dsRNA products dsF2 and dsF3 of the expected length.
Fig 3-13. Agarose gel electrophoresis verification of dsF2 and dsF3 transcription products.(a)dsF2:402bp，marker(left):sangon 1000bp marker，marker(right):BBI DNA Marker A (25~500 bp) (b)dsF3：391bp，marker:sangon 1000bp marker
Test
Lethal Efficiency Verification of Fused multi-target dsRNA
To investigate the synergistic lethal effect of fused dsRNA, we used nuclease-free water as the control and, based on the optimal concentration of 800 ng/μl determined in the single-target dsRNA experiments, treated aphid nymphs with solutions of dual-fusion dsF2 and triple-fusion dsF3, respectively. Each group contained 30 aphids, with a total of nine groups. After a 3-hour fasting pretreatment, the aphids were transferred to the corresponding feeding devices. The feeding devices and culture conditions were identical to those in the single-target dsRNA experiments. During the experiment, aphid survival and mortality were recorded every 12 hours, and fresh devices were replaced every two days, with continuous observation for 5 days (120 h). During the experiment, it was observed that some aphids turned red all over upon death, and some died during molting.
Fig 3-14. (a) Observed under a stereomicroscope at 40× magnification, after brown citrus aphids were treated with 800 ng/μl triple-fusion multi-target RNA (dsFA3), the individual on the left is a dead one with a whole-body reddening, and the individual on the right is a normal one that is black; (b) Observed under a stereomicroscope at 40× magnification, an individual of brown citrus aphid that died during the molting process after being treated with 800 ng/μl triple-fusion multi-target RNA (dsFA3).
The results showed that after 5 days, the corrected mortality rate of aphids in the dsF2 treatment group reached 40.8%, which was significantly higher than that of the single-target dsRNA (maximum 31.4%); the corrected mortality rate in the dsF3 treatment group further increased to 44.9%, which was not only superior to the single-target dsRNA but also higher than the dual-fusion dsRNA (Fig 3-15. (a)). These findings fully confirm the synergistic effect of the dual-target RNAi strategy and further highlight the stronger insecticidal advantage of triple-target synergy.
Silencing Efficiency Verification of Fused multi-target dsRNA
Following the same workflow of RNA extraction by the phenol–chloroform method, cDNA synthesis, and qRT–PCR detection, we verified the silencing effects of dual-fusion and triple-fusion dsRNA. The results showed that after dsF2 treatment, the silencing rate of the CYP450 gene was 31.5% (Fig 3-15. (b)); after dsF3 treatment, the silencing rates of the CHS, CYP450, and CP genes were 35.3%, 20%, and 45%, respectively(Fig 3-15. (c)). These results further demonstrate the role of fused multi-target dsRNA in gene silencing and provide a molecular basis for its synergistic insecticidal effect.
Fig 3-15. (a): Cumulative mortality rates of brown citrus aphids treated with 800 ng/μl dual-fusion dsRNA (dsF2) or triple-fusion dsRNA (dsF3) at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72 h, 84 h, 96 h, 108 h, and 120 h, compared with the control. Figure (b): Relative expression levels of CHS and CYP450 in brown citrus aphids treated with 800 ng/μl dsF2 for 120 h, compared with the control. Error bars represent the standard deviation (SD) derived from at least three biological replicates. All datasets were statistically analyzed using a Student's t-test (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001). Figure (c): Relative expression levels of CHS, CYP450, and CP19(CP) in brown citrus aphids treated with 800 ng/μl dsF3 for 120 h, compared with the control. Error bars represent the standard deviation (SD) derived from at least three biological replicates. All datasets were statistically analyzed using a Student's t-test (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001).
Multitandem short hairpin RNA stage
▼The results showed that after 120 h (5 days), the corrected mortality rate of aphids in the tri-shRNA treatment group reached 49.5% (Fig 3-18.), demonstrating effective killing of aphids and showing better insecticidal efficacy compared with the triple-fusion dsRNA (44.9%), thus providing a technical basis for subsequent RNAi product iteration.
Figure 3-18.Cumulative mortality rates of brown citrus aphids at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72 h, 84 h, 96 h, 108 h, and 120 h after treatment with 800 ng/μl tri-shRNA and in the control group.
If you are interested, please click here to learn more about our detailed experimental procedures.
Design
Due to the susceptibility of dsRNA to degradation and its relatively high off-target risk, we developed the third-generation RNA molecule design—shRNA—based on the multi-target approach. shRNA improves RNA stability and reduces off-target effects through more precise sequence design. We used the TTCAAGAGA universal hairpin sequence and linked multiple shRNA hairpin units with a linker, successfully designing a tri-shRNA targeting CHS, CP, and CYP450. We performed RNA secondary structure simulations using RNAfold to confirm the structural independence of each shRNA unit and analyzed off-target effects using dsRNAEngineer, ultimately obtaining tri-shRNA.
Figure 3-16.tri-shrna二级结构预测Secondary structure prediction of tri-shRNA
Build
Plasmid transformation
We transformed the plasmid containing the target dsDNA template into competent cells—DH5α, an E. coli strain suitable for efficient plasmid transformation—for subsequent plasmid amplification.
Plasmid Amplification
We inoculated the transformed E. coli via the streak plate method to obtain single colonies. Subsequently, we expanded the number of bacteria by culturing the obtained single colonies in a shaking flask, thereby obtaining more plasmids containing the target template for subsequent template preparation.
Preparation of Linear Template
We used restriction endonucleases to act on the restriction site located after the target sequence on the plasmid, obtaining restriction digestion products for the subsequent transcription system. The single-enzyme digestion gap plays a role as a physical terminator in the subsequent transcription experiment, which is also used for the subsequent transcription system. The gel electrophoresis results showed that we obtained a single restriction digestion product with the correct length.
Transcription Reaction
We added the obtained linear template to the system of the transcription kit, and obtained tri-shRNA products with the expected length through transcription.
Fig 3-17. Marker ( left ) : DL5000 DNA Marker 1 : negative control 2、3、4 : tri-shRNA transcription product——213bp Marker ( right ): D2000 DNA Marker
Test
Lethal Efficiency Verification of Multitandem short hairpin RNA
Using nuclease-free water as the control, we treated aphid nymphs with a multitandem short hairpin RNA (tri-shRNA) solution at a concentration of 800 ng/μl. The feeding devices and culture conditions were the same as in the single-target dsRNA experiments, and we continuously observed and recorded aphid survival and mortality. The results showed that after 120 h (5 days), the corrected mortality rate of aphids in the tri-shRNA treatment group reached 49.5% (Fig 3-18) demonstrating effective killing of aphids and showing better insecticidal efficacy compared with the triple-fusion dsRNA (44.9%), thus providing a technical basis for subsequent RNAi product iteration.
Fig 3-18. Cumulative mortality rates of brown citrus aphids at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72 h, 84 h, 96 h, 108 h, and 120 h after treatment with 800 ng/μl tri-shRNA and in the control group.
bi-Artificial miRNA stage
▼The results showed that after 120 h (5 days), the corrected mortality rate of aphids in the bi-amiRNA treatment group reached as high as 60.2% (Fig 3-22.), demonstrating the optimal insecticidal efficacy among the four generations of RNA molecules.
Fig 3-22. Cumulative mortality rates of brown citrus aphids at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72 h, 84 h, 96 h, 108 h, and 120 h after treatment with 800 ng/μL bi-amiRNA and in the control group (CTRL).
If you are interested, please click here to learn more about our detailed experimental procedures.
Design
We aimed to more effectively utilize endogenous RNAi-related proteins in organisms, and based on this, we designed the fourth-generation RNA molecule—multitandem artificial miRNA (amiRNA). Compared with the above three generations of RNA molecules (which induce RNAi through the exogenous siRNA pathway), this fourth-generation molecule uses the natural miRNA precursor scaffold of organisms, which can trigger the endogenous microRNA processing mechanism in eukaryotes to further improve RNAi efficiency. We screened the miRBase database and obtained two natural miRNA scaffolds from aphids—api-mir-71 and api-mir-3017a. The affinity of the scaffolds ensures the processing efficiency of the RNA molecules.
Figure 3-19.Skeletal structures of api-mir-71 (a) and api-mir-3017a (b), where the pink parts represent the final mature miRNA sequences.
We integrated the previously screened CYP450 siRNA sequence into api-mir-71 and the CHS siRNA sequence into api-mir-3017a, then linked them using a universal linker (AGGCAT) to obtain our artificial miRNA (amiRNA). Similarly, we performed RNA secondary structure simulation via RNAfold to ensure the independence of each hairpin unit.
Figure 3-20. Prediction of the secondary structure of bi-amiRNA
Build
Plasmid transformation
We transformed the plasmid containing the target dsDNA template into competent cells—DH5α, an E. coli strain suitable for efficient plasmid transformation—for subsequent plasmid amplification.
Plasmid Amplification
We inoculated the transformed E. coli using the streak plate method to obtain single colonies. Subsequently, we increased the bacterial count by culturing the obtained single colonies in a shaking flask, thereby acquiring more plasmids containing the target template for subsequent template preparation.
Preparation of Linear Template
We used restriction endonucleases to act on the restriction site located downstream of the target sequence on the plasmid, obtaining restriction digestion products for the subsequent transcription system. The single-enzyme digestion gap functions as a physical terminator in the subsequent transcription experiment and is also used in the subsequent transcription system. Gel electrophoresis results showed that we obtained a single restriction digestion product with the correct length.
Transcription Reaction
We added the obtained linear template to the system of the transcription kit, and obtained bi-amiRNA products with the expected length through transcription.
Figure 3-21. Agarose gel electrophoresis analysis of bi-amiRNA transcription products.Marker ( left ) : DL5000 DNA Marker；1 : negative control；2、3、4 : bi-amiRNA transcription product——157bp；Marker ( right ) : D2000 DNA Marker
Test
Lethal Efficiency Verification of Artificial miRNA (amiRNA)
Using nuclease-free water as the control, we treated young aphids with an artificial miRNA (amiRNA) solution at a concentration of 800 ng/μL. The feeding devices and culture conditions were consistent with those in the dsRNA experiment, and we continuously observed and recorded the survival and mortality of the aphids. The results showed that after 120 h (5 days), the corrected mortality rate of aphids in the bi-amiRNA treatment group reached as high as 60.2% (Fig 3-22), demonstrating the optimal insecticidal efficacy among the four generations of RNA molecules.
Figure 3-22.Cumulative mortality rates of brown citrus aphids at 6 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72 h, 84 h, 96 h, 108 h, and 120 h after treatment with 800 ng/μL bi-amiRNA and in the control group (CTRL).
Summary
In this study, by optimizing the high-efficiency in vitro dsRNA production system, it was found that the use of multi-terminator templates significantly increased the yield of dsRNA, and 6 hours was determined as the optimal transcription duration. Furthermore, four generations of RNA molecules were designed and their insecticidal effects on citrus aphids were verified. The data in the table below clearly shows the mortality rate and silencing efficiency of each generation of RNA molecules. Among them, the fourth-generation artificial miRNA (bi-amiRNA) exhibited the highest aphid mortality rate after 120 hours, reaching 60.2%. These results provide an important basis for the development of efficient and specific RNAi-based control strategies against citrus aphids.
Table: Summary Table of RNAi Therapeutic Outcomes
Delivery System
Background Information
Virus-like particles (VLPs) are hollow protein nanoparticles self-assembled from one or more structural proteins of viruses. They are highly similar to natural viruses in size, shape, and surface structure, but the most fundamental difference is that they completely lack the viral genetic material (DNA or RNA); thus, they possess no replication or infection capabilities and exhibit high biosafety.
The core advantage of VLPs stems from their unique property of being "virus-like but not viruses": the highly ordered and repetitively arranged antigenic epitopes on their surface can be efficiently recognized by the immune system, thereby triggering strong humoral and cellular immune responses—making them an ideal platform for vaccine development. Currently, vaccines based on VLP technology, such as human papillomavirus (HPV) vaccines and hepatitis B vaccines, have been successfully marketed and widely used worldwide. Beyond prophylactic vaccines, VLPs also show great potential as multifunctional nanocarriers. Their internal cavity can load and protect nucleic acid drugs (mRNA, siRNA, shRNA), and they can achieve cell- or tissue-targeted delivery through engineering modifications like surface display technology. Therefore, they hold broad prospects in fields such as targeted drug therapy and gene editing. Among them, VLPs derived from MS2 viruses can specifically bind to nucleic acids with pac sequences and rapidly assemble into a polymer composed of 180 capsid proteins (CP).
This year, the SZU-China team successfully constructed a VLP vector platform using MS2 phage capsid proteins. Through surface display technology, they co-displayed the TAT transmembrane peptide (which enhances cell penetration) and the GBP3.1 peptide (which specifically targets aphid intestinal cells) on the particle surface, thereby endowing this VLP system with the function of cross-species targeted delivery. In this section, we will present the characterization of the engineered VLPs and their stress resistance.
Implementation Objectives
We aim to construct a virus-like particle (VLP)-based RNA delivery system to address the key challenge of RNA being easily degraded during delivery to aphids. We will prepare VLPs capable of efficiently packaging target RNA through genetic engineering methods and focus on evaluating their capsid integrity and resistance to exogenous nucleases, thereby providing experimental evidence for the development of a stable and efficient RNA delivery platform.
Verification process
Plasmid construction and Characterization of Protein
▼The results showed that the eluate containing the target protein exhibited a distinct bright band at the predicted position, which is consistent with our prediction that the His-tagged VLPs have a molecular weight of approximately 29 kDa. This preliminarily confirms that we have successfully obtained the His-tagged VLP proteins.
Fig 4-5. SDS-PAGE analysis of the purified target protein expressed by pACYCDuet_1-CP-Histag-CP-amiRNA stained with Coomassie Brilliant Blue. Lane MW is the protein molecular weight marker; Lane 1 is the bacterial culture supernatant, Lane 2 is the flow-through during purification; Lanes 3 and 4 are the washing solutions in the purification process of the target protein, and Lanes 5–9 are the purified eluates. The theoretical molecular weight of the target protein is 28333.00 Da (approximately 28 kDa), and the actual size of the electrophoretic band is about 28 kDa, which is consistent with the theoretical molecular weight.
If you are interested, please click here to learn more about our detailed experimental procedures.
Plasmid construction
1.Construction of plasmid pACYCDuet_1-CP-Histag-CP-amiRNA
The construction of plasmid pACYCDuet_1-CP-Histag-CP-amiRNA is shown in Fig 4-1. In this construction, the MS2 capsid protein dimer–6×His tag fusion gene (CP–6×His–CP) was placed downstream of a T7 promoter. Simultaneously, two RNA interference sequence fragments targeting aphid genes—namely, chitin synthase (api-mir-71) and CYP450 (api-mir-3017a)—were cloned into specific sites of the plasmid. A 4,640 bp recombinant plasmid was successfully constructed. This plasmid can express the MS2 capsid protein and transcribe RNA carrying target gene sequences—specifically, a bifunctional amiRNA capable of silencing both the chitin synthase gene (api-mir-71) and the cytochrome CYP450 gene.
Fig 4-1. Schematic diagram of plasmid pACYCDuet_1-CP-Histag-CP-amiRNA.
2.Construction of plasmid pET28a_CP-GBP3.1-CP-TAT-Histag
The construction of plasmid pET28a_CP-GBP3.1-CP-TAT-Histag aimed to obtain MS2 capsid proteins displaying functional peptides on their surface. As shown in Fig 4-2, we constructed plasmid pET28a_CP-GBP3.1-CP-TAT-Histag (6076 bp). In this plasmid, a fusion gene sequence—MS2 capsid protein (CP)–aphid intestinal targeting peptide (GBP3.1)–second MS2 capsid protein (CP)—was obtained via gene synthesis. The C-terminus of this sequence was linked to the TAT cell-penetrating peptide via a flexible linker peptide, and a 6×His purification tag was added immediately after the TAT to facilitate affinity chromatography purification. This fusion gene was cloned downstream of the T7 promoter. Therefore, upon induction, the plasmid can express a single-chain version of the capsid protein dimer, whose surface can functionally display both the GBP3.1 targeting peptide and the TAT cell-penetrating peptide.
Fig 4-2. Schematic diagram of plasmid pET28a_CP-GBP3.1-CP-TAT-Histag.
3.Construction of plasmid pET28a_GBP3.1-eGFP-TAT-HisTag
To obtain a vector capable of expressing a multifunctional fusion protein incorporating the aphid intestinal targeting peptide (GBP3.1), the enhanced green fluorescent protein (eGFP) reporter gene, and the HIV-1 trans-membrane peptide (TAT), we constructed plasmid pET28a_GBP3.1-eGFP-TAT-HisTag (6214 bp). As shown in Figure X, the complete fusion gene sequence—GBP3.1-linker-eGFP-linker-TAT-linker-6×HisTag—was obtained via gene synthesis and cloned into the multiple cloning site (MCS) of the pET-28a(+) vector, placing its expression under the control of the T7/lac hybrid promoter. Under IPTG-induced conditions, this plasmid can be expressed at high efficiency in an E. coli host that expresses T7 RNA polymerase (e.g., BL21(DE3)), producing a fusion protein with a C-terminal 6×His tag for subsequent purification. The inclusion of eGFP provides a visual tracking tool for protein expression localization and the purification process.
Fig 4-3. Schematic diagram of plasmid pET28a_CP-GBP3.1-CP-TAT-Histag.
4.Construction of plasmid pET28a_eGFP
As a control, we constructed plasmid pET28a_eGFP (6085 bp). As shown in Figure 2, the eGFP gene sequence was cloned into the multiple cloning site (MCS) of the pET-28a(+) vector, placing its expression under the control of the T7/lac hybrid promoter. This plasmid is used to express the enhanced green fluorescent protein alone in the corresponding host strain.
Fig 4-4. Schematic diagram of plasmid pET28a_eGFP.
Characterization of the single-chain version of the capsid protein dimer
After constructing the above plasmids, these recombinant plasmids were transformed into Escherichia coli BL21 (DE3), and positive monoclonal colonies were selected on antibiotic-resistant plates for expansion culture and colony PCR. Our specific primers successfully amplified the corresponding target bands, confirming the successful transformation of the plasmids. After inducing protein expression with IPTG, the His-tagged virus-like particle (VLP) proteins were purified by nickel affinity chromatography. The results after SDS-polyacrylamide gel electrophoresis (SDS-PAGE) are shown in (Fig 4-5. and Fig 4-6.). The results showed that the eluate containing the target protein exhibited a distinct bright band at the predicted position, which is consistent with our prediction that the His-tagged VLPs have a molecular weight of approximately 29 kDa. This preliminarily confirms that we have successfully obtained the His-tagged VLP proteins.
Fig 4-5. SDS-PAGE analysis of the purified target protein expressed by pACYCDuet_1-CP-Histag-CP-amiRNA stained with Coomassie Brilliant Blue. Lane MW is the protein molecular weight marker; Lane 1 is the bacterial culture supernatant, Lane 2 is the flow-through during purification; Lanes 3 and 4 are the washing solutions in the purification process of the target protein, and Lanes 5–9 are the purified eluates. The theoretical molecular weight of the target protein is 28333.00 Da (approximately 28 kDa), and the actual size of the electrophoretic band is about 28 kDa, which is consistent with the theoretical molecular weight.
Fig 4-6. SDS-PAGE analysis of the purified target protein expressed by pET28a_CP-GBP3.1-CP-TAT-Histag, stained with Coomassie Brilliant Blue. Lane MW represents the protein molecular weight marker; Lane 1 is the bacterial culture supernatant, Lane 2 is the flow-through during the purification process; Lanes 3 and 4 are the bacterial pellets, and Lanes 5–9 are the purified eluates. The theoretical molecular weight of the target protein is 31042.25 Da (approximately 31 kDa), while the actual size of the electrophoretic band is about 30 kDa, which is consistent with the theoretical molecular weight.
Characterization of MS2 VLPs
▼In this experiment, we characterized the assembly of MS2 VLPs particles. Through TEM detection, individual particles showed uniform spherical structures with clear boundaries. These particle morphologies demonstrate that amiRNA and proteins can effectively assemble into VLPs nanoparticles.
Fig 4-8. Transmission Electron Microscopy (TEM) images of purified histidine-tagged MS2 phage-like particles (histidine-tagged VLPs). The scale bar represents 100 nanometers.
If you are interested, please click here to learn more about our detailed experimental procedures.
1.Characterization of VLPs
To verify whether histidine-tagged virus-like particles (VLPs) successfully encapsulate the target RNA inside their capsids, we designed and conducted a combined verification experiment. First, the samples were treated with nucleases for gradient time periods; subsequently, agarose gel electrophoresis was used to analyze and record the retention of the target RNA. After the nucleic acid detection was completed, the same gel was stained with Coomassie Brilliant Blue R250 to detect the presence of VLP capsid proteins. The experimental results showed that the target RNA could still be stably detected after nuclease treatment, and the capsid protein staining result was positive(Fig 4-7.). This confirms that the target RNA was successfully encapsulated inside the protein capsids of the VLPs.
Fig 4-7. Agarose gel images for characterization of the assembled products. We first stained with Super Red nucleic acid dye to detect RNA (Panel A), followed by Coomassie Brilliant Blue R-250 staining to detect MS2 coat protein (Panel B). Lane MW (the leftmost lane in both Panels A and B) shows the DNA ladder. In Panel A, Lanes 1–4 contain 50 µM virus-like particles (VLPs) samples treated with 100 U/mL Benzonase nuclease at 37 °C for 90 min, 60 min, 30 min, and 0 min, respectively; in Panel B, Lanes 5–8 correspond to Lanes 1–4 in Panel A. The results show that Lanes 1–4 corresponding to VLPs exhibit a bright RNA band above 1000 bp; due to the large overall size of RNA encapsulated within the capsid proteins, its migration speed in the gel is slower. It can be observed that the shorter the nuclease treatment time, the more distinct the band becomes; after nuclease treatment, VLP aggregates depolymerizes and releases amiRNA, thus a bright amiRNA band can be observed in the 100–200 bp range. It can be observed that the shorter the nuclease treatment time, the brighter the band; notably, under the condition of 0 min nuclease treatment, RNA is encapsulated within the protein capsid, making it difficult to observe discrete RNA bands. A comparison of the lanes patterns in Panels A and B confirms that the target RNA is successfully encapsulated inside the protein capsids of the VLPs.
2.Transmission electron microscopy (TEM) experiments
To verify whether amiRNA and capsid proteins can assemble into virus-like particles (VLPs) from a physical morphological perspective, we performed negative-staining transmission electron microscopy (TEM) analysis on the purified samples. Samples with a protein concentration of 50 μM (determined by the Bradford method) were dropped onto a hydrophilically treated ultra-thin support film copper grid, stained negatively with phosphotungstic acid (PTA), and observed under an accelerating voltage of 80 kV. The TEM results are shown in Fig 4-8.: in the low-magnification field of view in Fig 4-8. A, large-scale aggregates formed by capsid proteins can be observed, and this morphology is comparable to the research results of LaNell A. Williams et al. (2024); the high-magnification field of view in Fig 4-8. B clearly shows typical VLP clusters, in which individual particles exhibit a uniform spherical structure with clear boundaries. The morphology of these particles confirms that amiRNA and proteins can effectively assemble into VLP nanoparticles.
Fig4-8. Transmission Electron Microscopy (TEM) images of purified histidine-tagged MS2 phage-like particles (histidine-tagged VLPs). The scale bar represents 100 nanometers.
Verification of MS2 VLPs Function
▼The results showed that, after treatment with Benzonase Nuclease at 37°C for 10 min, the naked RNA was completely degraded; under the same conditions, the RNA encapsulated within MS2 capsid proteins remained significantly protected, indicating that VLP particles exhibit remarkable stability against nucleases.
Fig 4-9. Agarose gel image showing the nuclease resistance of the assembled products.The gel was stained with Super Red nucleic acid dye to detect amiRNA. Lanes MW (leftmost and rightmost lanes) contain a DNA ladder (molecular weight marker). Lanes 1–4 contain 50 µM virus-like particles (VLPs) treated with 100 U/mL Benzonase nuclease at 37 °C for 0 min, 10 min, 20 min, and 30 min, respectively. Lanes 6–9 contain 500 nM amiRNA treated under the same conditions as the VLPs (100 U/mL Benzonase nuclease, 37 °C) for 0 min, 10 min, 20 min, and 30 min, respectively. Lane 5 is a blank control, and Lane 10 is a control containing capsid protein (CP) dimers without amiRNA.The results show that Lanes 1–4, corresponding to VLPs, exhibit a bright RNA band above 1000 bp. Because the RNA encapsulated within the capsid proteins has a large overall size, its migration speed in the gel is slower. It can be observed that the shorter the nuclease treatment time, the more distinct the band becomes. After nuclease treatment, VLP aggregates depolymerize and release amiRNA, thus a bright amiRNA band can be observed in the 100–200 bp range. Again, the shorter the nuclease treatment time, the brighter this band appears. Notably, under the condition of 0 min nuclease treatment, the RNA is encapsulated within the protein capsid, making it difficult to observe discrete RNA bands. Furthermore, the RNA content in Lane 1 (VLPs treated for 0 min) is higher than that in Lane 6 (amiRNA treated for 0 min). This is presumably because RNA is prone to degradation in the natural environment, and the amiRNA degraded rapidly during loading and electrophoresis. A comparison of the two groups of lanes reveals that RNA encapsulated in VLPs still exhibits a better protective effect in a high-nuclease environment.
If you are interested, please click here to learn more about our detailed experimental procedures.
Investigation of the Protective Effect of VLP Nanoparticles on amiRNA
To evaluate the protective effect of VLP nanoparticles on amiRNA, we performed a time-gradient experiment using Benzonase Nuclease. The experimental setup was as follows: the control group contained 750 ng of naked RNA, and the experimental group contained 50 µM His-tagged VLPs. Both groups were treated with 100 U/mL Benzonase Nuclease at 37°C for 0 min, 10 min, 20 min, and 30 min. The remaining RNA content was then detected by agarose gel electrophoresis (Fig 4-9.).The results showed that, after treatment with Benzonase Nuclease at 37°C for 10 min, the naked RNA was completely degraded; under the same conditions, the RNA encapsulated within MS2 capsid proteins remained significantly protected, indicating that VLP particles exhibit remarkable stability against nucleases.
Fig 4-9. Agarose gel image showing the nuclease resistance of the assembled products.The gel was stained with Super Red nucleic acid dye to detect amiRNA. Lanes MW (leftmost and rightmost lanes) contain a DNA ladder (molecular weight marker). Lanes 1–4 contain 50 µM virus-like particles (VLPs) treated with 100 U/mL Benzonase nuclease at 37 °C for 0 min, 10 min, 20 min, and 30 min, respectively. Lanes 6–9 contain 500 nM amiRNA treated under the same conditions as the VLPs (100 U/mL Benzonase nuclease, 37 °C) for 0 min, 10 min, 20 min, and 30 min, respectively. Lane 5 is a blank control, and Lane 10 is a control containing capsid protein (CP) dimers without amiRNA.The results show that Lanes 1–4, corresponding to VLPs, exhibit a bright RNA band above 1000 bp. Because the RNA encapsulated within the capsid proteins has a large overall size, its migration speed in the gel is slower. It can be observed that the shorter the nuclease treatment time, the more distinct the band becomes. After nuclease treatment, VLP aggregates depolymerize and release amiRNA, thus a bright amiRNA band can be observed in the 100–200 bp range. Again, the shorter the nuclease treatment time, the brighter this band appears. Notably, under the condition of 0 min nuclease treatment, the RNA is encapsulated within the protein capsid, making it difficult to observe discrete RNA bands. Furthermore, the RNA content in Lane 1 (VLPs treated for 0 min) is higher than that in Lane 6 (amiRNA treated for 0 min). This is presumably because RNA is prone to degradation in the natural environment, and the amiRNA degraded rapidly during loading and electrophoresis. A comparison of the two groups of lanes reveals that RNA encapsulated in VLPs still exhibits a better protective effect in a high-nuclease environment.
Validation of Delivery Function
Previous studies have shown that MS2-encapsulated nucleic acid molecules exhibit poor ability to escape from endosomes after cellular phagocytosis (Reference). Furthermore, unmodified VLP-RNAi complexes distribute randomly within aphids, making it difficult for most small RNA molecules to be efficiently taken up and exert their effects. To enhance the uptake and delivery efficiency of VLPs in the aphid gut (Reference), we systematically fused two functional peptides to the surface of the MS2 capsid protein: GBP3.1 (a 12-amino-acid intestinal binding peptide: TCSKKYPRSPCM) and TAT (an 11-amino-acid cell-penetrating peptide: YGRKKRRQRRR).
We transformed the constructed plasmid into Escherichia coli BL21 (DE3) and expanded the culture. Following induction with IPTg, the His-tagged CP-GBP3.1-TAT fusion protein was purified using nickel affinity chromatography. SDS-PAGE electrophoresis results showed that the His-tagged CP-GBP3.1-TAT fusion protein was successfully expressed and purified (Fig 4-6.).
Western Blot Detection of Tree Tissues Following Vascular Injection of VLPs
To verify whether virus-like particles (VLPs) can be efficiently transported to leaves and stems via vascular tissue injection in trees, we designed an experimental system using red ink as a tracer molecule and VLPs as the functional carrier, referencing the principles of trunk injection technology. Purified VLP proteins were mixed with red ink at a specific ratio (for the preparation method, see Table 4-1.). A continuous injection device (Fig 4-10.) was used to perform targeted injection into the vascular tissue of citrus trees for three days. This device can provide a constant propelling force for the agent and record the injection dosage.
After the injection, young stem and leaf tissues were collected from the trees, and total protein was extracted for Western Blot (WB) analysis (Fig 4-11). The results showed that clear target protein bands were detected in leaf samples from all sampling sites. This indicates that VLPs can be transported to the apical young tissues along with the tree's transpiration stream, and the injection device achieved controlled and sustained delivery.
Table 4-1 Injection material
Fig 4-10. Schematic diagram of the injection device
Fig 4-11. Analysis of VLP distribution in the young stems of citrus trees after injection with the mixed agent of red ink and VLPs via Western Blot. MW represents the protein molecular weight marker; NC is the negative control of untreated young stems; Lane Stem-injection is the collected protein sample from young stems. The results show that a bright band is visible at approximately 28 kDa in the experimental group sample, while there is no corresponding band in the NC lane. This indicates that VLPs can be effectively delivered via vascular injection and distributed in young stem tissues.
To verify the tissue targeting and cell penetration efficiency of GBP3.1 and TAT transmembrane peptides in aphids, we designed a fusion protein expression system using enhanced Green Fluorescent Protein (EGFP) as the reporter gene. By constructing prokaryotic expression vectors of the pET series, we expressed the GBP3.1-EGFP-TAT fusion protein, which was then purified via His-tag affinity chromatography. The purified fusion protein was used to treat aphids through a feeding method, with a separate EGFP group set as the negative control.Observations were conducted respectively via in vivo imaging using a fluorescence microscope, and via observation of tissue sections subjected to immunofluorescent staining with an enhanced green fluorescent protein (EGFP) antibody using a laser confocal microscope.
The results showed that, compared with the control group, the TAT-EGFP and GBP3.1-EGFP treatment groups exhibited significantly enhanced retention of green fluorescent signals in aphids. This indicates that both TAT and GBP3.1 can effectively carry EGFP to penetrate the tissue barriers of aphids and enter the interior of cells, with GBP3.1 showing a certain degree of tissue targeting specificity. These results visually confirm the effectiveness and feasibility of GBP3.1 and TAT as delivery carriers in living aphids, and provide key evidence for the effectiveness of functional peptides for the subsequent construction of a delivery system that combines MS2 virus-like particles (MS2 VLPs) loaded with target RNA and the GBP3.1 and TAT transmembrane peptides. For more information and results regarding the delivery system, you can visit the Proof of Concept webpage to learn more about this section in detail.
Discussion
This collection of studies provides solid theoretical and experimental evidence for the core hypothesis—“MS2 Virus-Like Particles (VLPs) modified with GBP3.1 (aphid midgut-targeting peptide) and TAT (cell-penetrating peptide) enable efficient delivery of silencing small RNAs”—through systematic experiments spanning plasmid construction, VLP characterization, and functional validation.Based on the results in this section and the results regarding the tracking-based validation of the functions of GBP3.1 and the TAT cell-penetrating peptide (CPP) on the proof of concept (PoC) wiki page, we believe that achieving this objective is fully feasible.
The MS2 VLP system constructed in this study exhibits distinct advantages. Compared to traditional delivery methods (e.g., direct dsRNA spraying or unmodified VLPs), this system more effectively protects nucleic acids from degradation and significantly enhances cell-specific uptake efficiency due to the targeting capability of GBP3.1. In addition, unlike chemically synthesized polymeric nanocarriers, all components of this system possess excellent biodegradability and biocompatibility. Notably, VLP technology platforms have been extensively explored for drug delivery in humans, indicating that the system developed herein is an environmentally and human-friendly technology with broad application potential across multiple fields.
Currently, MS2 VLPs have been successfully self-assembled in Escherichia coli (E. coli). Transmission Electron Microscopy (TEM) observations revealed that their morphology does not consist of uniformly dispersed individual spherical particles; instead, they tend to aggregate in clusters. This finding is consistent with the report by LaNell A. Williams et al. (2024), who identified the stoichiometric ratio of components in the assembly system as a key factor determining VLP morphology. We hypothesize that the complexity of the intracellular environment in E. coli makes MS2 VLP difficult to precisely control this ratio, leading to the formation of capsid clusters. This represents one of the limitations of using in vivo expression systems for MS2 VLP production. Given this, the dimers of recombinant MS2 VLPs may hold great potential for RNA packaging in vitro. It is also crucial to evaluate the RNA protection efficacy of both VLP capsid clusters and dispersed morphologies. Importantly, the conformation of proteins fused to MS2 VLP dimers and their insertion sites appear to exert a significant influence on assembly stability. The final morphology of MS2 VLPs strongly depends on the conformation of the chimeric protein and its insertion site on the capsid, as well as the density of the targeting peptide. In-depth investigation into the regulatory mechanisms of these factors on VLP assembly will greatly expand the compatibility of MS2 VLPs with heterologous proteins. Notably, the use of covalent ligation systems such as SpyTag/SpyCatcher can effectively circumvent assembly challenges caused by the size or properties of fusion proteins, making this a highly promising solution.
In conclusion, the MS2 VLP system is poised to evolve into a modular RNA-based biopesticide delivery platform. Its application prospects are not limited to aphid control; by simply replacing the targeting peptide (e.g., with specific ligands targeting other pests or pathogens), the platform can be rapidly adapted to diverse biological scenarios. Therefore, establishing a modified modular platform for MS2 VLPs holds significant potential, demonstrating remarkable versatility and flexibility.
Fungal Contact-Killing Module
Background Information
As a type of natural entomopathogenic fungus, Metarhizium anisopliae (commonly known as green muscardine fungus) can germinate after its spores come into contact with the cuticle of brown citrus aphids. The germinated spores then differentiate to form a specialized infection structure—appressoria—which further penetrates the aphid’s cuticle, invades the hemolymph, and proliferates extensively. During this process, M. anisopliae secretes metabolites such as destruxins, which can disrupt the aphid’s tissue structure and ultimately lead to the aphid's death.Therefore, M. anisopliae has been registered as an environment-friendly biopesticide in many countries worldwide, and is currently widely used in the agricultural production of various crops such as citrus and rice.
Fig 5-1. Flowchart of Metarhizium Infecting Brown Citrus Aphids
However, the infection process of Metarhizium takes a certain amount of time, while brown citrus aphids reproduce extremely rapidly—this makes traditional biopesticides unable to completely eliminate the pests. In contrast, using genetic engineering to modify fungal strains for expressing host-derived small RNAs can effectively overcome the drawback of slow infection speed, achieving the desired pest control effect, especially during the local outbreak and migration of aphids.
Therefore, the SZU-China team designed dsRNAs targeting three genes of brown citrus aphids (the CHS gene, CYP450 gene, and CP19 gene), inserted them into the T-DNA region of the Ti plasmid, and then introduced them into Metarhizium via Agrobacterium-mediated transformation. After transformation, the target genes were integrated into the Metarhizium genome, enabling the stable expression of dsRNAs in Metarhizium anisopliae and ensuring their effective function.
Implementation Objectives
1. Using an Agrobacterium-mediated transformation method, the target fragment expressing the fusion dsRNA was inserted into the genome of Metarhizium anisopliae, enabling its stable expression in Metarhizium.
2. Clarify the infection effect of Metarhizium anisopliae on brown citrus aphids.
Verification process
Construction of plasmid pBar_PtrpC-dsF3
For the validation of the component function, we chose to carry it in the T-DNA of the pBar vector (kindly provided by Professor Duo Lai from Shenyang Agricultural University). This region contains a tryptophan constitutive promoter and a glufosinate resistance gene, which are used to screen for successfully transformed strains.Furthermore, the dsRNA we designed was inserted downstream of the tryptophan constitutive promoter (PtrpC), and transcription termination is mediated by the downstream tryptophan terminator (ttrpC) (see Fig 5-2.). Ultimately, we constructed a plasmid with a length of 10,376 bp, in which the T-DNA fragment will be inserted into the genome of Metarhizium via Agrobacterium-mediated transformation.This dsRNA is transcribed from an expression cassette consisting of a forward DNA sequence, a stem-loop, and a reverse DNA sequence, enabling simultaneous targeting of three genes in brown citrus aphids.
Fig 5-2. Schematic diagram of the plasmid pBar_PtrpC-dsF3-TtrpC.
Agrobacterium-mediated genetic transformation
The plasmid was first transformed into Agrobacterium. Successfully transformed Agrobacterium strains were able to grow normally on medium containing kanamycin, whereas strains transformed with the empty pBar vector could not grow. Single colonies from the resistant medium were picked for PCR verification, yielding amplified sequences of 711 bp and 955 bp in length (Fig 5-3. (a)), confirming that the transformed Agrobacterium carried the target plasmid. Under induction with acetosyringone (AS) and MES, Agrobacterium transferred the Ti plasmid into spores of Metarhizium anisopliae. After glufosinate resistance screening, successfully transformed Metarhizium strains were able to grow normally on SDAY medium containing cefotaxime and glufosinate, whereas Metarhizium transformed with the empty pBar vector could not grow (Fig 5-3. (b)).Twenty milligrams of mycelia from normally growing single-colony Metarhizium were used for genomic DNA extraction for PCR verification and for total RNA extraction for RT-PCR verification. The results showed that genomic amplification using primers F: gcccttcctccctttatttc and R: ctgacatcgacaccaacgat produced two distinct bands on a 1% agarose gel, with lengths of 711 bp and 955 bp (Fig 5-3. (c)). RT-PCR on RNA using the same primers yielded a distinct band of 80 bp on a 3% agarose gel (Fig 5-3. (d)), demonstrating that the T-DNA fragment had been inserted into the Metarhizium genome and was able to stably transcribe dsRNA.
Fig 5-3. Validation results of Metarhizium engineered via Agrobacterium-mediated transformation for dsF3 production. (a) Agarose gel electrophoresis results of PCR verification for Agrobacterium transformed with the pBar_PtrpC-dsF3-TtrpC plasmid. M, DNA molecular weight marker (from top to bottom: 2000 bp, 1000 bp, 750 bp, 500 bp, 300 bp, 100 bp). 1–5, PCR results of transformed Agrobacterium AGL1 colonies. +, PCR result of the positive control plasmid. −, negative control. (b) Growth of transformed Metarhizium after transfer from M-100 medium to SDAY medium containing glufosinate. Mpbar, Metarhizium transformed with the empty vector. Mpbar-dsRNA-1 and Mpbar-dsRNA-2, Metarhizium transformed with the pBar_PtrpC-dsF3-TtrpC plasmid. (c) Agarose gel electrophoresis results of genomic PCR for transformed Metarhizium. M, DNA molecular weight marker (from top to bottom: 2000 bp, 1000 bp, 750 bp, 500 bp, 300 bp, 100 bp). 1, Genomic PCR of Mpbar-dsRNA-1. 2, Negative control for Mpbar-dsRNA-1. 3, Genomic PCR of Mpbar-dsRNA-2. 4, Negative control for Mpbar-dsRNA-2. 5, Genomic PCR of wild-type Metarhizium. 6, Negative control for wild-type Metarhizium. +, PCR result of the positive control plasmid. (d) Agarose gel electrophoresis results of RT-PCR using total RNA extracted from successfully transformed Metarhizium. M, DNA molecular weight marker (from top to bottom: 1000 bp, 700 bp, 500 bp, 400 bp, 300 bp, 200 bp, 100 bp). 1–5, RT-PCR results of Mpbar-dsRNA-1. −, RT-PCR result of wild-type Metarhizium.
Phenotypic Results of Infection by Engineered Metarhizium
To further confirm the virulence of the modified Metarhizium against brown citrus aphids, we scaled up the culture of the verified successfully transformed strains to prepare spore suspensions. The spore suspension was adjusted to a concentration of 1×10⁷ spores/mL. Tender citrus branches were placed in EP tubes filled with distilled water and fixed inside 50 mL centrifuge tubes. Brown citrus aphids were immersed in the spore suspension for 15 seconds, then transferred to the tender branches for cultivation.
It was observed that after 3 days, white mycelia grew on the aphid body surface and the aphids’ movement was restricted. After 6 days, the aphids died, the mycelia decreased, and green spores began to appear on their body surface. After 9 days, the aphid cadavers were fully covered with spores (see Fig 5-4.).
Fig 5-4. Phenotypic results of brown citrus aphids infected by engineered Metarhizium. (a) Physiological state of brown citrus aphids without Metarhizium infection.(b) Physiological state of brown citrus aphids 3 days after infection with spores of the engineered Metarhizium.(c) Physiological state of brown citrus aphids 6 days after infection with spores of the engineered Metarhizium.(d) Physiological state of brown citrus aphids 9 days after infection with spores of the engineered Metarhizium.
Discussion
There are various types of natural fungi capable of infecting brown citrus aphids; however, their infection process takes too long, resulting in low pathogenicity against brown citrus aphids. Currently, the control effect of fungal preparations on brown citrus aphids is mostly unsatisfactory. Therefore, it is of great significance to improve the pathogenicity of fungal strains against brown citrus aphids through molecular approaches.
Metarhizium anisopliae was genetically modified via Agrobacterium-mediated transformation to express dsRNA that silences the CHS, CYP450, and CP19 genes of brown citrus aphids during infection, thereby enhancing the pathogenicity of the transgenic Metarhizium against brown citrus aphids. In this study, we successfully characterized the expression of this dsRNA and the phenotypic phenomena of brown citrus aphids infected by the engineered Metarhizium. However, due to time constraints, we aim to characterize the silencing effect of the transgenic Metarhizium on the three key genes of brown citrus aphids and determine its stability and biological activity in future research.
Based on the design of a responsible project, we expect to establish a model of the engineered Metarhizium biopesticide spreading in the field and infecting aphid populations, which simulates the process of the fungus spreading from treated plants to adjacent plants and causing infections. This result will provide guidance for farmers to determine the optimal spraying time, spraying location, and spraying dosage, enabling efficient, economical, and environmentally friendly aphid control. For detailed modeling, please refer to the "model" section.
Considering the safety issues of genetically modified microorganisms (GMOs) in application, we designed a safety switch system by referring to the designs of the previous NYMU_Taipei team and Hangzhou-SDG team and combining our own needs to ensure the safety of the engineered fungi. For detailed design, please refer to the "design" section.
Overlook
This year, the SZU-igem team has verified through multi-module experiments that the Aphigo Intelligent Citrus Aphid Management System is practical and valuable. The detection module uses engineered Bacillus subtilis to convert aphid honeydew signals into quantifiable MeSA gas, solving the "difficulty in detection" problem; the RNAi therapy module, after four generations of improvements, has increased the aphid mortality rate from 31.4% to 60.2%, and the optimized system is conducive to large-scale production; the delivery system uses MS2 VLPs as carriers to ensure the stable and efficient delivery of RNAi molecules; the fungal contact-killing module enables engineered Metarhizium to play a dual role, providing targeted prevention and control during the aphid outbreak period.
These achievements form a complete governance process of "early warning triggering → RNAi defense activation → fungal enhanced elimination," breaking through the limitations of traditional chemical pesticidesand conventional biological control. This system helps citrus growers achieve precise pest control, reduce costs and risks, and provides an innovative model of "synthetic biology + digital agriculture" for the green management of agricultural pests and diseases, thereby promoting the sustainable transformation of the citrus industry.
In the future, we are committed to transforming this innovative achievement into broader and deeper practical applications. With technology as the brush and green development as the ink, we will paint a new chapter of ecological harmony and industrial prosperity on the canvas of global agriculture, making the Aphigo Intelligent Management System a solid shield for safeguarding global agricultural ecological security and food security.
References
[1]Le, D. T., & Müller, K. M. (2021). In Vitro Assembly of
Virus-Like Particles and Their Applications. Life (Basel, Switzerland), 11(4), 334.
https://doi.org/10.3390/life11040334
Ashley, C. E., Carnes, E. C., Phillips, G. K., Durfee, P. N.,
Buley, M. D., Lino, C. A., Padilla, D. P., Phillips, B., Carter, M. B., Willman, C. L., Brinker, C. J.,
Caldeira, J.doC., Chackerian, B., Wharton, W., & Peabody, D. S. (2011). Cell-specific delivery of diverse
cargos by bacteriophage MS2 virus-like particles. ACS nano, 5(7), 5729–5745.
https://doi.org/10.1021/nn201397z
Zakeri, B., Fierer, J. O., Celik, E., Chittock, E. C.,
Schwarz-Linek, U., Moy, V. T., & Howarth, M. (2012). Peptide tag forming a rapid covalent bond to a
protein, through engineering a bacterial adhesin. Proceedings of the National Academy of Sciences of the
United States of America, 109(12), E690–E697.
https://doi.org/10.1073/pnas.1115485109
Cui, C., Wang, Y., Liu, J., Zhao, J., Sun, P., & Wang, S.
(2019). A fungal pathogen deploys a small silencing RNA that attenuates mosquito immunity and facilitates
infection. Nature communications, 10(1), 4298.
https://doi.org/10.1038/s41467-019-12323-1
de Groot, M., Bundock, P., Hooykaas, P. et al. Agrobacterium tumefaciens-mediated transformation of filamentous fungi. Nat Biotechnol 16, 839–842 (1998).
Chemla, Y., Sweeney, C.J., Wozniak, C.A. et al. Design and regulation of engineered bacteria for environmental release. Nat Microbiol 10, 281–300 (2025).
Ashley, C. E., Carnes, E. C., Phillips, G. K., Durfee, P. N., Buley, M. D., Lino, C. A., ... & Peabody, D. S. (2011). Cell-specific delivery of diverse cargos by bacteriophage MS2 virus-like particles. ACS nano, 5(7), 5729-5745.
Rohovie MJ, Nagasawa M, Swartz JR. Virus-like particles: Next-generation nanoparticles for targeted therapeutic delivery. Bioeng Transl Med. 2017 Jan 19;2(1):43-57.
Williams, L. A., Neophytou, A., Garmann, R. F., Chakrabarti,
D., & Manoharan, V. N. (2024). Effect of coat-protein concentration on the self-assembly of bacteriophage
MS2 capsids around RNA. Nanoscale, 16, 3121–3132. https://doi.org/10.1039/D3NR03292B
Chen, Y., Shi, Y., Wang, Z., An, X., Wei, S., Andronis, C., Vontas, J., Wang, J.-J. and Niu, J. (2025) dsRNAEngineer: a web-based tool of comprehensive dsRNA design for pest control. Trends in Biotechnology.
Vert, J. P., Foveau, N., Lajaunie, C., & Vandenbrouck, Y.
(2006). An accurate and interpretable model for siRNA efficacy prediction. BMC Bioinformatics,
7, 520. https://doi.org/10.1186/1471-2105-7-520
Weber, J. L., & Myers, G. W. (1997). Human whole-genome
shotgun sequencing. Genome Research, 7(5), 401–409.
https://doi.org/10.1101/gr.7.5.401
Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W., Mural,
R. J., Sutton, G. G., ... Zhu, X. (2001). The sequence of the human genome. Science, 291(5507),
1304–1351. https://doi.org/10.1126/science.1058040
Marçais, G., Pellow, D., Bork, D., Orenstein, Y., Shamir, R., &
Kingsford, C. (2017). Improving the performance of minimizers and winnowing schemes. Bioinformatics,
33(14), i110–i117. https://doi.org/10.1093/bioinformatics/btx235
