Generation of a Bivalent Recombinant Vaccine Conferring Dual Protection Against Potyvirus and Orthotospovirus in Cucurbits

TL;DR

Scientists created a dual-purpose vaccine that protects cucurbit crops from two different viruses, which are spread by insects.

Contribution

A bivalent recombinant vaccine was developed to confer dual protection against two distinct plant viruses in cucurbits.

Findings

ZAC-MYnp provided 70.4% and 87.0% protection against ZYMV in zucchini and muskmelon.

ZAC-MYnp reduced MYSV symptom severity in muskmelon with 66.7% protection and 79.0% efficacy.

Abstract

Climate warming has facilitated the expansion of insect vectors and plant viral pathogens, leading to increased incidence of viral diseases in crops. Cucurbit crops, including cucumber (Cucumis sativus), melon (Cucumis melo), squash (Cucurbita pepo), and watermelon (Citrullus lanatus), are of major economic importance worldwide, but their production is severely threatened by viral infections. Among the most damaging viruses are zucchini yellow mosaic virus (ZYMV; genus Potyvirus), transmitted by aphids, and melon yellow spot virus (MYSV; genus Orthotospovirus), transmitted by thrips, both of which cause significant yield losses in Asia, including Taiwan. Previously, an attenuated ZYMV mutant, ZAC, was shown to confer effective cross-protection against ZYMV in several cucurbit species. In the present study, we engineered a recombinant virus, ZAC-MYnp, by inserting the nucleocapsid protein (NP) open reading frame of MYSV into the ZAC genome. ZAC-MYnp retained the attenuated phenotype of ZAC and remained effective in protecting against ZYMV infection, with protection rates of 70.4% and 87.0% in zucchini and muskmelon plants, respectively. In addition, under both mechanical and thrips-mediated challenge conditions, ZAC-MYnp significantly reduced MYSV symptom severity in muskmelon, with a protection rate of 66.7% and a protective efficacy of 79.0%, respectively. These results demonstrate that ZAC-derived recombinant viruses can function as a bivalent viral vaccine, offering dual protection against an aphid-borne potyvirus and a thrips-borne orthotospovirus. Our study highlights the feasibility of using a bivalent recombinant vaccine to manage two distinct insect-borne viruses in cucurbit crops.

Figures

Full text

1. Introduction

Climate warming has accelerated the geographic expansion of insect vectors, thereby intensifying the incidence and impact of plant viral diseases worldwide. These diseases cause substantial yield losses and pose a growing challenge to sustainable crop production, underscoring the urgent need for effective, environmentally compatible virus management strategies.

Chemical control of vectors remains largely ineffective due to rapid insect adaptation and resistance. Although RNA interference (RNAi)-based transgenic plants confer strong virus resistance [1,2], regulatory restrictions and public concerns have limited their deployment. In contrast, mild virus-mediated cross-protection has emerged as a sustainable alternative. Cross-protection is a phenomenon first described by McKinney in 1929 between two strains of tobacco mosaic virus (TMV) [3]. Based on this scenario, prior infection with a mild strain of TMV prevented subsequent infection by a virulent TMV strain [4]. Since then, attenuated viruses have been successfully applied in large-scale viral disease control, such as African cassava mosaic virus (ACMV), Arabis mosaic virus (ArMV), citrus tristeza virus (CTV), cucumber mosaic virus (CMV), papaya ringspot virus (PRSV), Pepino mosaic virus (PepMV), soybean mosaic virus (SMV), tomato mosaic virus (ToMV), and zucchini yellow mosaic virus (ZYMV) [5]. Moreover, some attenuated strains have been commercialized as biopesticides or plant vaccines, including ZYMV in Japan and PepMV in Europe [5,6].

Effective cross-protection requires a suitable mild virus that replicates efficiently in the host, induces attenuated symptoms with no adverse effect, and provides protection against severe infection. Traditionally, attenuated strains were generated through physical or chemical mutagenesis, such as ultraviolet irradiation, heat treatment, or nitrous acid exposure [7]. Advances in reverse genetics and molecular engineering have developed methods enabling modification of viral pathogenicity determinants, facilitating the rational design of mild protective strains [8,9,10,11]. Cross-protection is generally strain- or species-specific, as demonstrated by mild strains of PRSV and ZYMV that confer protection only against closely related isolates [12,13,14]. This specificity limits their effectiveness against unrelated virus species.

Unlike live-attenuated vaccines in animals, plant cross-protection is mainly mediated by RNA silencing, including transcriptional and post-transcriptional gene silencing (TGS and PTGS), which operates in a sequence homology-dependent manner [9,15,16]. Additional mechanisms, such as sequestration of essential host factors to interfere with viral replication and induction of salicylic acid (SA)-mediated defense responses, are also involved [9]. Virus-induced gene silencing (VIGS) has been widely exploited for functional genomics and antiviral resistance [17,18]. Recombinant vaccines conferring protection against multiple plant viruses have previously been developed using apple latent spherical virus (ALSV) and pepper veinal mottle virus (PVMV). ALSV causes a latent infection in most host plants, and ALSV-based vectors have been shown to stably induce VIGS, thereby providing adequate protection against a variety of viruses, including cucumoviruses, orthotospoviruses, and potyviruses [19,20,21]. In contrast, the protective attenuated PVMV strain was generated through targeted mutagenesis of the 6K1 protein and then engineered as an expression vector. The derivative attenuated recombinant virus carrying the open reading frame (ORF) of the potato virus X (PVX) coat protein (CP) can successfully confer protection against the heterologous virus [22]. Mild virus strains, therefore, represent attractive platforms for delivering heterologous viral sequences to induce protection without compromising plant health.

Cucurbit crops, including cucumber (Cucumis sativus L.), melon (Cucumis melo L.), squash (Cucurbita pepo L.), and watermelon (Citrullus lanatus (Thunb.) Matsum. & Nakai), are of major economic importance, with an annual production value of approximately USD 250 million in Taiwan [23]. These crops are frequently affected by mixed infections of aphid-borne ZYMV and PRSV watermelon type (PRSV-W) (genus Potyvirus), thrips-borne melon yellow spot virus (MYSV) and watermelon silver mottle virus (WSMoV) (genus Orthotospovirus), and whitefly-borne cucurbit chlorotic yellows virus (CCYV) (genus Crinivirus), resulting in severe yield losses [24,25].

Previously, we developed an attenuated ZYMV strain, ZAC, by introducing two amino acid substitutions, R_180_I and E_396_N, into the helper component-protease (HC-Pro) of the severe Taiwanese isolate TW-TN3. ZAC induces mild symptoms in cucurbits following recovery, is not aphid-transmissible, and provides durable protection against severe ZYMV strains [8,26]. ZAC has also been engineered as a viral vector and used to elucidate the role of viral suppressors of RNA silencing in symptom development and virus synergism [27,28].

In the present study, we constructed a recombinant attenuated virus, ZAC-MYnp, by inserting the full nucleocapsid protein (NP) ORF of MYSV (designated MYnp) into the ZAC backbone. ZAC-MYnp retains the attenuated phenotype of ZAC while conferring dual protection against both ZYMV and MYSV in cucurbit hosts. Here, we present a novel, promising strategy for concurrent management of two noxious insect-borne viruses threatening cucurbit crops.

2. Materials and Methods

2.1. Virus Source and Inoculation

The MYSV TW isolate, collected from watermelon in Miaoli County, central Taiwan [29], was maintained in Chenopodium quinoa and Nicotiana benthamiana plants. The ZYMV TW-TN3 isolate, sourced from sponge gourd (Luffa cylindrica) in Tainan City, southern Taiwan [30], was maintained in zucchini (Cucurbita pepo L. var. zucchini) and C. quinoa plants. For mechanical inoculation, virus inoculum was prepared by grinding infected leaf tissue in 10 mM potassium phosphate buffer (pH 7.0) containing 0.1% sodium sulfite. All inoculated plants were kept in a temperature-controlled greenhouse at 25–28 °C.

2.2. Generation of ZAC Recombinant Construct Carrying the MYSV NP ORF

Total RNA was extracted from MYSV-infected N. benthamiana leaf tissue using the Plant Total RNA Miniprep Purification kit (GMbiolab, Taichung, Taiwan), according to the manufacturer’s protocol. Reverse transcription-polymerase chain reaction (RT-PCR) was performed to amplify the full-length MYSV NP ORF (MYnp) using the primer pair MYSV-N-f/MYSV-N-r (Table S1) as previously described [29]. SphI and KpnI restriction sites were incorporated at 5′ and 3′ ends of the amplicon, respectively, for cloning purposes. The PCR product was cloned into the pCR2.1-TOPO vector using the TOPO TA Cloning kit (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s standard procedure. The cloned insert was verified by Sanger sequencing. The MYnp fragment was then excised with SphI/KpnI and ligated into the corresponding sites of the attenuated ZYMV infectious clone p35SZYMVGAC [8] to generate the construct p35SZAC-MYnp.

2.3. Infectivity, Host Response, and Stability Analyses

Plasmids of p35SZYMVAC [26] and p35SZAC-MYnp were purified using the Plasmid Miniprep Purification kit (GMbiolab, Taichung, Taiwan), according to the manufacturer’s instructions. A 10 μL aliquot containing 1 μg plasmid DNA was introduced into two-cotyledon-stage zucchini seedlings by particle bombardment using a Biolistic PDS-1000/He particle delivery system (Bio-Rad, Hercules, CA, USA) as described previously [28]. Inoculated plants were maintained in a temperature-controlled greenhouse (25–28 °C) for symptom observation. Once symptoms appeared, extracts from systemically infected tissues were used as inoculum for mechanical inoculation of C. quinoa, muskmelon (Cucumis melo var. cantalupo), and zucchini squash plants. Local lesion development in C. quinoa leaves was examined under UV illumination (365 nm) [31].

The recovered virus ZAC-MYnp, derived from p35SZAC-MYnp, was mechanically transferred to zucchini and C. quinoa plants every 14 days after inoculation to maintain the virus and assess its stability, for a total of 20 passages. In addition, CP accumulation of ZAC-MYnp in zucchini plants (n = 3) was monitored by indirect enzyme-linked immunosorbent assay (ELISA) using the anti-ZYMV CP antiserum RAs-ZCP [30] every two days over a 30-day period.

2.4. Verification of the Recombinant Virus

RT-PCR and restriction fragment length polymorphism (RFLP) analyses were used to confirm the identity of ZAC and ZAC-MYnp. Total RNA was extracted from virus-infected plant tissue using the aforementioned kit and subjected to one-step RT-PCR using the One-Step RT-PCR kit (GMbiolab, Taichung, Taiwan). The primer sequences used are listed in Table S1. The primer pair PZCP8542/MZCP9378 corresponding to the ZYMV CP coding sequence was used to detect ZYMV and its derivatives. The primer pair MYSV-N-f/MYSV-N-r [29] was used to amplify the MYnp insert within the recombinant ZAC genome. The primer pair PZ866/MZ1155 was used to confirm the insert [28]. Amplicons were sequenced for further verification. RFLP was performed using the primer pair PZ1328/MZ2263 for RT-PCR amplification followed by StuI digestion to differentiate ZAC and its derivatives from ZYMV TW-TN3, as described previously [8].

2.5. Serological Detection

Crude sap extracted from infected leaf tissue was diluted 1:20 for serological assays. Indirect ELISA and immunoblotting were performed to detect viral infection and protein expression, following the previously reported procedures [32]. Rabbit antisera against ZYMV CP (RAs-ZCP) and MYSV NP (RAs-MNP) were used at 1:1000 dilution [29,30], followed by the 1:5000-diluted secondary antibody alkaline phosphatase-conjugated goat anti-rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA, USA).

2.6. Thrips Transmission Assay

First-instar Thrips palmi larvae were allowed a 48 h acquisition access period (AAP) on MYSV-infected muskmelon leaves, then transferred into glass tubes containing fresh common bean (Phaseolus vulgaris) leaves until adulthood. Five adult thrips were placed on test plant leaves and secured with a netted clip for a 48 h inoculation access period (IAP). Plants were maintained in a greenhouse for symptom development.

2.7. Detection of MYSV in Thrips

Thrips were collected to verify their MYSV carriage by real-time RT-PCR. Total RNA from individual thrips was extracted using the Total RNA Purification kit (GMbiolab, Taichung, Taiwan) with modifications. Each thrips was homogenized in 350 μL RNA lysis buffer, mixed with an equal volume of 70% ethanol, and processed according to the manufacturer’s protocol. The primer pair MY-N516f/MY-N774r was used for MYSV detection. The primer pair mtD-Tp-F/mtD-Tp-R, designed based on PCR product sequences amplified by the primer pair mtD-7.2F/mtD-9.2R [33], was used to amplify transcripts of the mitochondrial cytochrome c oxidase subunit I (COI) gene of T. palmi as an endogenous control. Primer sequences are listed in Table S1. SYBR Green I-based real-time RT-PCR was conducted using the StepOne™ Plus Real-Time PCR System (Applied Biosystems, Carlsbad, CA, USA). The reaction mixture (10 μL) included 2 μL total RNA, 5 μL KAPA SYBR^®^ FAST qPCR 2× Master mix (KAPA Biosystems Inc., Woburn, MA, USA), 50 U SAMscript reverse transcriptase (GMbiolab, Taichung, Taiwan), 10 U RNase inhibitor (GMbiolab, Taichung, Taiwan), and 100 nM of each primer. The reaction conditions were cDNA synthesis at 42 °C for 30 min, DNA polymerase activation at 95 °C for 5 min, followed by 35 cycles of 95 °C for 15 s, 65 °C for 30 s, and 75 °C for 20 s. Melting curve analysis was performed from 65 °C to 95 °C with a 0.3 °C increment. Reactions were considered positive if both an exponential increase in fluorescence and a specific melting peak were detected.

2.8. Cross-Protection Evaluation

Cross-protection was evaluated in plants of the zucchini cv. Dark Green (Western Hybrid Seeds, Inc., Hamilton, CA, USA) and muskmelon cv. Meihua (Known-You Seed Co., Ltd., Kaohsiung, Taiwan) under greenhouse conditions. Seedlings at the one-true-leaf stage were first inoculated with attenuated viruses using leaf-tissue extracts from infected zucchini plants as the inoculum. Buffer-mock inoculation served as a control. Ten days post-inoculation (dpi), plants that tested positive for attenuated viruses using RAs-ZCP via indirect ELISA were mechanically challenged with severe ZYMV TW-TN3 or MYSV TW. Additionally, challenge inoculation with MYSV TW was also conducted on 14-dpi muskmelon plants by T. palmi transmission.

Symptom development was monitored, and plants showing severe symptoms were further examined by bioassay or RT-PCR to confirm the presence of the challenge viruses. ZYMV infection was tested by C. quinoa bioassays. The crude extract from the tested plants was applied to C. quinoa leaves to distinguish between mild and severe virus infection. If no local lesions were observed, it was recorded as a protective effect against the challenge virus. MYSV infection was tested by one-step RT-PCR using the primer pair gM410/gM870c (Table S1) targeting the NSm ORF encoding the movement protein in M RNA, as previously described [34]. The protection rate was calculated as follows: (plants without conspicuous symptoms/total plants tested) × 100. Protective efficacy was estimated as: [Disease rate (DR) in the untreated population (DRU)—DR in the protected population (DRP)/DRU] × 100 [35].

3. Results

3.1. Infectivity and Host Responses of ZAC-MYnp

The recombinant virus ZAC-MYnp was successfully recovered by biolistic delivery of the construct p35SZAC-MYnp into zucchini cotyledons, the natural host of ZYMV. Similar to plants infected with the parental attenuated strain ZAC, zucchini plants inoculated with the recombinant ZAC-MYnp developed mild, transient mottling on systemic (uninoculated) leaves at 14 dpi, following gradual recovery without symptoms thereafter (Figure 1).

To evaluate host responses in additional plant species, the recovered recombinant ZAC-MYnp was mechanically transferred to the local lesion indicator C. quinoa and to muskmelon plants, a commercially important cucurbit highly susceptible to both ZYMV and MYSV. No visible local lesions were observed on C. quinoa leaves; however, discrete fluorescent spots were detected under UV illumination at 7 dpi, confirming successful infection. In muskmelon plants, ZAC-MYnp induced mild, transient symptoms comparable to those observed in zucchini plants, indicating that the recombinant virus retained the attenuated phenotype characteristic of ZAC. In contrast, plants infected with the wild-type (WT) ZYMV TW-TN3 isolate developed pronounced chlorotic local lesions on C. quinoa leaves at 7 dpi and exhibited severe mosaic symptoms and leaf deformation in both zucchini and muskmelon by 14 dpi (Figure 1).

Viral infection was confirmed by RT-PCR using ZYMV-specific primer pairs PZ866/MZ1155 and PZ1328/MZ2263 (Table S1). The presence and integrity of the MYnp insert, as well as the ZAC genomic backbone, were further confirmed by RT-PCR and RFLP analyses (Figure 2). Sanger sequencing of the amplified products verified the correct sequence of the MYnp insert within the ZAC genome.

3.2. Stability Assessment of ZAC-MYnp

The infectivity and pathogenicity of ZAC-MYnp were assessed through serial passage in zucchini and C. quinoa plants at a 14-day interval for a total of 20 passages. Throughout the passaging experiment, ZAC-MYnp consistently retained its infectivity and attenuated symptom phenotype. To examine viral replication dynamics, CP accumulation of ZAC-MYnp was monitored in zucchini plants (n = 3) at two-day intervals over a 30-day period by indirect ELISA using RAs-ZCP [30]. CP accumulation levels of ZAC-MYnp remained consistently lower than those of WT ZYMV and were comparable to those observed in plants infected with the parental attenuated strain ZAC (Figure 3a). Expression of the heterologous MYnp was evaluated by immunoblot analysis using RAs-MNP [29]. No detectable MYSV NP signal was observed in ZAC-MYnp-infected tissues, indicating that the translational product of MYnp insert was not stable (Figure 3b).

To further assess the long-term genetic stability of the MYnp insert, RT-PCR analyses were performed on ZAC-MYnp-infected zucchini plants at multiple time points. Amplification of the ZAC-MYnp CP ORF served as a control for viral genome integrity. The CP ORF remained intact from 4 dpi onward (Figure 3c), whereas the MYnp insert showed progressive degradation after 20 dpi, as indicated by the appearance of truncated MYnp fragments (Figure 3d,e). These results suggest that while ZAC-MYnp maintains stable infectivity and attenuation, the heterologous MYnp insert displays limited long-term stability during systemic infection.

3.3. Evaluation of the Protective Effect of ZAC-MYnp Against ZYMV

The protection efficacy of ZAC-MYnp against ZYMV was evaluated in greenhouse-grown zucchini and muskmelon plants. Seedling cotyledons were mechanically inoculated with ZAC-MYnp, with ZAC-treated plants serving as the protective control and mock-inoculated plants as the unprotected control. At 10 dpi, all plants were challenge-inoculated with the virulent ZYMV TW-TN3 strain. The evaluation assay was conducted in at least 3 independent trials.

In zucchini plants, all mock-inoculated control plants (n = 18) developed severe yellow mosaic and leaf distortion symptoms by 14 days post-challenge (dpc). In contrast, all ZAC- and ZAC-MYnp-treated plants (n = 27 per treatment) remained symptom-free at this time point (Figure 4a). By 21 dpc, mosaic symptoms were observed in seven ZAC-MYnp-treated plants and one ZAC-treated plant. At the end of the 30-day observation period, 19 of 27 ZAC-MYnp-treated plants (70.4%) and 25 of 27 ZAC-treated plants (92.6%) remained asymptomatic (Figure 4b).

In muskmelon plants, yellow mosaic symptoms were first observed in 11 mock-inoculated control plants (n = 35) at 7 dpc, and all control plants showed severe symptoms and growth retardation by 21 dpc. In contrast, only one ZAC-MYnp-treated plant (n = 23) developed symptoms at 14 dpc, and a total of three test plants exhibited symptoms by the end of the trial. Overall, 20 of 23 ZAC-MYnp-treated plants (87.0%) and all ZAC-treated plants (n = 20) remained asymptomatic throughout the 30-day assessment period (Figure 4c,d).

The protective effect was further validated by bioassay. Crude sap extracts from treated plants were mechanically inoculated onto C. quinoa leaves to assess the presence of WT ZYMV. No local lesions were observed on leaves inoculated with sap from asymptomatic ZAC-MYnp- or ZAC-treated plants, whereas sap from symptomatic plants and unprotected plants induced characteristic ZYMV-associated chlorotic lesions, as shown in Figure 1. Collectively, these results demonstrate that ZAC-MYnp confers significant protection against ZYMV in both zucchini and muskmelon plants.

3.4. Evaluation of the Protective Effect of ZAC-MYnp Against MYSV

Since zucchini plants are not susceptible to MYSV, the protective efficacy of ZAC-MYnp against MYSV was evaluated solely in muskmelon plants. MYSV was mechanically introduced to plants that had been pretreated for 10 days with ZAC-MYnp (n = 18), ZAC (n = 19), or mock inoculation (n = 19) in three independent trials. All plants were kept in a greenhouse and monitored for 60 days. Following the mechanical challenge, severe mosaic symptoms were observed at 14 dpc in 11 ZAC-treated plants (57.9%) and 16 mock-inoculated plants (84.2%). By 21 dpc, all plants in both groups (100%) developed bud necrosis and growth retardation (Figure 5a). In contrast, only five ZAC-MYnp-treated plants (27.8%) exhibited severe symptoms at 28 dpc, increasing to six plants (33.3%) by 35 dpc, corresponding to a protection rate of 66.7% (Figure 5b).

To further assess protection under conditions mimicking natural infection, MYSV challenge was conducted via transmission by T. palmi. Five thrips that had undergone a 48 h AAP were placed on each test plant 14 days after pretreatment with ZAC-MYnp, ZAC, or mock inoculation and allowed a 48 h IAP. Thrips individuals were retrieved after virus transmission to validate MYSV carriage by real-time RT-PCR testing (Figure 6).

The number of plants per treatment varied with thrips availability, resulting in 54, 67, and 68 plants with ZAC-MYnp, ZAC, and mock treatments, respectively. Plants were observed for 60 days. In this case, different disease severity was observed and scored using a disease severity level (DSL) scale, where DSL 1 indicates no symptoms and normal growth; DSL 2 indicates mild leaf mottling; DSL 3 indicates yellow or necrotic spots and mosaic symptoms; and DSL 4 indicates bud necrosis and growth restriction (Figure 7a). Plants classified as DSL 3 or 4 were considered unprotected. Severe symptoms were observed at 14 dpc in 13 ZAC-treated plants (19.4%) and nine mock-inoculated plants (13.2%). By the end of the experiment, severe diseases developed in 60 ZAC-treated plants (89.6%) and 60 mock-inoculated plants (88.2%). In contrast, only seven ZAC-MYnp-treated plants (13.0%) exhibited severe symptoms at 28 dpc, increasing to 10 plants (18.5%) by 35 dpc, resulting in a protective efficacy of 79.0% (Figure 7b).

At the conclusion of the observation period, MYSV infection in symptomatic plants was confirmed by RT-PCR targeting the NSm gene of the M RNA segment, which was used to identify the MYSV genome in ZAC-MYnp-infected plants. Amplification of the expected 488 bp product verified MYSV infection and correlated with the severe disease phenotypes observed in challenge assays (Figure 7c).

4. Discussion

Plant virus diseases transmitted by insect vectors remain a persistent and escalating constraint on cucurbit production, a challenge that is further intensified by climate-driven expansion of vector populations and by the frequent occurrence of mixed viral infections. An effective management strategy must therefore be durable, environmentally compatible, and capable of controlling multiple viruses simultaneously. In this study, we demonstrate that an attenuated recombinant virus, ZAC-MYnp, can function as a bivalent plant viral vaccine, conferring protection against both a potyvirus (ZYMV) and an orthotospovirus (MYSV). This work extends the concept of mild strain-mediated cross-protection by the evidence that a single engineered attenuated virus can provide dual protection against taxonomically distant viruses of different families in economically important cucurbit hosts [21,22].

The attenuated ZAC backbone represents a well-characterized and biologically safe platform for vaccine development. Previous studies have established that ZAC, generated by targeted mutations in HC-Pro, induces only highly attenuated symptoms followed by symptomless recovery, accumulates at lower levels than WT ZYMV, and lacks aphid transmissibility while maintaining strong cross-protective efficacy against severe ZYMV strains [8,16,26]. Importantly, ZAC-MYnp preserved these favorable properties, including stable attenuation and lower CP accumulation, indicating that insertion of the MYnp sequence did not compromise the core protective profile of the parental virus ZAC. Since the key mutation HC-Pro to abolish aphid transmissibility remains the same, it is expected that the mild recombinant should retain the criterion of non-transmissibility by aphids. Maintenance of attenuation is a critical requirement for practical deployment of mild virus-based vaccines, as reversion to virulence represents a major biosafety concern.

Our previous studies demonstrated that ZYMV- and ZAC-based vectors can efficiently express diverse heterologous plant viral genes in planta [27,28,32,36]. In the present study, however, we demonstrate that the heterologous viral protein is not expressed at detectable levels from the ZAC vector; yet, the recombinant ZAC-MYnp still confers significant protection against MYSV. Because the cloning strategy introduced additional aa residues at both ends, the heterologous protein may be expressed with instability, or the inserted protein may be removed from the viral polyprotein through unanticipated protein splicing events. This finding suggests a protective mechanism primarily mediated by VIGS rather than by protein-based interference. VIGS triggered by homologous viral RNA sequences is a well-established antiviral defense in plants, in which even translationally inactive inserts can induce effective resistance through sequence homology-dependent RNA silencing [17,20,21,22]. Our results indicate that the protection is not mediated by MYSV NP expression. Instead, the findings suggest that MYnp-derived RNA from the mild recombinant virus is sufficient to initiate a systemic silencing response targeting the MYSV genome. Future research will elucidate the nature and persistence of MYnp-derived small interfering RNAs and their contribution to antiviral protection.

Although the MYnp insert exhibited progressive instability during long-term systemic infection, this did not compromise the protective effect observed in challenge assays. This outcome is most likely attributable to the early induction of a robust and durable RNAi response, initiated by the relatively long insert sequence. From a practical standpoint, such transient stability may be sufficient for short-cycle crops such as cucurbits, in which effective early-season protection is often adequate to preserve yield and fruit quality. Nevertheless, the observed insert instability underscores an important design consideration for recombinant viral vaccines. Previous studies have demonstrated that a 201-nt fragment derived from orthotospoviral NP ORFs is sufficient to confer effective cross-protection [20]. Our findings suggest the existence of a viral homeostasis that maintains genomic stability during replication, thereby limiting the length of heterologous inserts that can be stably maintained in the viral vector. Accordingly, future optimization of the ZAC platform could focus on refining insert length and sequence composition to improve genetic stability while retaining strong RNAi-mediated antiviral efficacy.

ZAC-MYnp retained strong protective efficacy against ZYMV in both zucchini and muskmelon plants, although protection levels were slightly reduced compared with the parental ZAC strain. This modest decrease is likely attributable to fitness costs associated with carrying an additional heterologous insert, which may affect replication efficiency or RNA stability. Nonetheless, the protection conferred by ZAC-MYnp remained agriculturally meaningful, with high proportions of treated plants remaining asymptomatic throughout the observation period. Notably, ZAC-MYnp delayed symptom onset in some breakthrough infections, a feature particularly valuable for short-duration crops, where delaying disease progression can substantially mitigate economic losses.

Protection of ZAC-MYnp against MYSV was especially pronounced under thrips-mediated transmission, which more closely mimics natural infection conditions than mechanical inoculation. Compared with high-dose mechanical challenge, vector transmission typically delivers lower and more variable inoculum levels, allowing RNA silencing mechanism to become fully established before systemic virus spread. Under these conditions, ZAC-MYnp achieved high protective efficacy and markedly reduced disease severity. Verification of MYSV acquisition in individual thrips further confirmed that the observed protection was not due to failed challenge inoculation but reflected genuine antiviral resistance. To this end, we also developed a sensitive and specific real-time RT-PCR method to detect MYSV in a single thrips (Figure 5).

From an applied perspective, the ability of a single attenuated recombinant virus to protect against both aphid-borne potyviruses and thrips-borne orthotospoviruses is particularly attractive for integrated pest management (IPM) systems. In Taiwan and other cucurbit-producing regions, cultivation under net houses or protected facilities reduces but does not eliminate vector pressure like tiny thrips, and mixed viral infections remain a serious threat. Incorporation of ZAC-MYnp as a prophylactic treatment could complement existing IPM strategies by providing continuous antiviral protection, thereby reducing reliance on chemical control.

We previously demonstrated that the combined application of ZAC and an attenuated PRSV-W strain, WAC, confers simultaneous protection against mixed infections by both viruses in cucurbits [37]. Building on this work, the present study provides proof of concept that a single recombinant attenuated plant virus can function as a bivalent vaccine against unrelated plant viruses belonging to different families, offering a flexible and scalable strategy for managing complex viral disease pressures in the cucurbit production system. Consistent with this platform-based approach, we have also constructed PRSV WAC as an attenuated recombinant virus WAC-CP, which carries another aphid-borne unrelated CMV CP ORF, to protect cucurbits from both PRSV-W and CMV infections [38]. Future studies will focus on field-scale validation of ZAC-MYnp and further development of this strategy toward combinatorial applications, including a 2-in-1 combination of bivalent vaccines (ZAC-MYnp + WAC-CP) designed to protect cucurbit crops against four distinct viruses.