A New Strategy of Cross-Protection Based on Attenuated Vaccines: RNA Viruses Are Used as Vectors to Control DNA Viruses

Abstract

Plant viruses can infect various types of plants, including food and oil crops, and ornamental flowers, threatening agricultural production and food supply. Cross-protection is an efficient strategy against severe viral strains. Due to distinct infection mechanisms, cross-protection cases involving RNA viruses and DNA viruses often rely on the utilization of corresponding attenuated strains for control purposes. In this study, we utilized cucumber mosaic virus (CMV), a member of the RNA virus group, as the foundational framework for developing attenuated vaccines. We developed four vaccines by inserting relevant sequences from tomato yellow leaf curl virus (TYLCV), a DNA virus. All vaccines demonstrated effective prevention against TYLCV infection, with relative control efficacies exceeding 80%. Subsequently, we evaluated the preventive effects of these vaccines on mixed infections of CMV and TYLCV. Our findings demonstrated that CMV (R2-2bPTI-TYC1C4), CMV (R2-2bPTII-TYC1C4), and CMV (R2-2bPTIII-TYRep) displayed significant efficacy in preventing mixed infections. Following pre-inoculation with these vaccines, the disease index of tomato plants decreased from 100 to 56. This work provides theoretical foundations and tangible resources for controlling TYLCV through cross-protection while suggesting a feasible strategy for utilizing weak RNA virus vaccines to control DNA viruses.

1. Introduction

Viral plant diseases result in annual economic losses of up to USD 60 billion worldwide. Cross-protection with a mild strain of a disease-causing virus is considered an effective control measure. Cross-protection refers to the phenomenon in which plants infected with the mild strain of a virus are protected from subsequent infection by a virulent strain of the same virus [1]. Cross-protection mechanisms widely acknowledged in the industry include “RNA-mediated gene silencing” and “superinfection exclusion (SIE)”. RNA-mediated gene silencing involves the cleavage of double-stranded RNA generated during viral replication by dicer-like enzymes. This results in small interfering RNAs that selectively target and degrade the corresponding RNA sequences, thus causing gene silencing. SIE refers to the spatial segregation observed between different strains of the same virus within dually infected plant tissues [1].

Notably, cross-protection has been used as a control measure against viral plant diseases since the mid-20th century. In Brazil, cross-protection was achieved against the extensively disseminated citrus tristeza virus (CTV). Eighty million sweet orange trees were effectively protected by employing mild CTV strains, thereby mitigating the impact of the disease [2]. Similarly, the cross-protection of zucchini plants was achieved under field conditions in France by using a mild zucchini yellow mosaic virus WK (ZYMV-WK) strain. This approach resulted in significantly higher and more marketable fruit yields than those observed in control plants subsequently infected with severe ZYMV strains [3]. Numerous studies have reported the successful use of mild RNA strains as vaccines against RNA viruses, including tomato mosaic virus (TMV) and cauliflower mosaic virus (CaMV) [4,5]. An illustrative case involves African cassava mosaic virus (ACMV) DNA-A, which induced resistance against both ACMV and East African cassava mosaic Cameroon virus (EACMCV) in cassava plants [6]. Though remarkable outcomes have been achieved using mild strains of RNA viruses, the use of mild strains of DNA viruses for cross-protection remains relatively limited.

The use of mild strains of RNA viruses to control DNA viruses has not been reported yet. Hence, this study aims to explore the potential of utilizing mild RNA virus strains as vaccines against DNA viruses, specifically focusing on cucumber mosaic virus (CMV) and tomato yellow leaf curl virus (TYLCV) as the RNA and DNA viruses, respectively. CMV has a wide host range and can infect over 1200 plant species from 100 families, including monocotyledonous and dicotyledonous plants [7]. TYLCV exhibits a diverse host range and has been reported in 49 plant species across 16 families. Infection in tomatoes during the seedling stage leads to stunted plant growth, leaf curling, yellowing, and severe yield loss [8]. Previous studies have demonstrated that the overlapping region within the TYLCV coding frame generates substantial siRNAs. The transgenic expression of N-terminal amino acids of the TYLCV-Rep protein strongly inhibits TYLCV replication, while the transcription of R-Rep produces anti-sense RNA sequences that can specifically target and suppress gene expression [9,10].

In this study, a feasible strategy for utilizing RNA virus vaccines to control DNA viruses through cross-protection was explored. Four mild mutant strains of cucumber mosaic virus (CMV) RNA2 containing viral fragments related to tomato yellow leaf curl virus (TYLCV) were constructed and used in combination to prevent infections of TYLCV and CMV in tomato plants. This strategy offers valuable insights into the use of attenuated RNA virus vaccines for the control of DNA viruses or simultaneous management of RNA and DNA viruses.

2. Materials and Methods

2.1. Materials

The infectious clone plasmids containing CMV_Fny RNA1(D00356), RNA2(D00355), and RNA3(D10538) were donated by Professor Tao of Nanjing Agricultural University. The Agrobacterium tumefaciens carrying the infectious clone of TYLCV (MN432609) was generously provided by researcher Li Fangfang from the Plant Protection Institute of the Chinese Academy of Agricultural Sciences. The attenuated infectious clone mutant plasmids pCCFR2-2bPTI (constructed by inserting multiple cloning sites (BamHI, SpeI, and SmaI), as well as the TAATA sequence, immediately after the stop codon of the 2a protein at position 2662 G in the full-length RNA2 sequence of CMV_Fny), pCCFR2-2bPTII (constructed by inserting a fragment containing the TAATAG sequence and multiple cloning sites (BamHI, SpeI, and SmaI) at position 2662 G), and pCCFR2-2bPTIII (constructed by deleting a 90 nt fragment from position 2662 G to 2751 A and inserting the TAATAG sequence and BamHI, SpeI, and SmaI cloning sites) were preserved at the Plant Virus Laboratory, College of Plant Protection, Shandong Agricultural University [11]. The mutants of pCCFR2-2bPTII and pCCFR2-2bPTIII used as basic vectors have been shown to have weak pathogenicity. However, the mutant pCCFR2-2bPTI used as a basic vector has been shown to have strong pathogenicity (Figure S1). Tomato plants of the commercially available “Yellow Saint Maiden” variety were selected.

2.2. Plasmid Construction

To evaluate the cross-protective efficacy of recombinant mutants, which were inserted into the identical TYLCV fragment but utilized different CMV vectors, we established CMV-R2-2bPTI-TYC1C4 and CMV-R2-2bPTII-TYC1C4. Furthermore, to assess the potential cross-protection of various TYLCV fragments within a consistent CMV vector context, targeting the potent toxicity of TYLCV, we developed CMV-R2-2bPTIII-TYRep and CMV-R2-2bPTIII-TYR-Rep for comparative analysis. Using Agrobacterium containing the infectious cloning plasmid of TYLCV as a template, fragments of overlapping regions of different coding frames of TYLCV, including TYLCV-C1C4, TYLCV-Rep, and TYLCV-R-Rep (Supplementary Materials), were cloned. A polymerase chain reaction (PCR) was performed using TYBJ-C1C4-F/TYBJ-C1C4-R, TYBJRep-F/TYBJRep-R, and TYBJR-Rep-F/TYBJR-Rep-R primers (

Table 1. Primer sequences in this study.

Primer Name	Primer Sequence (5′-3′)	Product Size	Purpose
TYBJ-C1C4-F	ATGGATCCTGACATCTGTTGAGCTCTTAGCT	150 bp	Amplification of TYLCV-C1C4
TYBJ-C1C4-R	ATCCCGGGACGAGAATGGGGAACCACATCT
TYBJRep-F	ATGGATCCTTACGCCTTATTGGTTTCTTCTTGG	150 bp	Amplification of TYLCV-Rep
TYBJRep-R	ATCCCGGGGATGACGTAGACCCGCATTATTTAA
TYBJR-Rep-F	ATCCCGGGTTACGCCTTATTGGTTTCTTCTTGG	150 bp	Amplification of TYLCV-R-Rep
TYBJR-Rep-R	ATGGATCCGATGACGTAGACCCGCATTATTTAA
CMV-Fny2-2510-F	AGAATCGACGGGAACGAGGT	540 + 150 bp or 540 + 210 bp	Detection of mutant stability
CMV-Fny2-3050-R	TGGTCTCCTTTTGGAGGCC

Remarks: The protective bases are displayed in a bold font. The enzyme digestion sites are underlined.

).

After enzymatic digestion, fragments were individually ligated into the pCCFR2-2bPTI, pCCFR2-2bPTII, and pCCFR2-2bPTIII vectors, resulting in the plasmids pCCFR2-2bPTI-TYC1C4, pCCFR2-2bPTII-TYC1C4, pCCFR2-2bPTIII-TYRep, and pCCFR2-2bPTIII-TYR-Rep.

2.3. Preparation and Inoculation of Agrobacterium Solution

Plasmids containing wild-type CMV_Fny RNA1, wild-type CMV_Fny RNA2, or mutant CMV_Fny RNA2 and wild-type CMV_Fny RNA3 were used to transform competent cells of Agrobacterium GV3101 by the freeze–thaw method [12]. Positive clones were screened to obtain successfully transferred Agrobacterium, and resuspended in an agro-infiltration buffer (10 mmol/L MgCl₂, 10 mmol/L MES, 150 μmol/L AS) to adjust the OD₆₀₀ value to 1.2. The suspension containing equal amounts of wild-type CMV_Fny RNA1, CMV_Fny RNA2, and CMV_Fny RNA3 was used as a positive control, while the mutant vaccine group comprised a suspension containing equal amounts of wild-type CMV_Fny RNA1, CMV_Fny RNA2 with inserted heterologous fragments, and wild-type CMV_Fny RNA3. The solution containing Agrobacterium without the vector was used as the healthy control. A disposable needleless syringe was used for inoculating tomatoes at the four-leaf stage. Plants were grown at 25 °C under 16 h light/8 h dark conditions.

2.4. Analysis of Pathogenicity and Stability of Mutants

On the third day after inoculation with the mutants, RNA was extracted from the systemic leaves of tomato plants. cDNA was synthesized with Reverse Transcriptase M-MLV (Takara) at 42 °C for 0.5–1 h using 1 μg RNA. A PCR procedure was performed as follows: 3 min at 95 °C; 15 s at 95 °C, 15 s at 52 °C, 60 s at 72 °C for 30 cycles; 5 min at 72 °C. RT-PCR analysis was performed using the primers CMV-Fny2-2510-F and CMV-Fny2-3050-R (Table 1) for amplifying 690 bp and 750 bp bands, respectively. Our findings proved the existence of stably inserted exogenous TYLCV fragments in the four mutants. On the 31st day after inoculation, symptoms were observed in tomato plants with diseases to assess the grade of symptom severity. The disease severity was graded according to plant height based on the methods of Wang et al. [13]. The disease index was calculated to display pathogenicity.

Level 0: The entire plant is disease-free.

Level 1: The disease-affected plant does not show obvious dwarfing and has a standard plant height.

Level 3: Dwarfing of disease-affected plants to over three-quarters of the standard plant height.

Level 5: Dwarfing of disease-affected plants to two-thirds to three-quarters of the standard plant height.

Level 7: Dwarfing of disease-affected plants to half to two-thirds of the standard plant height.

Level 9: Dwarfing of disease-affected plants to more than half the standard plant height.

Disease index: 100 × ∑ (Number of diseased plants at each level) × (Disease level)/(Total number of plants × Highest disease level).

Relative control efficacy (%): 100% × (Disease index of control group − Disease index of treatment group)/Disease index of control group.

3. Results and Discussion

3.1. Construction and Virulence Analysis of Attenuated CMV Mutants Containing TYLCV Specific Fragments

Mutant variants of CMV RNA2, namely R2-2bPTI-TYC1C4, R2-2bPTII-TYC1C4, R2-2bPTIII-TYRep, and R2-2bPTIII-TYR-Rep, containing the C1C4 overlapping regions and the Rep coding sequence, were constructed (Figure 1A). Agrobacterium-mediated transformation was performed, the various RNA2 mutants mixed with wild-type RNA1 and RNA3 were inoculated into “Yellow Saint Maiden” cultivars of tomato plants. After 31 days post-inoculation, the height of the tomato plants was comparable to that of the healthy control, with flat-leaf morphology and an absence of viral disease symptoms. In contrast, tomato plants inoculated with CMV_Fny exhibited severe leaf wrinkling, curling, and conspicuous stunting, indicating that all four mutants displayed attenuated virulence (Figure 1B). RT-PCR analysis confirmed the presence of stably inserted fragments in each mutant variant (Figure 1C).

The 2b protein of CMV is a suppressor of viral gene silencing that is closely related to viral pathogenicity in plants [14,15]. In this study, the 2b protein was mutated to construct protein translation premature termination mutants R2-2bPTI-TYC1C4, R2-2bPTII-TYC1C4, R2-2bPTIII-TYRep, and R2-2bPTIII-TYR-Rep were constructed under the premise of maintaining the integrity of the 2b gene nucleic acid sequence in CMV RNA2. This was the reason why the mutant had both weak virulence and stability.

3.2. Analysis of the Cross-Protection Effect of Tomato Plants Pre-Inoculated with Attenuated Mutants against TYLCV Infection

After pre-inoculation with mild mutants for five days, the tomato plants were challenged with TYLCV (OD₆₀₀ = 0.006). Plant phenotypes, disease indexes, and relative control efficacies were assessed after 34 days.

The effects of varied basic vectors containing a 150 bp fragment of the target virus were initially analyzed. Tomato plants not pre-inoculated with mild mutants exhibited significant stunting, yellowing, and leaf wrinkling after inoculation with TYLCV. Disease indexes of 11 and 19 were observed along with relative control efficacies of 89% and 81% for tomato plants that were pre-inoculated with R1/R2-2bPTI-TYC1C4/R3 and R1/R2-2bPTII-TYC1C4/R3, respectively, and then inoculated with TYLCV (

Table 2. The relative control efficacy of attenuated vaccines against TYLCV.

Vaccine Name	Disease Index	Relative Control Efficacy
TYLCV	100	0%
R1/R2-2bPTI-TYC1C4/R3	11	89%
R1/R2-2bPTII-TYC1C4/R3	19	81%
R1/R2-2bPTIII-TYRep/R3	26	74%
R1/R2-2bPTIII-TYR-Rep/R3	33	67%
R1/R2-2bPTIII-TYRep + R-Rep/R3	11	89%

). Both mutants demonstrated commendable protective effects (Figure 2A). RT-PCR analysis confirmed the presence of stably inserted fragments in each mutant variant and these results were verified by sequencing (Figure 2B, Figures S6 and S7). Notably, R1/R2-2bPTI-TYC1C4/R3 exhibited superior cross-protection in tomatoes compared to R1/R2-2bPTII-TYC1C4/R3. This was possibly attributable to the translated amino acids derived from the inserted overlapping region (C1C4) eliciting a discernible “SIE” response. The occurrence of “SIE” may be attributed to the partial TYLCV sequence fragment (TYC1C4) located before the double-stop codon TAATAG in the mutant CMV-R2-2bPTI-TYC1C4 constructed with CMV-R2-2bPTI as the skeleton, where exogenous fragments could express amino acids. Conversely, in the mutant CMV-R2-2bPTII-TYC1C4 constructed with CMV-R2-2bPTII as the skeleton, the partial TYLCV sequence fragment (TYC1C4) was located after TAATAG and could not be translated into amino acids normally, resulting in a failure to induce “SIE”.

Furthermore, the effects of using the same foundational vector carrying a 300 bp fragment of the target virus were analyzed. Disease severity indexes of 26 and 33 and relative control efficacies of 74% and 67% were observed with the pre-inoculation of tomato plants with R1/R2-2bPTIII-TYRep/R3 and R1/R2-2bPTIII-TYR-Rep/R3, respectively, followed by inoculation with TYLCV (Table 2). Remarkably, after adjusting the concentration of CMV-R2-2bPTIII-TYRep and CMV-R2-2bPTIII-TYR-Rep vaccines to be consistent, tomatoes were inoculated with the mixture at a ratio of 1:1, resulting in a significantly reduced disease index of 11, and a remarkable relative control efficacy of 89% (Table 2, Figure 2C). The presence of stably inserted fragments in each mutant variant was confirmed via RT-PCR analysis and these results were verified by sequencing (Figure 2D, Figures S8 and S9). The stability of two mutants in tomato plants inoculated with CMV-R2-2bPTIII-TYRep and CMV-R2-2bPTIII-TYR-Rep was specifically detected using two pairs of primers, namely TYBJRep-F and CMV-Fny2-3050-R, as well as TYBJR-Rep-F and CMV-Fny2-3050-R. Amplification of a 600 bp fragment confirmed the expected results, indicating the stable presence of both mutant loci in the mixed inoculation. In summary, TYLCV vaccines carrying both forward- and reverse-inserted segments demonstrated comparable levels of protection, when individually administered to tomato plants. However, co-inoculation with both variants resulted in more substantial and synergistic cross-protective effects.

3.3. Analysis of the Cross-Protection Effect of Tomato Plants Pre-Inoculated with Attenuated Mutants against Co-Infection of TYLCV and CMV

Several attenuated CMV mutants have demonstrated the ability to confer cross-protection against TYLCV. Further investigations were undertaken to determine whether these attenuated mutants could effectively combat mixed infections of CMV and TYLCV simultaneously. Tomato plants were pre-inoculated with the attenuated mutants and subsequently co-inoculated with TYLCV (OD₆₀₀ = 0.006) and CMV (OD₆₀₀ = 1.2) after a 5-day interval. Plant phenotypes were carefully monitored for 53 days, and disease indexes and relative control efficacies were calculated. The results revealed substantial stunting and severe leaf curling and wrinkling in tomato plants following co-inoculation with TYLCV and CMV. Notably, in tomato plants pre-inoculated with R1/R2-2bPTI-TYC1C4/R3, R1/R2-2bPTII-TYC1C4/R3, and R1/R2-2bPTIII-TYRep/R3, the disease indexes and relative control efficacies were determined to be 56, 56, and 63, and 44%, 44%, and 37%, respectively (

Table 3. The relative control efficacy of attenuated vaccines against TYLCV and CMV.

Vaccine Name	Disease Index	Relative Control Efficacy
TYLCV + CMV	100	0%
R1/R2-2bPTI-TYC1C4/R3	56	44%
R1/R2-2bPTII-TYC1C4/R3	56	44%
R1/R2-2bPTIII-TYRep/R3	63	37%

). These plants were subsequently challenged with both CMV and TYLCV (Figure 3A). RT-PCR analysis confirmed the presence of stably inserted fragments in each mutant variant (Figure 3B).

Cross-protection is an environmentally safe method to control viral plant diseases [16] and has been used to control multiple plant viruses in the laboratory or field [4,17,18,19]. However, isolating and screening mild strains in nature for cross-protection is a time-consuming task. In addition, plants are usually infected not only by one virus, but by multiple viruses [16]. Therefore, the construction of stable multivalent attenuated vaccines is an effective measure to prevent and control complex infectious viral diseases.

4. Conclusions

In this study, we explored the potential of utilizing mild RNA virus strains as vaccines against DNA viruses. Four mild mutant strains of cucumber mosaic virus (CMV) RNA2 containing viral fragments related to tomato yellow leaf curl virus (TYLCV) were constructed. The aforementioned attenuated CMV vaccines demonstrated efficacy in both the independent TYLCV control and control co-inoculated with TYLCV and CMV. Thus, it is feasible to employ attenuated RNA virus vaccines as a strategic approach to combat DNA virus targets. This strategy offers valuable insights into the use of attenuated RNA virus vaccines for the control of DNA viruses or the simultaneous management of RNA and DNA viruses.