A recessive gene pepy-1 encoding Pelota confers resistance to begomovirus isolates of PepYLCIV and PepYLCAV in Capsicum annuum

A begomovirus resistance gene pepy-1, which encodes the messenger RNA surveillance factor Pelota, was identified in pepper (C. annuum) through map-based cloning and functional characterization. Pepper yellow leaf curl disease caused by begomoviruses seriously affects pepper (Capsicum spp.) production in a number of regions around the world. Ty genes of tomato, which confer resistance to the tomato yellow leaf curl virus, are the only begomovirus resistance genes cloned to date. In this study, we focused on the identification of begomovirus resistance genes in Capsicum annuum. BaPep-5 was identified as a novel source of resistance against pepper yellow leaf curl Indonesia virus (PepYLCIV) and pepper yellow leaf curl Aceh virus (PepYLCAV). A single recessive locus, which we named as pepper yellow leaf curl disease virus resistance 1 (pepy-1), responsible for PepYLCAV resistance in BaPep-5 was identified on chromosome 5 in an F2 population derived from a cross between BaPep-5 and the begomovirus susceptible accession BaPep-4. In the target region spanning 34 kb, a single candidate gene, the messenger RNA surveillance factor Pelota, was identified. Whole-genome resequencing of BaPep-4 and BaPep-5 and comparison of their genomic DNA sequences revealed a single nucleotide polymorphism (A to G) located at the splice site of the 9th intron of CaPelota in BaPep-5, which caused the insertion of the 9th intron into the transcript, resulting in the addition of 28 amino acids to CaPelota protein without causing a frameshift. Virus-induced gene silencing of CaPelota in the begomovirus susceptible pepper No.218 resulted in the gain of resistance against PepYLCIV, a phenotype consistent with BaPep-5. The DNA marker developed in this study will greatly facilitate marker-assisted breeding of begomovirus resistance in peppers.


Introduction
Five domesticated species of Capsicum have been identified to date, among which Capsicum annuum is economically the most important around the world (Pickersgill 1997;Bosland and Votava 2000). In 2019, 4.3 million tonnes of dried pepper fruits and 38 million tonnes of fresh pepper fruits were produced in the world (FAOSTAT). The largest producers of fresh peppers include China, Mexico, Turkey, Indonesia, and Spain, and together, they produce approximately 75% of the world's fresh market pepper. In Indonesia, 2.6 million tonnes of fresh market pepper fruits were produced. Chili peppers are an important source of income for vegetable farmers in Indonesia, occupying an area of at least 155,000 ha and involving more than 500,000 farmers in the country (Vos and Duriat 1995;Mustafa et al. 2006).
Pepper yellow leaf curl disease (PepYLCD) seriously affects pepper production in Indonesia (Kenyon et al. 2014b;Koeda et al. 2016). PepYLCD was first identified in West Java in 1999 (Sulandari et al. 2001(Sulandari et al. , 2006Sumardiyono et al. 2003), and an outbreak of PepYLCD was detected based on the annual crop census data from 2000 to 2006 (De Barro et al. 2008 1 3 carry a circular single-stranded DNA genome encapsulated in twinned icosahedral particles and rely on whitefly (Bemisia tabaci) as vectors for plant-to-plant transmission (ICTV 2019). Most begomoviruses exhibit bipartite genomes composed of two circular DNA components (A and B) of approximately 2800 nt each (Hanley-Bowdoin et al. 2013). By contrast, monopartite begomoviruses carry only a DNA A-like genome component.
Over the past three decades, diseases caused by begomoviruses have caused substantial yield losses in solanaceous crops, particularly tomato (Solanum lycopersicum), pepper, and eggplant (Solanum melongena), in many tropical and subtropical regions of the world (Kenyon et al. 2014b). With the increasing international importance of yellow leaf curl disease in tomato, immense efforts have been made to identify begomovirus resistance sources and to incorporate this resistance into improved tomato cultivars (Kenyon et al. 2014b). To date, six major loci (Ty-1-Ty-6) conferring resistance to the monopartite tomato yellow leaf curl virus (TYLCV) have been identified in wild tomato accessions (Zamir et al. 1994;Agrama and Scott 2006;Anbinder et al. 2009;Ji et al. 2009a, b;Hutton and Scott 2014), and three of these gene (Ty-1/Ty-3/Ty-3a, Ty-2, and ty-5) have been cloned. The Ty-1, Ty-3, and Ty-3a loci are allelic and originated from Solanum chilense accessions LA1969, LA2779, and LA1932, respectively (Zamir et al. 1994;Agrama and Scott 2006). These allelic loci encode an RNA-dependent RNA polymerase (RDR) (Verlaan et al. 2013). The Ty-2 locus cloned from Solanum habrochaites accession B6013, encodes a nucleotide-binding leucine-rich repeat (NB-LRR) protein (Yamaguchi et al. 2018). The ty-5 gene is the only reported recessive resistance gene effective against TYLCV, which encodes the messenger RNA surveillance factor Pelota (Lapidot et al. 2015). The above-mentioned genes are the only begomovirus resistance genes identified to date, although cloning of additional resistance genes in other crops and vegetables is urgently needed.
BaPep-5, a C. annuum accession grown locally in the Aceh province of Indonesia, was identified as a begomovirus resistant genetic resource in our preliminary analyses. Several pepper-infecting begomovirus species commonly occur in Aceh province, including bipartite begomoviruses, such as pepper yellow leaf curl Indonesia virus (PepYL-CIV), pepper yellow leaf curl Aceh virus (PepYLCAV), and tomato yellow leaf curl Kanchanaburi virus (TYLCKaV), and a monopartite begomovirus, ageratum yellow vein virus (Koeda et al. 2016;Kesumawati et al. 2019Kesumawati et al. , 2020. In the present study, we used a map-based cloning approach to identify the begomovirus resistance gene in an F 2 population derived from a cross between BaPep-5 and the begomovirus susceptible C. annuum accession BaPep-4. The work presented here demonstrates that the recessive resistance of BaPep-5 against PepYLCIV and PepYLCAV results from the putative loss-of-function mutation in the messenger RNA surveillance factor Pelota.

Plant material
Two Capsicum annuum accessions, BaPep-5 (locally called Perintis) and BaPep-4 (locally called Kencana), were used in this study. F 1 and F 2 populations obtained by crossing BaPep-5 with BaPep-4 were used for restriction site-associated DNA sequencing (RAD-seq) and fine mapping of the candidate gene. Plants were grown in a growth room with temperatures ranging from 23 to 30 °C under 13 h light/11 h dark photoperiod.

Begomovirus isolates
Isolates of two bipartite begomoviruses, originally isolated in Indonesia, were used in this study: PepYLCIV isolate BA_D1-1 (Accession number of GenBank for DNA A: LC051114, DNA B: LC314794) (Koeda et al. 2016(Koeda et al. , 2018 and PepYLCAV isolate BAPep-V2 (DNA A: LC387327, DNA B: LC387329) . PepYL-CAV, originally isolated from pepper, tomato, and tobacco (Nicotiana tabacum) plants, is a recombinant begomovirus derived from PepYLCIV, the putative major parent, and pumpkin yellow mosaic virus and tomato leaf curl New Delhi virus as the minor parents ).

Infectious clones
The infectious clone of PepYLCIV isolate BA_D1-1 was previously constructed by our group through the transformation of Agrobacterium tumefaciens strain GV2260 with pGreenII-p35S-PepYLCIV-DNA-A+B (Koeda et al. 2018). To construct an infective PepYLCAV clone, fulllength PepYLCAV DNA A and DNA B sequences of isolate BAPep-V2 were cloned into pGreenII-p35S, as described previously (Koeda et al. 2017). To clone PepYL-CAV DNA A, two fragments were amplified by PCR: fragment 1 (full-length copy of viral DNA A) with PepYLCAV fra1 F/R primer pair, and fragment 2 (partial-length copy of viral DNA A) with PepYLCAV fra2 F/R primer pair. The linearized pGreenII-p35S plasmid (Koeda et al. 2017) was ligated with fragments 1 and 2 using the In-Fusion HD Cloning Kit (Takara Bio, Kusatsu, Shiga, Japan), and the resultant construct (pGreenII-p35S-PepYLCAV-DNA-A) was introduced into Escherichia coli (Stellar Competent Cells) (Takara Bio). To clone PepYLCAV DNA B, two fragments were amplified: fragment 1 (partial-length copy of viral DNA B) with PepYLCAV DNA-B fra1 F/R primer pair, and fragment 2 (full-length copy of viral DNA B) with PepYLCAV DNA-B fra2 F/R primers. Ligation of linearized pGreenII-p35S and fragments 1 and 2, followed by E. coli transformation of the resultant construct (pGreenII-p35S-PepYLCAV-DNA-B) were conducted as described above. To clone both DNA A and DNA B of PepYLCAV into the same pGreenII-p35S vector, the DNA B partial repeat in pGreenII-p35S-PepYLCAV-DNA-B was amplified by PCR. The amplified fragment and SmaΙ-digested pGreenII-p35S-PepYLCAV-DNA-A were ligated using the In-Fusion HD Cloning Kit (Takara Bio). The resultant construct, pGreenII-p35S-PepYLCAV-DNA-A+B, was transformed into E. coli. Primers used for plasmid construction are presented in Supplementary  Table S1. A. tumefaciens (GV2260) was transformed with pGreenII-p35S-PepYLCAV-DNA-A+B and pSoup (Hellens et al. 2000). A frozen stock of the transformed Agrobacterium was stored at − 80 °C until needed for subsequent experiments.

Inoculation of pepper plants with begomoviruses
A single inoculation experiment was conducted for graft transmission of PepYLCIV into BaPep-5, BaPep-4, and F 1 plants. The begomovirus susceptible C. annuum accession No.218 was agroinfiltrated with pGreenII-p35S-PepYLCIV-DNA-A+B, as described previously (Koeda et al. 2018). The plant developmental stage is critically important for the successful infection of begomovirus by agroinfiltration. Agroinfiltration was conducted with the optical density of 0.1 to the abaxial surface of the cotyledons which has just come out from the seed coat. The agroinfiltrated No.218 plants were used as scions, and uninoculated BaPep-5, BaPep-4, and F 1 plants were used as rootstocks. Scion plants were grafted onto rootstocks 30 days after agroinfiltration, and the grafted plants were covered by polybags to maintain high humidity. Plants were acclimatized 14 days after grafting, and most of the scion was decapitated to stimulate lateral branch growth of the rootstock. The newly developed lateral branches of BaPep-5, BaPep-4, and F 1 plants were evaluated for disease symptoms at 86 days after grafting, and young upper leaves were collected and stored at − 80 °C until needed for DNA extraction.
Single inoculation experiments with PepYLCIV or Pep-YLCAV were conducted by agroinfiltration. The abaxial surface of the cotyledons of BaPep-5, BaPep-4, and F 1 plants were agroinfiltrated with pGreenII-p35S-PepYLCIV-DNA-A+B or pGreenII-p35S-PepYLCAV-DNA-A+B, as described previously (Koeda et al. 2017(Koeda et al. , 2018. Disease symptoms were surveyed at approximately 60 days postinoculation (dpi), and young upper leaves were collected and stored at − 80 °C until needed for DNA and/or RNA extraction.
The cotyledons of F 2 plants (n = 536) were agroinfiltrated with pGreenII-p35S-PepYLCAV-DNA-A+B, and symptoms surveys were conducted at 54 and 88 dpi. Young upper leaves were collected at 21 and 54 dpi, and stored at − 80 °C until needed for DNA extraction.
The symptoms of each plant were scored on a disease severity index (DSI) ranging from 0 to 4, as follows: 0, no symptoms; 1, very mild symptoms with slight yellowing of the leaf vein; 2, mild yellowing of the leaf with or without the distortion of young upper leaves; 3, moderate yellowing of the leaf with or without the distortion of leaves; 4, heavy yellowing of the leaf with or without distortion. Statistical analysis of average DSI was performed using the Bonferroni-Dunn test of Excel Toukei ver. 7.0 at a p value of 0.05.

PepYLCIV and PepYLCAV viral DNA detection
DNA was extracted from pepper leaves using the Nucleon PhytoPure Kit (GE Healthcare, Little Chalfont, Buckinghamshire, UK). DNA from the leaves of F 2 individuals (n = 536) collected at 21 dpi was extracted using a simple method, as described previously . The DNA A component of PepYLCIV was detected using Pep uni F and PI uni R primers, while PepYLCAV was detected using Pep uni F and PA uni R primers, which amplified 666-and 466-bp fragments, respectively . PCR was performed using Emeral-dAmp PCR Master Mix (Takara Bio). The amplified PCR products were subjected to electrophoresis using 1.0% (w/v) agarose gel. Primer sequences used for viral DNA detection and PCR condition are listed in Supplementary Table S1 and Table S2.

Quantification of PepYLCIV and PepYLCAV titer by real-time PCR
The PepYLCIV and PepYLCAV DNAs were quantified using the CFX Connect Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA), as described previously , with a minor modification. The DNA A components of PepYLCIV and PepYLCAV were detected by quantitative PCR (qPCR) using IV-AV Real F and R primers, which amplified 102-bp fragments. Data from the total DNA extracts were normalized relative to the 25S ribosomal RNA gene (93 bp) amplified by qPCR using 25S-rRNA 2F and 2R primers. The qPCR was performed using SsoAdvanced Universal SYBR Green Supermix (Bio-Rad). Statistical analysis was performed using the Tukey-Kramer test of Excel Toukei ver.7.0 with a p value of 0.05. Primer sequences used for real-time PCR and PCR condition are listed in Supplementary Table S1 and Table S2.

Whole-genome resequencing
DNA was extracted from pepper leaves using the Nucleon PhytoPure Kit (GE Healthcare). Whole-genome resequencing of BaPep-5 and BaPep-4 was performed by the Macrogen Japan sequencing service (Macrogen Japan, Tokyo, Japan). Each sequenced sample was prepared according to the Illumina TruSeq DNA sample preparation guide to obtain a final library with an average insert size of 300-400 bp. After quality control analysis of the sample library and quantification of the DNA library templates, sequencing was conducted using NovaSeq 6000 (Illumina). Trimming and mapping of the sequence reads, followed by variant calling were performed as described above for RADseq. Genetic variant annotation and functional effect prediction were performed using SnpEff (v.4.3.1t) (Cingolani et al. 2012).

Fine mapping
To reduce the target region of the locus, one indel and five SNP markers were developed. To analyze indels, PCR was performed using the EmeraldAmp PCR Master Mix (Takara Bio). The amplified PCR products were subjected to electrophoresis using 8% polyacrylamide gels. To analyze SNPs, high resolution melting (HRM) analysis was conducted. Each PCR reaction contained 5 μL of Precision Melt Supermix (Bio-Rad), 0.2 μM of each primer, and 2.0 μL of genomic DNA (30 ng/μL) in a final volume of 10 μL. PCR was performed using the CFX Connect Real-Time PCR Detection System (Bio-Rad). Primer sequences used for fine mapping and PCR condition are listed in Supplementary  Table S3 and Table S2.
The co-dominant cleaved amplified polymorphic sequence (CAPS) marker was used to determine the allelic state at the candidate gene in all 346 F 2 individuals. PCR was performed with S05_14208507 F/R primer pair using the EmeraldAmp PCR Master Mix (Takara Bio). The PCR products were then digested with StuI restriction enzyme.
Each digestion reaction (10 μL volume) contained 2 μL of the PCR amplicon, 1 μL of buffer (provided with the enzyme), and 0.05 μL of StuI, and was incubated at 37 °C for 2 h. The digested products were subjected to electrophoresis using 1.0% (w/v) agarose gel. Primer sequences used for CAPS genotyping and PCR condition are listed in Supplementary Table S3 and Table S2.

Sequencing candidate gene
Total RNA was extracted from pepper leaves using the Sepasol-RNA I Super G extraction buffer (Nacalai Tesque, Kyoto, Japan), according to the manufacturer instructions. The isolated total RNA was purified using the High-Salt Solution for Precipitation (Plant) (Takara Bio). To perform reverse-transcription PCR (RT-PCR), all RNA samples were purified and treated with DNase using the NucleoSpin RNA (Takara Bio) to remove any traces of contaminating DNA. Then, 2 μg of RNA was reverse-transcribed with the oligo(dT) 21 primer using the ReverTra Ace (Toyobo, Osaka, Japan) by incubation at 42 °C for 60 min. To amplify the open reading frame (ORF) of the candidate gene, RT-PCR was performed with CaPelo full F/R primer pair using the cDNA template and KOD-plus Neo (Toyobo). The PCR products were cloned into the pTAC1 vector (BioDynamics Laboratory, Tokyo, Japan), and the resultant construct was sequenced on ABI PRISM 3100 genetic analyzer with the ABI PRISM BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA). Primer sequences used for PCR are listed in Supplementary  Table S1.

Expression analysis of candidate gene
The expression of the candidate gene was analyzed by realtime quantitative reverse-transcription PCR (real-time qRT-PCR) using the CFX Connect Real-Time PCR Detection System (Bio-Rad). Each 10-μL reaction contained 5 μL of SsoAdvanced Universal SYBR Green Supermix (Bio-Rad), 0.4 μM of each primer, and 2.0 μL of tenfold diluted cDNA. Transcript level of the candidate gene was normalized relative to that of CaActin (AY572427) reference gene, and relative expression of the candidate gene was calculated using the 2 −ΔΔCt method. Five biological replicates, each containing three technical replicates, were analyzed. Statistical analysis was performed using Student's t test with a p value of 0.05. Primer sequences used for real-time qRT-PCR and PCR condition are listed in Supplementary Table S1  and Table S2.
Agrobacterium carrying pGreenII-p35S-PepYLCIV-DNA-A+B, pTRV1, pTRV2::CaPDS, pTRV2::GFP, or pTRV2::CaPelota was grown overnight at 28 °C in 6 mL of LB medium containing rifampicin (50 μg/mL) and kanamycin (50 μg/mL), as described previously (Chung et al. 2004). The transformed Agrobacterium cells were harvested by centrifugation at 13,000 × g for 2 min at 20 °C and then resuspended in 10 mM MES, 10 mM MgCl 2 , and 200 μM acetosyringone to the final optical density of 0.1 at 600 nm. Cell suspensions were incubated at room temperature with shaking for 4 h. Agrobacterium cultures containing pGreenII-p35S-PepYLCIV-DNA-A+B and pTRV1 with pTRV2::PDS, pTRV2::GFP, or pTRV2::CaPelota were mixed at a ratio of 1:1:1 for inoculation. The inoculated plants were incubated in a growth chamber maintained at 23 °C and 12-h light/12-h dark photoperiod. Disease symptoms were surveyed at 23 dpi, and young upper leaves were collected and stored at − 80 °C until needed for DNA and/ or RNA extraction. PCR-based diagnosis of PepYLCIV and TRV was conducted using Pep uni F/PI uni R and TRV2 insert F/R primer pairs, respectively (Table S1). Statistical analysis was performed using Student's t test with a p value of 0.05.

Genetic mapping of the candidate gene
The F 2 individuals (n = 536), derived from a cross between BaPep-5 and BaPep-4, were agroinfiltrated with PepYL-CAV BAPep-V2 isolate; PepYLCAV was chosen for agroinfiltration because of its higher infectivity (less inoculation escape) than PepYLCIV (Table 1). At 21 dpi, DNA was extracted from the young upper leaves of F 2 individuals for PCR-based diagnosis of PepYLCAV. Out of 536 F 2 individuals, 346 were positive for PepYLCAV infection.

BaPep-5
BaPep Next, linkage analysis of PepYLCAV resistance in BaPep-5 was conducted using 316 SNPs obtained from RAD-seq data. The linkage map consisted of 12 linkage groups, which was equivalent to the chromosome number of pepper (C. annuum), and the average distance between DNA markers was 3.4 cM ( Supplementary Fig. S1). Linkage analysis performed by CIM resulted in the detection of a single peak on chromosome 5 with the highest logarithm of the odds (LOD) score of 118 (DSI at 54 dpi), 172 (DSI at 88 dpi), and 54.4 (viral DNA accumulation at 54 dpi) (Fig. 4). We named this single locus as pepper yellow leaf curl disease virus resistance 1 (pepy-1), considering its recessive nature.
Initial mapping inferred that the candidate gene is located between RAD markers S05_9614395 and S05_17254971, spanning a physical distance of 4622 kb in the reference genome sequence (CM334, Pepper.v.1.6) (Kim et al. 2014), and S05_14236545 was identified as the nearest RAD marker (Fig. 5). To reduce the size of the candidate region, one indel marker (S05_13024889) and five HRM SNP markers (S04_12817392, S05_14202764, S05_14228102, S04_14299330, and S04_14312617) were developed, based on the comparison of whole-genome sequences of BaPep-5 and BaPep-4. A total of 29 recombinants were identified based on genotypic and phenotypic comparisons. The susceptible phenotype of recombinant #297 did not match the homozygous recessive (BaPep-5 allele) genotype of the S05_14202764 marker. Similarly, the resistance phenotype of recombinant #324 did not match the heterozygous genotype of the S05_14236545 marker. The S05_14228102 marker genotype perfectly co-segregated with the phenotype in all 346 F 2 individuals. These results indicated that the candidate gene was located within the genomic region of 14,203-14,237 k on chromosome 5. Consequently, the candidate region was delimited to a 34 kb region flanked by S05_14202764 and S05_14236545 markers on either side (Fig. 5).

Candidate gene analysis
Analysis of the annotation data of the reference genome of CM334 (Pepper.v.1.6) (Kim et al. 2014) revealed that the candidate region contained a single gene, CA.PGAv.1.6.scaffold186.68 (Fig. 5). Investigation of the reference genome of Zunla (Qin et al. 2014), another C. annuum accession, also showed that a single gene (LOC107870304) was located within the candidate region. BLAST analysis of the putative amino acid sequence of LOC107870304 (XP_016572281) revealed that this gene was a homolog of the messenger RNA surveillance factor Pelota, which corresponds to the TYCLV resistance gene ty-5 in tomato (Lapidot et al. 2015). We named this candidate gene as CaPelota.
Phylogenetic analysis was conducted using the putative amino acid sequences of Pelota belonging to diverse species, with the yeast Dom34 as an outgroup (Fig. 6). The Drosophila, zebra fish, and human Pelota proteins constituted a single clade, while the Pelota proteins of dicot and monocot plants constituted independent clades. CaPelota showed high sequence similarity with the amino acid sequences of Pelota from C. chinense (CcPelota; PHU17047.1), C. baccatum (CbPelota; PHT54191.1), and tomato (SlPelota; (AGJ52123.1). OsPelota, which confers resistance against bacterial blight in rice (Qin et al. 2018;Zhang et al. 2018), was grouped in the clade of monocot plants.

pepy-1 encodes CaPelota
Whole-genome resequencing of BaPep-5 and BaPep-4 and comparison of their genome sequences revealed a SNP (A to G in BaPep-5) at the splice site of the 9th intron (Fig. 7a). According to the Chambon's rule, the first two and the last two nucleotides of introns are GT and AG, respectively (Rédei 2008 showed that the 9 th intron was not spliced, and rather inserted into the BaPep-5 transcript (Fig. 7a). Pelota has a tripartite structure with three domains (Lee et al. 2007). Analysis of the putative amino acid sequence of CaPelota of BaPep-5 revealed that 28 amino acids were inserted into domain 2 without causing a frameshift (Fig. 7b). The substitution of valine 16 (V) (susceptible M82 line) by glycine (G) (resistant TY172 line) in domain 1 is reportedly responsible for TYLCV resistance conferred by ty-5 in tomato (Lapidot et al. 2015). Similarly, the bacterial blight resistant O. sativa mutant lml1 carried an amino acid substitution in domain 1, while the genome-edited lml1 mutants (lml1-1, lml1-2, and lml1-3) and HM47 mutant carried a mutation in domain2 (Qin et al. 2018;Zhang et al. 2018). Thus, the position of the amino acid substitution in CaPelota of BaPep-5 differed from that in the Pelota of TY172, lml1 mutants, and HM47 mutant (Fig. 7b). Based on the A to G SNP in BaPep-5, we developed a codominant CAPS marker (Fig. 8a, Supplementary Fig. S2). When the PCR amplicon was digested with StuI, two fragments of 258 and 501 bp were detected in BaPep-4 (Fig. 8b), whereas a single 759-bp fragment was detected in BaPep-5, as it lacked the StuI recognition site (AGG CCT ) (Fig. 8b). Genotyping all 346 F 2 individuals with this CAPS marker S05_14208507 revealed that the SNP in CaPelota perfectly co-segregated with PepYLCAV resistance in the F 2 population ( Table 2). The genotype of this marker explained 86.5% (DSI at 54 dpi), 99.1% (DSI at 88 dpi), and 50.6% (viral DNA accumulation at 54 dpi) of the phenotypic variation.

Expression analysis and VIGS of CaPelota
The expression of CaPelota in the leaves of PepYLCAV-and mock-inoculated pepper plants was analyzed by real-time qRT-PCR (Fig. 9a). The expression of CaPelota was significantly lower in mock-inoculated BaPep-5 than in mock-inoculated BaPep-4 at 46 dpi ( Supplementary Fig. S3). Similar results were observed in PepYLCAV-inoculated BaPep-5 and BaPep-4 plants at 46 and 53 dpi. The expression of CaPelota in PepYLCAV-inoculated BaPep-5 and BaPep-4 plants was higher at 53 dpi than 46 dpi. DNA sequence analysis suggested CaPelota as a strong candidate gene responsible for PepYLCIV and PepYLCAV resistance in BaPep-5. To analyze the function of CaPelota in begomovirus resistance, we performed VIGS of CaPelota in the begomovirus susceptible accession No.218. VIGS is a powerful tool for reverse genetics in pepper, which is recalcitrant to transformation (Chung et al. 2004). The SGN VIGS tool was used to design the VIGS construct, pTRV2::CaPelota, specific for CaPelota without any offtarget results.
Our preliminary investigation showed that No.218 plants co-inoculated with PepYLCAV and TRV exhibited severe disease symptoms due to the synergistic interaction of the two viruses. Therefore, PepYLCIV, which shows lower pathogenicity than PepYLCAV, was used for VIGS analysis of CaPelota. At 23 dpi, No.218 plants infected with PepYLCIV and TRV2::PDS (n = 7) showed photobleaching due to the silencing of the phytoene desaturase (PDS) gene (Fig. 9b). Infection by PepYLCIV and TRV was confirmed by PCR prior to further analysis. CaPelota expression, PepYLCIV DNA quantification, and DSI evaluation were conducted in No.218 plants co-infected with Pep-YLCIV and TRV harboring partial sequences of the green fluorescent protein (GFP) gene or CaPelota. The expression of CaPelota in the young upper leaves of plants inoculated with TRV2::CaPelota (n = 4) was significantly lower than that in plants inoculated with TRV2::GFP (n = 6) (Fig. 9c). qPCR analysis showed that the accumulation of PepYLCIV DNA in the young upper leaves was significantly lower in plants inoculated with TRV2::CaPelota compared to those infected with TRV2::GFP (Fig. 9d). Furthermore, PepYL-CIV symptoms and average DSI were significantly reduced in TRV2::CaPelota-inoculated plants (Fig. 9b, e). Since the VIGS of CaPelota in No.218 plants resulted in the gain of resistance against PepYLCIV, we concluded that begomovirus resistance in BaPep-5 is caused by the putative lossof-function mutation in CaPelota.
Most of the virus resistance genes in plants either prevent viral replication or restrict viral replication to the cells targeted by the virus to enter the host (Kang et al. 2005a, b). The pepy-1-conferred resistance of BaPep-5 is a levels of virus tolerance rather than an immune response. In plants harboring the pepy-1 gene, a low level of PepYLCIV and PepYLCAV DNA accumulation was detected in systemic leaves, but with nearly no or mild symptoms (Figs. 1, 2, 3). The TYLCV resistance genes, Ty-1/Ty-3, Ty-2, and ty-5, are a level of virus tolerance as well rather than immunity, consistent with our results (Verlaan et al. 2013;Lapidot et al. 2015;Yamaguchi et al. 2018).
The effectiveness of Ty gene-mediated resistance against different begomoviruses is well studied in tomato. The Ty-3a gene is fully effective against PepYLCIV but only partially effective against TYLCKaV . Additionally, Ty-2-conferred resistance is effective only against some monopartite begomoviruses, such as TYLCV-IL, but not against TYLCV-Mld and other monopartite or bipartite begomoviruses (Hanson et al. 2000;Barbieri et al. 2010;Tsai et al. 2011;Shahid et al. 2013;Prasanna et al. 2015;Ohnishi et al. 2016;Yamaguchi et al. 2018). Furthermore, ty-5 is effective against TYLCV but completely ineffective against the bipartite tomato mottle virus (Gill et al. 2019). In BaPep-5, the resistance conferred by pepy-1 was more effective against PepYLCIV than against PepYLCAV at approximately 60 dpi. PepYLCIV-infected BaPep-5 plants showed constant disease symptom with average DSI of 0.5 (Table 1). In contrast, PepYLCAV-infected BaPep-5 plants showed DSI score of 0-3 and plants with a higher disease scores tended to show higher viral accumulation (Table 1; Fig. 2c). PepYLCAV-infected pepy-1 homozygous F 2 individuals also showed the symptom range of DSI 0-3 at 54 dpi, but symptom recovery was observed at 88 dpi (Fig. 3). It is possible that the disease symptoms of PepYLCAVinfected BaPep-5 would have recovered until 88 dpi; however, the DSI scores of these plants were recorded only until 62 dpi (Table 1). Our preliminary data showed that Pep-YLCAV exhibits higher pathogenicity than PepYLCIV in the begomovirus susceptible tomato cultivar 'Momotaro' and pepper accession No.218 . Overall, we conclude that the pepy-1 gene-conferred resistance is effective against PepYLCIV as well as PepYLCAV, and the difference between the response to PepYLCIV and PepYL-CAV corresponds to the higher pathogenicity of PepYLCAV, which is a recombinant virus derived from PepYLCIV .
The segregation ratio of phenotype in the F 2 population did not fit the expected 3:1 by χ 2 test. In our previous study, graft-inoculation was a reliable method to avoid inoculation escapes, which are inevitably observed in agroinfiltrated pepper plants (Koeda et al. 2018). However, because the graft-inoculation method is much more laborious compared to agroinfiltration, we chose efficient agroinfiltration in the study of the F 2 population. There is a possibility that selecting 346 PepYLCAV-infected F 2 plants out of 536 inoculated plants at 21 dpi has led to experimental bias in the segregation ratio. However, because linkage analysis conducted by CIM detected a single peak above the threshold only in chromosome 5 (Fig. 4), the genotype of SNP located at the splice site of the 9th intron of CaPelota showed perfect co-segregation with the observed phenotype (Table 2), and the phenotypic variation explained by the developed CAPS marker was considerably high (99.1% for DSI at 88 dpi), we concluded that the begomovirus resistance of BaPep-5 is controlled by a single recessive gene pepy-1. The field test in Indonesia using the same F 2 populations is ongoing, this will further clarify the segregation ratio of each genotype.
Protein translation is a strictly controlled cellular process and a part of the mechanism that eliminates aberrant 1 3 transcripts and proteins (Gerovac and Tampé, 2019). Rescue of nonproductive stalled ribosomes and the associated mRNA decay is mediated by Pelota (Dom34 in yeast) and Hbs1. In the transposon-inserted Drosophila mutants or knockout mutant of Pelota, the production of Pelota protein is disrupted or eliminated, and these mutants show resistance against drosophila C virus (DCV), a single-stranded RNA virus (Wu et al. 2014). It is presumed that the lack of Pelota protein impairs the recycling of stalled ribosomes, which reduces the availability of free ribosomes and limits the high-level synthesis of DCV capsid proteins, thus restricting the replication of DCV. Interestingly, the pelota mutant pelo −/− of Drosophila also restricted the replication of cricket paralysis virus (a single-stranded RNA virus), In female mosquitoes (Aedes aegypti) co-existing with Wolbachia, a bacterial endosymbiont, the Pelota protein was downregulated, and its subcellular localization was altered, which possibly contributed to the reduction in dengue virus replication (Asad et al. 2018). Furthermore, the micro RNA (miRNA) Bta-miR-2411 induced the downregulation of the Pelota mRNA in the Madin-Darby bovine kidney cells, which restricted the replication of bovine viral diarrhea virus . These studies strongly suggest that Pep-YLCIV and PepYLCAV resistance in BaPep-5 (pepy-1) is the result of restricted replication of begomoviral DNA by Pelota deflection (Fig. 2).
Restricted viral replication caused by a mutation in Pelota seems to be the common phenomenon observed in plants and animals; however, the effect of mutation on other traits varies among organisms. The knockout mutation of Pelota is lethal in mice (Mus musculus) but not in yeast and Drosophila (Eberhart and Wasserman 1995;Davis and Engrebrecht, 1998;Adham et al. 2003;Wu et al. 2014). However, the growth of the dom34 mutant strain of yeast is particularly slow at low temperature because of the reduced ribosome supply during the transition from the stationary phase, and male infertility is observed in pelota mutant of Drosophila. Although the mutation of OsPelota in rice resulted in resistance against bacterial blight and rice blast, the mutant plants showed lesion mimicking phenotype and exhibited poor performance in agronomic traits, including plant height, tiller number, panicle length, seed-setting, and 1000-grain weight (Feng et al. 2013;Qin et al. 2018;Zhang et al. 2018). On the other hand, in a near isogenic BC 4 F 3 population of tomato developed by crossing TY172 (TYLCV resistant) and M82 (TYLCV susceptible), although begomovirus noninoculated ty-5 homozygous plants displayed a reduction in fruit size and total fruit yield compared with Ty-5 homozygous plants, they showed no lesion mimicking phenotype or arrested growth (Lapidot et al. 2015). BaPep-5 showed no visible growth defects or lesion mimicking phenotype in begomovirus non-infected plants ( Supplementary Fig. S3). The less negative effects of Pelota mutations in pepper and tomato genetic materials enable their use for agricultural production. Because almost all pepper plants cultivated in the field are infected by begomoviruses in Indonesia (Koeda et al. 2016), even if there was a small penalty caused by the Pelota mutation, the benefit of resistance would most likely exceed the penalty cost. Evaluation of agronomic traits of BaPep-5 and its progeny in the field under non-infected and begomovirus-infected conditions will be highly informative, and such experiments are currently being set up.
Analysis of the putative amino acid sequence of CaPelota in BaPep-5 revealed that 28 amino acids were inserted into domain 2 without causing a frameshift (Fig. 7b). Based on this finding, we conducted VIGS of CaPelota in the begomovirus susceptible pepper No.218 and revealed that silencing Fig. 7 Molecular genetic identification of CaPelota as the candidate gene responsible for the PepYLCAV resistant phenotype of BaPep-5. a Schematic diagram of CaPelota in BaPep-4 and BaPep-5. Exons (closed white boxes) were predicted from the corresponding cDNA sequences. The splice site contained an A in BaPep-4 and a G in BaPep-5, which resulted in a splicing variant involving the 9 th intron of CaPelota in BaPep-5. b Alignment of the predicted amino acid sequences of CaPelota and its homologs in other organisms. The amino acid sequence of CaPelota of BaPep-4 (C. annuum) and BaPep-5 (C. annuum) was aligned with those of its homologs in Oryza sativa, Arabidopsis thaliana, Solanum lycopersicum (TYLCVresistant TY172 and -susceptible M82), Drosophila melanogaster, and Homo sapiens, along with Dom34 of S. cerevisiae using Clustal Omega. Red underbar represents the inserted 28 amino acid residues in BaPep-5; black underbars represent the sequence motifs 1, 2, and 3 of domains 1, 2, and 3, respectively (Lee et al., 2007). Mutation points 1, 2, 3, and 4 indicate mutated residues in TY172 (ty-5) (S. lycopersicum), lml1 (O. sativa), genome-edited lml1 mutants (lml1-1, lml1-2, and lml1-3) (O. sativa), and HM47 mutant (O. sativa), respectively (Lapidot et al. 2015;Qin et al. 2018;Zhang et al. 2018  resulted in the gain of resistance against PepYLCIV (Fig. 9). Our result is also supported by recently reported patent that the artificial Maor (C. annuum) mutant of CaPelota possessing 3 amino acids deletion in the 4th exon showed resistant against PepYLCIV, pepper leaf curl virus, and pepper huasteco yellow vein virus (Prins et al. 2019). Interestingly, the position of mutation in CaPelota in BaPep-5 differed from previously reported mutations of Pelota genes. The amino acid substitutions caused by SNPs in TY172, lml1, and HM47 mutants conferred resistance against begomovirus or bacterial blight (Lapidot et al. 2015;Qin et al. 2018;Zhang et al. 2018). Introducing the deletion or insertion of several nucleotides at the splice site will be relatively easy to  achieve by genome editing using the CRISPR/Cas9 system compared with introducing nonsynonymous substitutions, which needs the additional delivery of the oligonucleotide donor sequences (Bortesi and Fischer 2015). Thus, our findings provide a new opportunity for the control of begomoviruses using genome editing techniques targeted on Pelota.
As part of the Columbian Exchange, Capsicum was first introduced into Europe at the end of the fifteenth century, and its use as a horticultural crop spread rapidly throughout the Old World. Since Indonesia is not the place of origin of Capsicum, the genetic diversity for local Capsicum cultivars is presumably not as high as in Central and South America. A large outbreak of PepYLCD caused by begomoviruses in Indonesia might have increased the chance for the local farmers to select BaPep-5, which has a mutation in a plant disease susceptibility (S) gene, such as Pelota. In practice, the majority of resistance breeding programs aim to introgress plant resistance (R) genes from wild species into their cultivated relatives (Pavan et al. 2010). In most cases, resistance conferred by R genes can be overcome by pathogens, resulting in widespread epidemics ( Van der Hoorn et al. 2002;Panstruga and Dodds 2009). In contrast to R genes, the loss-of-function mutation of S genes often leads to durable and broad-spectrum resistance, such as mlo-based and eIF4-based resistance (Büschges et al. 1997;Lellis et al. 2002;Piffanelli et al. 2004;Kang et al. 2005a, b;Humphry et al. 2006;Nieto et al. 2006;Bai et al. 2008;Pavan et al. 2008). BaPep-5 is a highly valuable genetic resource with recessive resistance to begomoviruses, and the DNA marker developed in this study will greatly facilitate the marker-assisted breeding of begomovirus resistance in peppers.