Reviews
Possibility of genome editing for melon breeding
2024 年 74 巻 1 号 p. 47-58
詳細
2024 年 74 巻 1 号 p. 47-58
Genome editing technologies are promising for conventional mutagenesis breeding, which takes a long time to remove unnecessary mutations through backcrossing and create new lines because they directly modify the target genes of elite strains. In particular, this technology has advantages for traits caused by the loss of function. Many efforts have been made to utilize this technique to introduce valuable features into crops, including maize, soybeans, and tomatoes. Several genome-edited crops have already been commercialized in the US and Japan. Melons are an important vegetable crop worldwide, produced and used in various areas. Therefore, many breeding efforts have been made to improve its fruit quality, resistance to plant diseases, and stress tolerance. Quantitative trait loci (QTL) analysis was performed, and various genes related to important traits were identified. Recently, several studies have shown that the CRISPR/Cas9 system can be applied to melons, resulting in its possible utilization as a breeding technique. Focusing on two productivity-related traits, disease resistance, and fruit quality, this review introduces the progress in genetics, examples of melon breeding through genome editing, improvements required for breeding applications, and the possibilities of genome editing in melon breeding.
Melons (Cucumis melo L.) are annual herbaceous plants. Its main production areas range from the Mediterranean Region to East Asia. According to FAOSTAT statistics, in 2021, the global harvested area of melon was 1,077,369 ha, the production was 28,617,598.39 t, and the yield was 265,625 t/ha. It is a commercially important horticultural crop worldwide in terms of production quantity. Especially in Asia, melon is an important vegetable crop. Asia has the highest production, about 10 times higher than other areas, compared to Africa, America, and Europe, the Food and Agriculture Organization of the United Nations (FAO) (https://www.fao.org/faostat/en/#data/QC) (FAOSTAT 2021, Accessed on 01 November). Melons are used in various methods. In addition to being eaten as a dessert, some bitter or bland melons are consumed as vegetables. Due to its content of potassium, β-carotene, and vitamin C (ascorbic acid), the flesh of the melon fruit is considered a source of vitamins, minerals, and other health-promoting substances. Melon seeds are used as food in several countries, including India and Turkmenistan. Some melon fruits, such as small-fruited agrestis types and Waharman-type melons, are used after drying as preserved food when resources are scarce (Grumet et al. 2021). Melon can be widely utilized as food, thus improving its quality and productivity is essential, and a variety of new varieties have been developed.
Although melons have been classified into at least 19 horticultural subgroups and 6 groups (Pitrat 2016), there is little incompatibility or interbreeding between subgroups. Therefore, it can be said that melons have abundant genetic resources for breeding. Crossbreeding using DNA markers has been actively carried out using this rich genetic resource. Low incompatibility has accelerated the quantitative trait locus (QTL) analysis. Several QTL analyses with genetic diversity have provided practical DNA markers for important traits in breeding programs, such as improved fruit quality, plant disease resistance, and stress tolerance, and identified several associated genes. Selective breeding using DNA markers is a highly effective method, but it is not necessarily a panacea. The efficiency of this breeding method depends on the existing genetic diversity, which is limited in crops having experienced genetic bottlenecks during domestication. That is, if single nucleotide polymorphisms (SNPs) responsible for beneficial traits do not exist as genetic resources, selective breeding methods using DNA markers will not function satisfactorily. Mutagenesis and target-induced local lesioning (TILLING) platforms have been proposed as complementary techniques to complement this point (Dahmani-Mardas et al. 2010, Triques et al. 2007). This technology can introduce SNPs that did not exist in genetic resources. However, mutations may be presented simultaneously at multiple locations on the genome, and backcrossing is required to remove unnecessary mutations to develop cultivars. In this case, linkage dragging with the linkage of undesirable to desirable genetic traits should be considered. Although many excellent lines have been created by DNA maker selection and TILLING platforms, the limiting factors (genetic bottlenecks and linkage dragging) make conventional breeding a long process that can take, on average, a decade to develop a new plant variety. Expectations have been placed on genome editing as a new technology to accelerate breeding. This technology directly introduces mutations such as deletions and insertions into the target gene, so it does not introduce unnecessary mutations or chain resistance during the breeding process, and there is no need for backcrossing. Therefore, this technique can modify parental lines with excellent traits in the short term.
In genome editing technology, chimeric nucleases, such as zinc finger nucleases, transcription activator-like effector nucleases (TALENs), and clustered, regularly interspaced, short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (CRISPR/Cas9), have been used as mutagens. CRISPR/Cas9 is relatively easier to design and create than the other two (Fichtner et al. 2014). However, it is said to have lower target specificity than the TALEN method. These nucleases induce double-strand breaks (DSBs) in specific genomic regions, which are then repaired. DSB repair can be broadly divided into homologous recombination (HR) repair, which uses a template to accurately repair, and non-homologous end joining (NHEJ), which involves error-prone repair. The DSB repair in plants frequently utilizes the non-homologous end joining (NHEJ) pathway (Fichtner et al. 2014). That is, the repair after cleavage of DNA on a target sequence by a chimeric nuclease involves insertion or deletion of bases. As a result, mutations are induced on the target sequence, resulting in loss or gain of gene function (first generation genome editing technology) (Giordano et al. 2022, Ito et al. 2015, Liu et al. 2022, Nizan et al. 2023, Nonaka et al. 2017). Several genome-edited crops have been commercialized in the United States and Japan, including tomatoes with high GABA content, waxy corn, and soybeans with high oleic acid content (Demorest et al. 2016, Ezura 2022, Gao et al. 2020, Ku and Ha 2020, Nonaka et al. 2017). Recently, base editing (targeted AID) (Bastet et al. 2019, Hunziker et al. 2020, Shimatani et al. 2017) and/or prime editing (Lin et al. 2020) have become available, allowing base substitutions. These technologies have made it possible to introduce SNPs (second generation genome editing technology). With the advent of second-generation genome editing technology, much more advanced genome editing is now possible, and it is becoming possible to meet the more detailed demands of breeding sites. For this reason, we can expect that the use of genome editing technology in breeding will further accelerate in the future.
For using genome editing technology as a breeding technology, it is necessary to obtain genome and genetic information. Genetic analysis of melon was performed and the genome size was estimated to be approximately 454 Mb based on nuclear DNA content (Arumuganathan and Earle 1991, Garcia-Mas et al. 2012). Whole genome sequences are already available through the useful databases Melonomics v4.0 (Ruggieri et al. 2018, https://www.melonomics.net) and CuGenDBv2 (Yu et al. 2023, http://cucurbitgenomics.org/v2/). When selecting target genes and regions using genome editing technology in breeding, not only information on gene function but also information on the site and timing of gene expression is important. A gene expression database is publicly available, and the site, timing, and intensity of gene expression are disclosed. (Yano et al. 2018, 2020, https://melonet-db.dna.affrc.go.jp/. For selection of CRISPR/Cas9 target sites and design guide RNA, databases (CRISPR-P 2.0, Liu et al. 2017, http://crispr.hzau.edu.cn/CRISPR2/ and/or CRISPR Plant, Minkenberg et al. 2019, http://omap.org/crispr//) are available.
In this way, tools to actually utilize genome editing technology in melon breeding are being developed. In this paper, we examine the possibility of breeding melons using genome editing technology. First, we will introduce research on gene function analysis to determine the most important target genes in genome editing technology. Many studies have provided information on loci and genes associated with important traits. In particular, we will introduce progress in functional genomics analysis of traits related to important traits targeted for melon breeding (productivity and fruit quality). Next, we will introduce an attempt to utilize the CRISPR/Cas9 system for melon breeding. The final section discusses improvements in genome editing techniques for breeding applications.
Traits related to productivity and quality, such as disease resistance, fruit quality, sugar content, aroma, nutrition, and fruit color, are the most important in melon breeding. Therefore, many studies have been conducted to identify the genes involved in these traits. This information is important for selecting target genes to introduce traits into melons via genome editing. This section introduces information on the genes or loci related to important traits for melon breeding.
Disease resistanceDisease resistance traits are essential for increasing productivity, thus disease resistance is an important trait, and many efforts have been made to clarify the functions of resistance genes. Major diseases in melon are caused by viruses (33), fungi (10), and bacteria (3) (McCreight 2010–2011). We report on the research progress on resistance genes, using representative examples of several diseases. Genes introduced in this section are possible targets of genome editing.
Viral disease resistanceViral diseases seriously damage melon productivity. Since viruses always parasitize cells, using chemicals to control the corresponding diseases is not an option. Therefore, the use of genetic resistance in breeding is the most reliable and cost-effective means to minimize losses. This review introduces results on resistance genes and loci for resistance of five viruses that cause the most significant economic losses on melons: Cucumber mosaic virus (CMV), Watermelon mosaic virus (WMV), Zucchini yellow mosaic virus (ZYMV), Papaya ringspot virus (PRSV), and Melon necrotic spot virus (MNSV). For a review of viral resistance causing other diseases, see Martín-Hernández and Picó (2021).
CMV: CMV causes typical mosaic symptoms on melon leaves, plant stunting, mottle or mosaic on fruits, and yield losses. This virus may have the widest host range of any known plant virus. Major QTL mapping has demonstrated that a single gene, cmv1, can confer complete recessive resistance to CMV subgroup II strains (Essafi et al. 2009, Guiu-Aragonés et al. 2015). At least two additional QTLs (cmvqw3.1 and cmvqw10.1) are required to confer resistance to subgroup I strains (Guiu-Aragonés et al. 2014). cmv1 encodes the vacuolar protein sorting 41 gene (CmVPS41). An amino acid residue (L348R or G85E) was identified as a polymorphism associated with a resistance phenotype (Giner et al. 2017, Pascual et al. 2019). These mutations alter the cellular localization of CmVPS41, restricting viral movement within bundle sheath cells and causing resistance to CMV (Pascual et al. 2019, Real et al. 2023).
ZYMV: ZYMV induces vein clearing and yellowing, blisters and enations on leaves, and severe stunting. On fruits, ZYMV induces mosaic or necrotic cracks, marbling, and hardening of the flesh. Moreover, isolates belonging to the pathotype F induce wilting in melons carrying the Fn gene (Risser et al. 1981) instead of mosaic in melons carrying the Fn+ allele. The Fn gene (for Flaccida necrosis) is present in numerous melon accessions. Three resistance candidate loci, namely Zym-1, Zym-2, and Zym-3, have been identified (Danin-Poleg et al. 1997, 2002, Díaz et al. 2003, Périn et al. 2002). More detailed analysis of the Zym-1 locus has found NBL-1 (MELO3C015354), NBL-2 (MELO3C015353), and NBL-3 with typical resistance genes of the coiled coil-nucleotide binding site of the leucine rich repeat family (CC-NB-LRR, abbreviated as NBL). Two NAC family transcription factors (MELO3C015355 and MELO3C015357) are present in the adjacent region (Adler-Berke et al. 2021).
WMV: WMV induces mosaic, vein banding and deformation, such as blisters, filiformis and size reduction. It induces severe discoloration on fruits, with slight deformation in some cases. A major QTL for recessive resistance to WMV is located on chromosome 11 (Palomares-Ríus et al. 2011). Further analysis on chromosome 11 narrowed the locus of the resistance gene to a range of 141 kb. This locus was named wmv1551, and the SNP marker is known (Pérez-de-Castro et al. 2019). Three minor QTLs on chromosomes 4, 5, and 6 have been identified (Pérez-de-Castro et al. 2019).
PRSV: PRSV causes severe mosaicism, blisters, and malformations on leaves. Fruits may also show varying degrees of discoloration and deformity. Map-based cloning revealed that the PRSV resistance candidate gene (Prv) and the resistance candidate gene Fom-1 of Fusarium oxysporum f.sp. melonis races 0 and 2 are located adjacent to each other on chromosome 11. Both of these genes encode proteins of the Toll/Interleukin-1 receptor domain intracellular nucleotide-binding leucine-rich repeat family (TIR-NLR) (Brotman et al. 2013). Because the two proteins (Prv and Fom-1) have similar structures, it was necessary to prove that the putative Prv is resistant to PRSV. To achieve this goal, mutations were introduced into Prv using the CRISPR/Cas9 system. Prv knockout mutant individuals showed susceptibility to PRSV. This revealed that Prv is a PRSV resistance gene (Nizan et al. 2023). An unusual nonconserved domain, encoding a second NBS domain was found at the C-terminus of Prv, which is likely to be related to PRSV resistance.
MNSV: MNSV induces small, greasy-looking to transparent spots on new leaves. On the fruit surface, it induces relatively large spots, resulting in a sunk and distorted appearance. Necrotic spots can be produced on stems. The recessive nsv resistance gene was identified (Coudriet et al. 1981), and hybrid melon cultivars carrying such resistance are widely used in commercial agriculture. nsv encodes Cm-elF4E, which is involved in translation initiation factors. One amino acid residue on Cm-elF4E (228Leu) showed resistance against all strains of MNSV except MNSV-264 (Nieto et al. 2006). An RNA sequence in the 3ʹ UTR of MNSV was essential for the resistance/recessive phenotype (Díaz et al. 2004). Using the CRISPR/Cas9-mediated cytosine base editor, two kinds of mutations were introduced on Cm-elF4E, Cm-elF4E (C322T/C323G) with a stop codon and Cm-elF4E (C322T/C323T, P108L) (Shirazi Parsa et al. 2023).
Viral disease transmitted by Aphis gossypiiAphids are known pests that affect melon production. Heavy aphid colonization causes stunting, and severe leaf curling leads to plant death. Aphids excrete honeydew on the surface of leaves and fruits. This sticky, sweet substance allows the growth of sooty mold, which significantly reduces fruit quality. The aphid that colonizes melon crops is A. gossypii, which causes direct damage to melons and contributes to the spread of viral diseases as an efficient vector of viruses such as CMV, WMV, ZYMV, PRSV, and Cucurbitaceae-borne yellow virus (CABYV).
Entomologists, virologists, and plant geneticists have been working together to search for resistance genes in melons against aphid-mediated CMV infection. The Indian line PI 414723 and the Korean line PI 161375 show complete resistance to aphid-mediated CMV infection. In these lines, few aphids survived, and those that did survive had a low fecundity. These two phenotypes are thought to be co-segregated and controlled by a single sexual gene (Pitrat and Lecoq 1980, 1982). The genetic locus controlling this phenotype was named Vat, meaning “viral aphid transmission”. Aphid density is lower on Vat plants than on non-Vat plants (Thomas et al. 2016), this means Vat is involved in the resistance to aphid. Further genetic analysis revealed that the Vat locus encodes a 1467 amino acid protein belonging to the coiled-coil (CC)-nucleotide binding site (NBS)-leucine-rich repeat (LRR) family (Dogimont et al. 2014). An 11 kb DNA fragment containing the Vat resistance allele with a single coding sequence, native promoter, and terminator was transformed into an aphid-susceptible melon line (Vedrantais-Vat) and showed no response to the NM1 melon aphid clone. It showed high levels of resistance and complete resistance to the viruses upon inoculation with CMV, WMV, and ZYMV. Therefore, the 11 kb region containing the Vat gene seems to contain genes that function in both aphid resistance and virus resistance (Dogimont et al. 2014). To test this, resistance patterns to aphids and CMV were characterized by infection of nine aphid clones with PI 161375 and Vedrantais-Vat. In PI 161375, infection with eight aphid species showed CMV resistance, whereas in Vedrantais-Vat, infection with only three aphid species showed CMV resistance (Boissot et al. 2016a). These results indicate that at least additional loci are involved in CMV resistance induced by aphid infection in PI 161375. Accordingly, the Vat gene identified from PI 161375 was renamed Vat-1, and the presumed additional locus was named Vat-2. The genomic region spanning these genes is now called the Vat cluster (Boissot et al. 2016b). Five Vat homologs coding for proteins were identified in this cluster (Chovelon et al. 2021).
Resistance to other diseasesPowdery mildew (PM), Fusarium head blight (FW), gummy stem blight (GSB), and downy mildew (DM) are serious problem in melon production. these diseases are caused by fungi, oomycetes, and filamentous fung. Because these grow in high temperatures and humid conditions, these diseases are widespread in greenhouse. Most melon production are in greenhouse, thus the spread of these diseases leads to a decrease in melon production and causes economic losses. Therefore, controlling this disease is one of the important issues in melon breeding. Unlike viral diseases, some diseases caused by fungi, oomycetes, and molds can be controlled to some extent with fungicides. However, considering the negative effects of using fungicides on the environment and the human body, introducing resistance genes into melons is extremely effective. Therefore, introducing resistance to these pathogens into melons is essential for melon breeding. Abundant genetic resources have been analyzed, and many attempts have been made to search for resistance genes. This section introduces the loci, genes, and markers that have been discovered to confer resistance to these pathogens.
PM: Melon PM is often caused by Podosphaera xanthii (Px) and Golovinomyces cichoracearum (Gc) (Křístková et al. 2009, Li et al. 2017). G. cichoracearum occurs powderly in field environments in temperate and cold regions. P. xanthii is more frequently found in subtropical and tropical regions as well as in greenhouse crops (Natarajan et al. 2016, Ning et al. 2014). P. xanthii is emerging as the predominant pathogen in most countries (Bardin et al. 1999, Hosoya et al. 1999, Hudson et al. 2018, McCreight 2006, Ning et al. 2014, Pino et al. 2002, Zhang et al. 2012). Based on the reactions of isolates to melon differential lines, more than 28 physiological races of P. xanthii have been characterized (McCreight 2006). Due to race differentiation, resistant varieties are becoming susceptible to the disease in melons. It is necessary to breed varieties with multiple race-specific resistance genes and field resistance genes, and to date, several genes and QTL associated with resistance to PM have been mapped in different melon populations derived from several genetic sources (Table 1, Haonan et al. 2020). The identified QTLs have been listed by López-Martín et al. (2022).
Identifried PM resistnce loci
| Chromosome/Linkage group | Loci | Ref. |
|---|---|---|
| Chr02 | Pm-x | Périn et al. 2002 |
| Pm-x1.5, Pm-x3 | Fazza et al. 2013 | |
| Pm2F | Zhang et al. 2013 | |
| QTL (AR5) | Fukino et al. 2008 | |
| Pm-Edisto47-2 | Ning et al. 2014 | |
| Chr04 | qPx1-4 | Branham et al. 2021 |
| Chr05 | Pm-w | Pitrat 1991 |
| Pm-R1-2 | Yuste-Lisbona et al. 2011 | |
| PM-R5 | Yuste-Lisbona et al. 2011 | |
| Pm-AN | Wang et al. 2011b | |
| qPx1-5 | Branham et al. 2021 | |
| QTL PmV-1-Pi124112 | Perchepied et al. 2005 | |
| Chr10 | CmPMrs | Cui et al. 2022 |
| qPx1-10 | Branham et al. 2021 | |
| Chr12 | Pm-y | Périn et al. 2002 |
| QTL PmXII-1-Pi124112 | Perchepied et al. 2005 | |
| QTL (AR5) | Fukino et al. 2008 | |
| Pm-Edisto47-1 | Ning et al. 2014 | |
| CmPMR1 | Cui et al. 2022 | |
| Cmpmr2F | Zhang et al. 2023a | |
| qPx1-12 | Branham et al. 2021 | |
| qCmPMR-12 | Cao et al. 2021 | |
| BPm12.1 | Li et al. 2017 | |
| LGIX | Pm-1 | Teixeira et al. 2008 |
Several candidates or resistance genes have been found from detailed analysis of chromosomes 5 and 12. The gene MELO3C004311, which encodes TMV resistance protein N-like, has been identified on chromosome 5 as a candidate resistance gene (López-Martín et al. 2022). Another P. xanthii resistance gene for races 1 and 3 (Pm-wWMR 29) has been identified in a cluster of nucleotide-binding leucine-rich repeat receptors (NLRs) on chromosome 5. Although Pm-wWMR 29 is a homolog of the A. gossypii resistance gene Vat-1. Pm-wWMR 29 does not confer resistance to aphids, and Vat-1 does not confer resistance to PM (Boissot et al. 2024). The recessive resistance gene for race 2F “Cmpmr2F” (MELO3C002403), encoding allantoate amidohydrolase, was identified on chromosome 12 (Zhang et al. 2023a).
FM: FM is incited by the fungal pathogen Fusarium oxysporum Schlechtend f. sp. Melonis. Four physiological races of the pathogen (0, 1, 2, and 1.2) have been identified (Risser et al. 1976). Fom-1 and Fom-2 confer resistance to race 0 and race 2 and to race 0 and race 1, respectively. Fom-1 protein belongs to the TIR–NB–LRR type (Brotman et al. 2013). The putative Fom-2 protein includes the NB-ARC domain, an LRR-1domain, one Sfi1C (spindle body associated protein C-terminus) domain, and EAF (ELL-associated factor) domain (Wang et al. 2011a). These two genes have been utilized as commercial cultivars or as rootstocks. Race 1.2 can overcome these two resistance genes and include race 1.2y and 1.2w; race 1.2 yellowing (1.2y) induces yellowing symptoms before the death of the plants, and 1.2 wilting (1.2w) leads to wilting and death without the yellowing symptoms. Nine QTLs that provide resistance to race 1.2 have been found (Perchepied et al. 2005). Among the QTLs, fomIII.1 and fomVI.1 were specific for race 1.2y, while fomV.2 and fomXII.1 were only identified following inoculation with race 1.2w. Five QTL, fomIII.2, fomIII.3, fomV.1, fomXI.1, and fomXII.1, were effective against race 1.2y and race 1.2w. These results suggested that partial resistance to race 1.2 is governed by pathotype-shared loci and pathotype-specific loci. We screened 294 melon accessions, mainly of African and Asian origin, and found that the Indian accession PI124550 is resistant to both race 1.2y and race 1.2w. The highest resistance level was observed for both races in PI 124550, making it a promising breeding material with novel genes that confer useful resistance to both disease types in races 1.2 (Ishikawa et al. 2023). We look forward to future results regarding the discovery of resistance genes.
DM: Downy mildew (DM), caused by the obligate oomycete pathogen Pseudoperonospora cubensis, is a major foliar disease that causes great economic losses in melon production. Nine QTLs, including two major QTLs, for DM resistance were isolated from RILs (N = 169) generated from a cross between the resistant melon breeding line MR-1 and susceptible cultivar ‘Ananas Yok’neam’. The phenotype of the RILs was tested in both the greenhouse and growth chamber. Four QTLs (competitive allele-specific PCR [KASP] markers) were identified; of the major QTLs, qPcub-10.1 was stable across the growth chamber and greenhouse tests, whereas qPcub-8.2 was detected only in the growth chamber tests (Toporek et al. 2021, 2023). On chromosome 9, another QTL was identified from two F2 populations, which were bred from crossing the DM-resistant accession PI 442177 and two landraces, ‘Huangtu’ or ‘Huangdanzi’. Both QTL-seq and linkage map-based QTL mapping approaches have been used to identify a major QTL (DM9.1) associated with DM resistance (Zhang et al. 2023b).
GSB: GSB is a fatal fungus disease affecting most Cucurbit species, causing severe yield losses, especially in humid tropics and sub-tropics (Keinath et al. 1995, Zhang et al. 2017). In melon, the ascomycetous fungus, Didymella bryoniae. Rehm, is pathogenic. In the last sixty decades, although some GSB-resistance genetic loci for D. bryoniae have been reported in different resistant germplasm (Table 2), a functional GSB-resistant gene has not been identified.
Identifried GBS resistnce loci
| Loci | Hereiary type | Plant Intrduction accession | Ref. |
|---|---|---|---|
| Gsb-1 | Dominance | PI140471 | Sowell et al. 1966 |
| Frantz and Jahn 2004 | |||
| Gsb-2 | Dominance | PI157082 | Zhang et al. 1997 |
| Frantz and Jahn 2004 | |||
| Gsb-3 | Dominance | PI511890 | Zhang et al. 1997 |
| Frantz and Jahn 2004 | |||
| Gsb-4 | Dominance | PI482398 | Zhang et al. 1997 |
| Frantz and Jahn 2004 | |||
| Gsb-5 | Recessive | PI420145 | Frantz and Jahn 2004 |
| Gsb-6 | Dominance | PI482399 | Zhang et al. 1997 |
The rapid development of molecular biology and sequencing technologies has accelerated the isolation of GSB-resistant genes. In the past 5 years, several genes controlling resistance to GSB have been identified. A single recessive resistance gene (MELO3C022157) was isolated on chromosome 9. MELO3C022157 encodes a nucleotide-binding site leucine-rich repeat (BS-LRR), which is a resistance (R) genes. A comparison of the sequence of MELO3C022157 between the resistant and susceptible lines indicated polymorphism, such as deletions in the first intron, a 2-bp frameshift deletion from the second exon, and a 7-bp insertion in the fourth exon of the resistant line. Based on this information, a molecular marker has been developed (Hassan et al. 2018). Another single dominant candidate resistance gene (MELO03C012987) was identified on chromosome 4, which encodes a protein similar to the uncharacterized Avr9/Cf-9 rapidly elicited (ACRE) protein 146 (Hu et al. 2018). A new single dominant resistance gene, MELO3C010403, has been isolated on chromosome 7. MELO3C010403 encodes a protein with a wall-associated receptor-like kinase (WAK-RLK), critical for plant resistance to various pathogens; hence it is potentially associated with GSB resistance. Therefore, MELO3C010403 is the most likely candidate gene (Ma et al. 2023).
Sex-based expressionSex-based expression in melon flowers is an important trait in melon breeding programs. Commercially available melon varieties usually have bisexual flowers. Monoecious plants, with female and male flowers, are of great value in melon breeding as they obviate emasculation to ensure hybrid purity. Therefore, many studies have been conducted, and the genes and mechanisms involved in sex determination have been clarified. Genes involved in the regulation of sex expression in melons are dominated by the expression of andromonoecious (M), androecious (A), and gynoecious (G) genes, which define the dioecious pathway (Boualem et al. 2015).
The A and M genes encode CmACS11 and CmACS7, respectively, two aminocyclopropane-1-carboxylic acids (ACC) synthase (ACS) enzymes, which catalyze the rate-limiting step of ethylene production in the plant (Boualem et al. 2008, 2015). The G gene encodes a C2H2 zinc finger transcription factor of the transcription factor WIP protein subfamily (CmWIP1) (Martin et al. 2009). The expression of CmACS11 suppresses the expression of CmWIP1 (Boualem et al. 2015). CmWIP1 suppresses the expression of the carpel formation gene CRABS CLAW (CmCRC), inhibits differentiation into female flowers, and promotes male flower formation (Zhang et al. 2022). Ethylene generated by CmACS7 mediates a signal transduction system and controls the expression of the homoeodomain class I transcription factor HD-ZIP I, CmHB40, suppressing stamen formation in hermaphrodite flowers and suppressing female flower formation (Rashid et al. 2023).
Recently, four promising candidate genes related to sexual expression in oriental melon were identified on Chr. 1 and 8, namely MELO3C015898 (transport inhibitor response 1), MELO3C015904 (SWR1 complex protein 4/DNA methyltransferase 1-associated protein 1), MELO3C024563 (putative UDP-N-acetylglucosamine-peptide N-acetylglucosaminyltransferase SPINDLY), and MELO3C024565 (MRNA decapping enzyme-like protein) (Kishor et al. 2021).
QTLs, tpbf2.1, Opbf3.1, and tpbf8.1 have been found on chromosomes 2, 3, and 8, respectively. Opbf3.1 and tpbf8.1 promote the development of pistil and stamen primordia by inhibiting the functions of CmWIP1 and CmACS-7, respectively, indicating the possibility of inducing hermaphroditism (Nashiki et al. 2023).
Sexual expression in developing flower buds is plastic, and it has been suggested that environmental factors influence sexual expression. A heterochromatin protein (CmLHP1) affects sex determination, and the relationship with epigenetic regulation is becoming clearer (Rodriguez-Granados et al. 2022).
Fruit traitMelon fruit, an economically and agriculturally important plant, shows huge diversity in pericarp and fruit flesh color, nutrition, sugar content, aroma, and ripening style. These traits are deeply involved in fruit quality and are among the most important traits for breeding. Therefore, several studies have been conducted that are related to fruit traits such as the color of fruit flesh, sugar content, aroma, and ripening. Many QTLs have been listed in the review by Shahwar et al. (2023). Here, we introduce the identified gene-related fruit traits.
Fruit color: Two genes (CmOr and CmFBN1) are involved in fresh, orange-colored fruit. There are three colors: orange, green, and white. The orange color is caused by the accumulation of β-carotenoid, a precursor of vitamin A, and its synthesis is controlled by the CmOr gene (Chayut et al. 2017, Tzuri et al. 2015). Carotenoid accumulation is regulated by the carotenoid sequestration protein FIBRILLIN1 (CmFBN1) (Zhou et al. 2023). A comparison of proteomic analysis of a high β-carotene melon variety and its isogenic line with a defective CmOr revealed impaired chromoplast formation and identified CmFBN1 as differentially expressed. CmFBN1 enhances CmOR-triggered carotenoid accumulation by stimulating plastid globule proliferation for carotenoid sequestration within chromoplasts.
Sugar content: Several related genes have been identified, including sucrose phosphate synthase (CmSPS1), acidic invertase (CmAI) (Hubbard et al. 1989), and sucrose transporters (CmTST1-3) (Cheng et al. 2018). QTL analysis with ‘PS’ and ‘SC’ showed three important loci that were related to sugar content on chromosomes 4, 5, and 7 (Argyris et al. 2017).
Aroma: Eighty-two volatile organic compounds (VOCs) biosynthesized in melon rind and flesh were detected by gas chromatography-mass spectrometry (GC-MS). An RIL population from the cross ‘PS’ × ‘VED’ showed 166 QTLs and a major QTL cluster was identified on chromosome 8, which was the same as the ripening-related QTL ETHQV8.1. QTLs related to esters, lipid-derived volatiles, and apocarotenoids were identified. Candidate genes have been proposed for ethyl 3-(methylthio) propanoate and benzaldehyde biosynthesis (Mayobre et al. 2021). A set of Near-Isogenic Lines (NILs) containing overlapping introgressions from the Korean accession ‘SC’ showed climacteric ripening QTL, ETHQB3.5, were also involved in VOCs biosynthesis (e.g., 3-methylbutyl acetate, benzyl acetate, 2-methylbutan-1-ol, 2-methylpropyl acetate, and 1-methylsulfanylbutan-1-one) (Zhao et al. 2023).
Ripening: Melon employs two types of ripening systems: climacteric (ethylene-dependent) and non-climacteric (ethylene-independent) ripening, depending on the horticultural subgroups. Other fruits do not include two systems, usually within the same species; only a single system is used (for example, tomato is only a climacteric type, and strawberries are only non-climacteric). Therefore, melon is a good source-material for the genetic analysis of climacteric and non-climacteric species. In climacteric fruits, ethylene triggers multiple responses (i.e., changes in texture, fruit firmness, expression of cell wall-degrading enzymes, fruit color, and aroma). Genetic analyses were performed to clarify the mechanism of ethylene evolution. Three major QTLs (ETHQB3.5, ETHQV6.3, and ETHQ8.1) were found in the NILs derived from the non-climacteric melon parental lines ‘SC’ and ‘PS’ and population of RILs, obtained by crossing a climacteric ‘VED’ and a non-climacteric variety ‘PS’ (Pereira et al. 2020, Vegas et al. 2013). ETHQV6.3 includes CmNOR related to the ethylene signaling pathway (Liu et al. 2022), and ETHQ8.1, which contains CmCTR involved in epigenetic control and CmROS1, which is a transcription factor (Giordano et al. 2022). Using the CRISPR/Cas9 system, we confirmed that these genes are key for controlling climacteric ripening. The knockout of CmNOR using the CRISPR/Cas9 system delays ripening in the climacteric type of melon (Liu et al. 2022). The destruction of CmCTR and CmROS in climacteric melons results in earlier ethylene production. This result indicates that the loss-of-function of these genes accelerates the ripening stage in climacteric fruits (Giordano et al. 2022).
Shelf-life: A long shelf-life is a crucial trait in fruit crop breeding because it directly affects food loss and waste. Globally, 14% of the total food production is lost before retail sales (English et al. 2019), and an additional 17% is wasted at the retail and consumer levels (Forbes et al. 2021), resulting in a staggering loss of USD 400 billion. These losses are especially significant for fruits and vegetables because of their poor shelf-lives compared to cereals and non-perishable goods (English et al. 2019). Therefore, the ability to modify plants to improve their shelf-life should improve the sustainability of the global food system by reducing food loss and waste. Moreover, a long shelf-life is crucial for exporting fruits and vegetables, as maintaining fruit quality and freshness requires close control of temperature, humidity, oxygen, and carbon dioxide, leading to high energy usage and costs. Therefore, producing a parental line with a long shelf-life can reduce the energy and costs required for export (Lamberty and Kreyenschmidt 2022).
Ethylene, a gaseous plant hormone, is a key regulator of fruit shelf-life. Ethylene is synthesized from the amino acid methionine and first converted into S-adenosyl-L-methionine (SAM) by SAM synthase. SAM is the primary methyl donor in plants and is involved in the methylation of lipids, proteins, and nucleic acids. SAM is converted by aminocyclopropanecarboxylate (ACC) synthase into 5-methylthioadenosine, which is converted back into methionine and then into ACC, the precursor of ethylene. ACC is oxidized by ACC oxidase (ACO) into ethylene. Therefore, ACO is a key enzyme involved in the regulation of ethylene production. In many plant species, the ACO gene has several homologs with highly conserved sequences. Melonet-DB (https://melonet-db.dna.affrc.go.jp/ap/top) shows that the melon genome contains five CmACOs, and the gene CmACO1 is predominantly expressed in harvested fruits (Yano et al. 2020). Previous studies have shown that suppressing the ACO gene via antisense or RNA-i genetic modifications extends fruit shelf-life by 5–14 d in climacteric melons (Ayub et al. 1996, Nuñez-Palenius et al. 2006). Substitutional mutation of ACO via ethyl methanesulfonate (EMS) treatment and TILLING selection also extended the shelf-life of melon (Dahmani-Mardas et al. 2010). The TILLING study showed that the G194D mutation resulted in longer shelf-life in melon fruit (Dahmani-Mardas et al. 2010). Based on these results, CmACO1 is expected to be a key gene for the shelf-life of melons.
Fruits and vegetables produced in Japan are highly regarded for their extremely high quality, safety, and reliability, and demand is gradually increasing. Introducing long-lasting traits is important for promoting exports, including further expanding sales channels. Japanese melon varieties are known for their high quality and are popular overseas. However, problems persisted, including the short shelf life and high transportation costs for exportation. It has been shown that CmACO1 is one of the key genes to control shelf-life through reverse genetic analysis with anti-sense and EMS technology. To breed a new melon lien with a long shelf-life, mutations should be introduced only in CmACO1 into existing melon varieties without affecting other genes. Antisense and RNAi technologies that have been used to date are based on transformation technologies, and crops produced using these technologies are classified as genetically modified crops (GM crops). GM crops are strictly regulated and require safety evaluation. Mutagenesis by EMS is also used as a breeding technique. Backcrossing is required to introduce a mutation into a single gene in an existing variety due to the simultaneous introduction of mutations into genes other than the target genes by this technique. The original genome set may be lost during backcrossing due to effects such as linkage drag. Crops produced by genome editing can be considered non-GM crops because genome editing technology does not leave foreign genes in the final crop. Additionally, since modifying only the target gene is possible, the breeding period is expected to be considerably shorter than conventional breeding methods such as mutation breeding. Recently, using genome editing technology, we succeeded in modifying CmACO1 to ‘Earl’s Favourite’, which is used as the parent variety line of Japanese high-grade melon (C. melo var. ‘reticulatus’). Modification of CmACO1 suppressed ethylene production from fruit. Wild-type and genome-edited melons were stored at room temperature for 14 days, and the storage periods were compared. In the wild type, fruit softened on the 14th day of storage, the epidermis collapsed, and deterioration of the fruit was observed. In contrast, in the genome-edited plants, the fruit remained firm even after 14 days of storage and did not deteriorate. Based on the above, we succeeded in introducing a long shelf-life trait by modifying his CmACO1 using genome editing technology (Nonaka et al. 2023).
For genome editing for crop breeding, genomic information is essential for select target genome. Beside, stable techniques are also important for introducing the CRISPR/Cas9 system into crop genome. In melons, genomic information with important traits has been made available through considerable efforts (see “Important traits for melon breeding”). Moreover, few genetic databases have been made available, such as Melonomics v4.0 (Ruggieri et al. 2018, https://www.melonomics.net), CuGenDBv2 database (http://cucurbitgenomics.org/v2/), and Melonet DB (Yano et al. 2018, 2020, https://melonet-db.dna.affrc.go.jp/ap/top).
In a recent study that used a CRISPR/Cas9 system in melons, 40–100% of transgenic lines with CRISPR/Cas9 showed a mutation in the target gene, these result meanst that the occurrence of mutations was high with CRISPR/Cas9 system (Giordano et al. 2022, Nonaka et al. 2023). Generally, to introduce the CRISPR/Cas9 system into crop genome, Agrobacterium-mediated transformation technology is used. Although several transformation protocols have been reported for melons, the transformation frequency is usually quite low and depends on the cultivar. Additionally, melons tends to become tetraploid in tissue culture with high frequency. To avoid these problems, tissue culture systems such as selection and redifferentiation and genome editing might be done without tissue culture, such as the iPB method (Imai et al. 2020).
AAs discussed in “Important traits for melon breeding”, many useful traits, such as disease resistance, fruit traits, and sexual expression, are often due to SNPs rather than genetic defects. Therefore, in breeding, genome editing technology that can induce SNPs is more practical than the simple CRISPR/Cas9 system that induces gene disruption. Single nucleotide substitution enzymes (base editor/target-AID) that can replace C to T, G to A, A to G, and T to C, have developed and can be used in rice and tomatoes. (Bastet et al. 2019, Hunziker et al. 2020, Shimatani et al. 2017). Regarding the targeted AID method, it can also be used in melon, where base and amino acid substitutions were successfully made in the MNSV resistance gene Cm-elF4E (Shirazi Parsa et al. 2023). Furthermore, prime editing, which specifies the location of the target site and replaces the target DNA nucleotide by linking reverse transcriptase to nCas9, has been developed and can be used in rice and wheat (Lin et al. 2020). CRISPR/Cas9, base editor/target-AID, and prime editing utilize the Cas9 protein. Due to the Cas9 protein, the end of the target sequence (PAM sequence) must be NGG. This point is the only factor that limits the selection of target sequences. The development of Cas proteins that can recognize various PAM sequences is progressing, and there will be almost no restrictions on target arrays.
Genome editing technology is advancing rapidly and is becoming an easy-to-use technology for breeding new varieties. From a simple technical perspective, there is great potential to accelerate the development of new varieties. However, there are some hurdles that must be overcome for commercial use, such as patent issues. In order to use it as a breeding technology and commercially utilize new varieties, there is a need to develop new technologies that keep patent fees low, as well as technology transfer systems, packaging, and consulting that make it easier to utilize existing genome editing technologies. becomes important. Widespread use also requires expanding public understanding of this technology.
This manuscript was mainly written by S.N. S.N. and H.E. planned and refined the overall structure of the text.
This work was supported by the Cabinet Office, Government of Japan, Cross-Ministerial Strategic Innovation Promotion Program and the ‘Technologies for Creating Next-Generation Agriculture, Forestry and Fisheries’ grant (funding agency: Bio-Oriented Technology Research Advancement Institution, NARO) to H.E. This research was conducted at Gene Research Center Facility at Tsukuba Plant Innovation Research Center.
