Interspecific and Intraspecific Hybrid Rootstocks to Improve Horticultural Traits and Soil-Borne Disease Resistance in Tomato

Tomato rootstocks are important to increase yield and control soil-borne pathogens, increasing vigor for a longer crop cycle and tolerance to biotic and abiotic stress. This study, conducted in the greenhouse of Sunchon National University during the period from 2019 to 2022, aimed to identify local soil-borne-disease resistant interspecific and intraspecific tomato hybrid rootstocks. The 71 interspecific hybrids (S. lycopersicum × S. habrochaites) showed that the germination vigor (GV) was less than Maxifort, except for several combinations. The germination rate (GP) of cross-species hybrids showed a different pattern according to the hybrid combinations, of which three combinations showed less than 30%. The horticultural traits, such as GV and GP, of the intraspecies hybrid (S. l × S. l) combination were significantly improved compared to that of Maxifort. In 71 combinations (S. l × S. h) and 25 combinations (S. l × S. l), MAS was used to evaluate the resistance of eight genes related to soil-borne pathogens, four genes related to vector-mediated pathogens, and three genes related to air-borne pathogens. The results showed that the new hybrid combination had improved resistance over the commercial-stock Maxifort. Therefore, interspecies and intraspecies hybrid techniques for breeding commercial rootstocks can be utilized as a way to improve horticultural properties and resistance to soil-borne diseases in tomato.


Introduction
Tomatoes (Solanum lycopersicum, Solanaceae; 2n = 2x = 24, 25, 26) are an important vegetable crop grown worldwide from temperate to tropical and subtropical regions and are particularly valued for their nutritional qualities [1][2][3]. World production of fresh tomatoes for 2020 was about 182 million tons, planted on 4.76 million hectares in 168 countries [4]. Tomatoes are supposed to have originated in western South America and were domesticated in Central America. The tomato plant has a number of distinguishing properties, including fleshy fruit, a sympodial stalk, and compound leaves, which are not seen in other model plants (such as rice and Arabidopsis) [5]. The wild tomato Solanum habrochaites S. Knapp et D.M. Spooner (formerly Lycopersicon hirsutum Dunal) is the most resilient and has the showiest floral displays of all the wild tomato species. This species can be found on the western slopes of the Andes at heights ranging from 400 to 4000 m, from central Ecuador to central Peru [6]. Peralta et al. [7] found that the cultivated tomato is closely related to 13 wild Solanum species, all of which can be crossed with tomatoes with varying degrees of difficulty. All wild tomatoes are diploid (2n = 24), may be crossed with cultivated tomatoes and serve as a breeding source for desirable qualities such as enhanced production yield, fruit quality, disease, and abiotic stress resistance. Wild tomato species are very useful in evolutionary research [8,9]. The current effort on the tomatogenome-sequencing project has yielded significant data for tomato research [5]. This wild

Tomato Crossing and Production of F1 Hybrids
Specific crosses were conducted using 17 inbred parental lines as female parents (S. l), 5 accessions of wild species as male parents (S. h), and 2 inbred lines as male parents (S. l) The selection of these parents was based on growth traits such as high germination, vigor, many fruit sets, high seed products and diseases resistance, especially soil-borne diseases, and tolerance to abiotic stresses. All crosses were performed by hand pollination. The female plants (S. l) were emasculated before the flower opened (removing stamens, petals and sepals), typically a day before the anthesis (evening time). Pollen was collected from the male parent (S. h). A handle vibrator was used to collect pollen on the tip (volume 1.5 mL) and apply pollen to the stigma surface (morning time). All the crosses were made in the morning between 09:00 to 11:30 a.m. local time. After pollination, all flowers were tagged with labels that included names and dates. Harvesting of tomato fruit was carried out daily until the end of the season. The fruit set rate was determined as the total number of fruits divided by the total number of pollinated flowers on each plant. Seed yield was determined as the total seed obtained divided by the total fruit harvested from each plant. The F 1 hybrids were transplanted on both sides of the bed (width: 1.2 m; row space: 0.8 m). The in-row distance between plants was 30 cm. Each experimental unit (EU) consisted of 2 plants. All cultural practices (fertilization, irrigation, weeding, and disease and insect control) were performed as recommended for commercial greenhouse tomato production.

Evaluated Traits and Marker-Assisted Selection (MAS)
The germination percentage (GP) was counted at the time germination was completed (100 seeds per line were sown in then-rolled towel papers) [33]. The germination vigor was measured by counting the number of seedlings emerging daily (7-14 days) from the day of planting the seeds in a medium till the time germination was complete (one hundred seeds were sown in 105-cell trays containing cocopit soil mix). Germination Index (GI) or Germination Vigor (GV) was computed by using the following formula: GV = n/d (n: number of seedlings emerging on the day, d: the day after sowing) [33]. The GV rating was scored for each line/hybrid combination based on a 1 to 9 scale (note, 1 = very weak, 2 = very weak to weak, 3 = weak, 4 = weak to medium, 5 = medium, 6 = medium to strong, 7 = strong, 8 = strong to very strong, 9 = very strong) following Juss and Shaw et al. [34]. Plant growth measurements, internode length (IL) F 1 (from the base to the leaf 3rd, 5th, 7th, 9th, 11th), total root length (TRL) (from the root collar to the end of the root by meters), root fresh mass (RFM) (after washing for 3 h with scales), were measured 60 days after transplanting (DAT). The seedling length (SL) and the plant height (PH) (from the base to the end of the stem in meters), seedling stem diameter (SSD), and plant stem diameter (PSD) (from cotyledon to 1st leaf, the leaf 9th to 10th internode from the base by digimatic caliper) were measured twice, 30 days after sowing (DAS) and 60 days after transplanting (DAT). The method of root collection was to dig from the ground by spraying water gradually because all tomato plants were planted on the ground directly. The yield was measured by the average number of seeds per fruit (ANSF) for each plant (harvesting of tomato fruit was carried out daily until the end of the season for S. h, but S. l was collected only 2 or 3 times for good fruit (big size, no blossom-end rot, no cracked fruits and disease). The methods of marker-assisted selection (MAS) were performed based on HRM curve method and judged with resistance or susceptibility instead of trait values [35]. Then, the experiment evaluated resistance with a marker such as: Fusarium wilt; I2 [36]

DNA Extraction
For extraction of genomic DNA, young leaves (1 g) of 172 tomato cultivars, wild species, and F1 hybrids were collected and genomic DNA was isolated by CTAB method [49]. PCR was performed in a total volume of 10 µL containing 2 µL of genomic DNA, 0.5 µL of forward and 0.5 µL of reverse primers (10 pmol), 5 µL of Prime Taq Premix and 2 µL of distilled water. The reaction condition was as follows: samples were primarily denatured at 95 • C for 5 min; followed by 30 cycles of 95 • C for 30 s, 60 • C for 30 s, and 72 • C for 30 s and final elongation at 72 • C for 5 min in a GenAmp PCR system 9700 (Applied Biosystems, Seoul, Korea). The amplicon was run on a 1.2% agarose gel. PCR conditions for Ty2 and cf9 were: initial denaturation at 95 • C for 5 min followed by denaturation at 95 • C for 30 s, annealing at 60 • C for 30 s, elongation at 72 • C for 30 min repeated for up to 30 cycles and final elongation at 72 • C for 5 min. Finally, the reaction mixture was cooled down to 4 • C and the amplicon was loaded on 1.2% agarose gel concentration.

Horticultural Traits and Marker Selection of S. lycopersicum
A total of 101 lines were developed from 43 cultivars. The selection was based on growth traits such as germination percentage (GP), germination vigor (GV), a high number of fruit sets, high seed products, and disease resistance, especially to soil-borne diseases.

Horticultural Traits and Marker Selection of Solanum habrochaites
A total of 42 lines were selected from 44 accessions. The selection was based on growth traits such as germination percentage (GP), germination vigor (GV), many fruit sets, and high seed products. The seedlings of S. h had purple hypocotyls above the soil level, and the length of the hypocotyl was shorter than S. l under the same condition ( Figure 1). In the experiments, the S. h were indeterminate (ID). In the experiments, S. h had genes resistant to Fusarium crown and root rot (J3), tomato spotted wilt virus (Sw5/TSWV), tomato yellow leaf curl virus (Ty2/TYLCV), late blight (Ph3), gray leaf spot (Sm-565), and Fusarium wilt (I2); root-knot nematode (Mi23) and leaf mold (Cf9) were not amplified ( Table 2). The germination percentage (GP) of S. h was as follows: less than 50% was recorded by SN-15 (2.38%), between 50 to 85% were 14 lines (33.33%), and more than 85% were 27 lines (64.29%), as shown in ( Table 2). The germination vigor (GV) of S. h was described; 3 = weak were eight lines (19.05%), 5 = medium were 13 lines (30.95%), 7 = strong were 13 lines (30.95%), and 9 = very strong were eight lines (19.05%) ( Table 2). The seedling length (SL) of S. h was measured from the base to the end of the stem at 30 DAS. The SL of S. h was as follows: longer than 10.5 cm were 21 lines (50%), and shorter than 10.5 cm were 21 lines (50%) ( Table 2). The maximum SL was recorded by SN-31 (16 cm), and the minimum SL was recorded by SN-10 (7 cm) ( Table 2). The seedling stem diameter (SSD) was measured between cotyledon to 1st leaf at 30 DAS. The SSD of S. h was as follows: bigger than 3.01 mm were 22 lines (52.38%), and smaller than 3.01 mm were 20 lines (47.62%) ( Table 2). The maximum SSD was recorded by SN-31 (3.66 mm), and the minimum SL was recorded by SN-11 (2.23 mm) ( Table 2). In the experiments, the plant height (PH) and the plant stem diameter (PSD) of S. h were non-significantly different, except SN-14 (14.85 mm) at 60 DAT (Table 2). In this research, high fruit setting and many seeds per fruit were target-specific for commercial rootstock. The average number of seeds per fruit (ANSF) of S. h was: less than 14.5 seeds/fruit for 27 lines (64.29%), and more than 14.5 seeds/fruit for 15 lines (35.71%) ( Table 2). ANSF values were highest in SN-12 lines (33 seeds/fruit) and lowest in SN-14 and SN-37 lines (6 seeds/fruit) ( Table 2).

Genetic Control and Horticultural Traits of F1 hybrids
A total of 96 new hybrid seed products, 71 interspecific hybrids (S. l × S. h) and 25 intraspecific hybrids (S. l × S. l) were identified (Table 3, Table 4). During crossing time, the fruit setting and seed yield of interspecific hybrids were determined by the phenotype of female parents. As a result, JTS01-3 produced high fruit setting and high seed product even though it had a small fruit size. In contrast, JTS37-3 had a larger fruit size but less fruit setting, and seed products.In the experiments, we observed that the female parent of D type was better than the female parent of ID-type in hybrid rootstock with respect to horticultural traits, such as germination percentage, germination vigor, and stem girth (Table 4). In experiments, the germination speed and seedling vigor of intraspecific hybrids were better than interspecific hybrids, respectively. The GP was also significantly affected by female-and-male-parent interaction, revealing genetic variation among hybrids for germination response. As a result, the GP of commercial rootstock (Maxifor) was only 85% ( Table 4). The GP of intraspecific hybrids was detailed: there were 98% three new hybrid combinations and 100% 23 new hybrid combinations ( Table 4). The GP of interspecific hybrids was: more than 85% were 38 new hybrid combinations (37.62%), and lower than 85% were 63 new hybrid combinations (62.38%) ( Table 4). In experiments, GV was important for commercial breeding such as rootstock grafting. The evaluation of GV was based on a 1-9 scale. The GV of intraspecific hybrids was described: 7 = strong were five new hybrid combinations (20%), and 9 = very strong, were 20 new hybrid combinations (80%) ( Table 4). The GV of F 1 in interspecific hybrids was described as follows: 1 = very weak were seven new hybrid combinations (9.86%), 3 = weak were 19 new hybrid combinations (26.76%), 5 = medium were 38 new hybrid combinations (53.52%) and 7 = strong were seven new hybrid combinations (9.86%). Therefore, the Maxifort was at a 7 on the scale and most of the intraspecific hybrids were at a 9 (Table 4, Figure 3). The SSD was measured between cotyledon to 1st leaf at 30 DAS. The SSD of Maxifort was 4.25 mm ( Table 4). The SSD of interspecific hybrids was as follows: smaller than (3.99 mm) were three new hybrid combinations (12%), bigger than (4.00 mm) were 22 new hybrid combinations (88%) ( Table 4). The SSD of interspecific hybrids was as follows: smaller than (3.99 mm) were 53 new hybrid combinations (74.65%), and bigger than (4.00 mm) were 18 new hybrid combinations (25.35%) ( Table 4). The seedling length (SL) was measured from the base to the end of the stem at 30 DAS. The SL of Maxifort was 23 cm (Table 4). The SL of intraspecific hybrids was as follows: shorter than 18.5 cm were 7 new hybrid combinations (28%), and longer than 18.5 cm were 18 new hybrid combinations (72%) ( Table 4). The SL of interspecific hybrids was: shorter than 18.5 cm were 46 new hybrid combinations (64.79%), longer than 18.5 cm were 25 new hybrid combinations (35.21%) ( Table 4). The PH of Maxifort was 257 cm (Table 4). All the PH of interspecific hybrids were higher than intraspecific hybrids. The PH of intraspecific hybrids was as follows: lower than 199 cm were 20 F 1 new combinations (80%), and higher than 199 cm were five new hybrid combinations (20%) ( Table 3). The PH of interspecific hybrids was: lower than 249 cm were 13 new hybrid combinations (18.31%), higher than 249 cm were 58 new hybrid combinations (81.69%) ( Table 4). The plant stem diameter (PSD) of new hybrid combinations and Maxifort were measured twice (between cotyledon to 1st leaf and the 9th to 10th leaf at 60 DAT). The PSD between cotyledon to 1st leaf of intraspecific hybrids was smaller than 14.30 mm. Interspecific hybrids were as follows: smaller than 15 mm were 43 new hybrid combinations (60.56%), bigger than 15 mm were 28 new hybrid combinations (39.44%), and Maxifort was 15.20 mm ( Table 4). The PSD between the 9th and 10th leaf of intraspecific hybrids was described as follows: smaller than 15 mm were eight new hybrid combinations (32%), between 15 to 18 mm were 18 new hybrid combinations (56%), bigger than 18 mm 3 new hybrid combinations (12%); intraspecific hybrids were as follows: smaller than 15 mm was recorded by JTS07-2 × SN-08 (1.41%), between 15 to 18 mm were 55 new hybrid combinations (77.46%), bigger than 18mm were 15 new hybrid combinations (21.13%), and Maxifort was 20.16 mm, as shown in Table 4. Internode length is an important agronomic characteristic affecting plant architecture and crop yield. The IL was measured from the base to the 3rd leaf, 5th leaf, 7th leaf, 9th leaf, and 11th leaf at 60 DAT. In the study, we selected new hybrid combinations that had a short internode length; therefore, as a result, among 71 Table 4). The IL of three new intraspecific hybrid combinations (JTS01-3 × JTS33-3, JTS05-2 × JTS33-3, JTS37-3 × JTS35-4) were similar to Maxifort and other new hybrid combinations of intraspecific hybrids were longer than Maxifort ( Table 4). The root system of interspecific hybrids was more than intraspecific hybrids (Figure 4). The total root length (TRL) of Maxifort was 75 cm, and the root fresh mass (RFM) was 258.56 g at 60 DAT ( Table 4). As a result, all the TRL of intraspecific hybrids were shorter than Maxifort, and the RFM of intraspecific hybrids was lighter than Maxifort, except (JTS35-3 × JTS35-4) was heavier than Maxifort (Table 4). All the RFM of interspecific hybrids were heavier than Maxifort. The TRL of interspecific hybrids was described as follows: 11 new hybrid combinations (15.50%) were shorter than Maxifort and 60 new hybrid combinations (84.50%) were heavier than Maxifort (Table 4). Table 3. List of gene/loci of hybrid new combination; interspecific (S. l × S. h), intraspecific (S. l × S. l), and commercial rootstock (Maxifort).       Hybrid tomato varieties have multiple disease resistances, especially to soil-borne pathogens, such as Maxifort being resistant to Fusarium wilt (I2), Verticillium wilt (Ve2), Fusarium crown and root rot (J3), Root-Knot nematode (Mi23), Tomato Spotted wilt virus (Sw5/TSWV), Tomato mosaic virus (Tm2a/ToMV), and Leaf mold (Cf9). As a result, all new hybrid combinations were more resistant than commercial rootstock Maxifort, except only one new hybrid combination (JTS35-3 × SN-08) had the same resistance as Maxifort but to different diseases (Table 3).

Discussion
Eight horticultural traits such as germination percentage (GP), germination vigor (GV), seedling length (SL), plant high (PH), seedling stem diameter (SSD), plant stem diameter (PSD), the average number of seeds per fruit (ANSF), plant type (PT), and marker-assisted selection (MAS) were used in our study for selection of tomato S. l (Table 1). These plant growth characteristics were an important indicator for commercial breeding. The ANSF was affected by fruit setting and fruit phenotype. In addition, tomato seed yield and quality are largely determined by the variety chosen for seed production [50]. According to Patwary et al. [51], the number of seeds per fruit varied from 26.0 to 107.70 in the winter to 4.02 to 49.39 in the summer. Tomato fruit set is best around 17-18 • C at night and 20-25.6 • C during the day [52,53]. Given that the maternal parent decides the quantity of ovules, supplies resources to the new embryo, and develops the seed coat, these findings are not surprising [54]. The germination percentages and the seed germination vigor were influenced by several factors, including the genetic constitution, mother plant environment and nutrition, harvest maturity, seed weight and size, mechanical integrity, degradation and ageing, and infections [55]. Additionally, disease resistance was selected as a single resistance and a combination of multiple resistances by molecular markers. Marker-assisted selection (MAS), which permits the selection of a single resistance gene or a combination of many resistance genes, has been widely and successfully used in tomato breeding projects, particularly for disease resistance [56,57].
Six horticultural traits, germination percentage (GP), germination vigor (GV), seedling length (SL), seedling stem diameter (SSD), plant stem diameter (PSD), average number of seeds per fruit (ANSF), and marker-assisted selection (MAS), were used in our study for the selection of tomato S. h (Table 2). In contrast, Ibrahim et al. [31], reported that seed germination rates are low, seed homogeneity is poor, and seed dormancy is high in wild species. The experiment revealed that seed germination of S. h was strong (Table 2), seed homogeneity was strong, and seed dormancy was low (data not shown). In addition, wild species are valuable sources of disease resistance and agronomic features in breeding efforts [58]. Earlier studies showed that S. h contain disease-resistance genes [59][60][61], pest resistance [62][63][64], cold tolerance, and quality traits [65,66] in some of these genes [67]. Additionally, Peralta et al. [2] found that the species is extremely vigorous, with a big spreading habit and a corolla up to 5 cm. This high vigor may be a major reason for its success in rootstock hybrids. In contrast, Huarachi Morejon et al. [68] reported that seed germination can be predicted by the genetic distance between female and male parents; however, some wide crossings can perform as well as or better than crosses with small-genetic-distance parents. The experiment revealed the seed germination percentage and germination vigor of intraspecific hybrids (S. l × S. l) were better than interspecific hybrids (S. l × S. h) ( Table 4). In this study, these two lines (JTS33-3 and JTS35-4) were high GP and strong GV and the five wild lines (SN-42, SN-06, SN-08, SN-20 and SN-33) were high GP and strong GV. However, after crossing these two lines, the result showed that the seed germination of intraspecific hybrids was better than interspecific hybrids (Table 4). Horticultural traits such as plant height (PH), internode length (IL) and stem girth were used in this study at 60 DAT. Furthermore, these traits of interspecific hybrids were non-significantly compared to Maxifort, and these traits were better than intraspecific hybrids (Table 4). Plant height and stem girth are usually strong indicators of plant vitality, which can lead to higher yields. It is important to understand the relationship between plant characteristics, growth parameters, and yield. The height of the tomato plant and the diameter of the fruit have a strong positive correlation [69]. Plant height has also been found to have a substantial positive relationship with leaf metrics such as the number of leaves, leaf area, and leaf area index, as well as the number of branches [70]. The most frequent tomato rootstocks are tomato hybrids (intraspecific hybrids) and interspecific hybrids [25]. Interspecific hybrids are more vigorous and usually produce highquality rootstocks with a large genetic diversity [71]. The stem diameter of intraspecific hybrids and interspecific hybrids was non-significantly different under the same condition (Table 4). Thus, the shortest stem diameter could be related to inadequate mineral, water, and photosynthetic transport from the earth to the plant [72]. This study indicated that quantifiable morphological differences exist between intraspecific hybrids and interspecific hybrids of root systems (Table 4 and Figure 4). In addition, Oztekin et al. [73] reported that two commercial rootstocks found variations in root density but not in average root diameter when it came to tomato rootstock root systems. Except for total root length, the root system morphology in tomato rootstocks varies by cultivar and is consistent through time. These distinctions could be used to classify cultivars for their suitability for use in certain growing situations, as well as to explain why specific rootstocks produce better growth and productivity [74]. The root system is a critical part of plant growth because it plays important functions in absorbing water and nutrients as well as a mechanical support and a storage organ as a barrier against pathogens [75,76]. S. l × S. h F 1 hybrids with multiple resistance to soil-borne diseases are the most frequent commercial rootstocks. However, the genetic potential of Solanum spp. for rootstock development has yet to be completely realized.
In conclusion, tomato rootstock with multiple resistances and tolerances to biotic and abiotic stresses are required in order to justify the extra cost added in the production. At the same time, it is important to obtain high rootstock seed quality based on high germination and vigor. Screening multiple inbred lines crossed with multiple wild relatives can help to achieve these goals. The production of seeds is a complex interaction of genetics and environmental factors.