Turkish Journal of Agriculture and Forestry Turk J Agric For (2018) 42: 124-135 http://journals.tubitak.gov.tr/agriculture/ © TÜBİTAK Research Article doi:10.3906/tar-1709-52

Use of phenotypic selection and hypocotyl properties as predictive selection criteria in pumpkin (Cucurbita moschata Duch.) rootstock lines used for grafted cucumber (Cucumis sativus L.) seedling cultivation

1, 2 3 3 4 Onur KARAAĞAÇ *, Ahmet BALKAYA , Münevver GÖÇMEN , İsmail ŞİMSEK , Dilek KANDEMİR 1 Black Sea Agricultural Research Institute, Samsun, Turkey 2 Department of Horticulture, Faculty of Agriculture, Ondokuz Mayıs University, Samsun, Turkey 3 Antalya Tarım Hybrid Seeds Inc., Antalya, Turkey 4 Samsun Vocational School, Ondokuz Mayıs University, Samsun, Turkey

Received: 15.09.2017 Accepted/Published Online: 14.01.2018 Final Version: 26.04.2018

Abstract: In recent years, grafted cucumber seedling use has been rapidly increasing worldwide, especially for providing tolerance to stress conditions and positively affecting yield potential. However, graft incompatibility is still an important issue in grafted seedling production. Hypocotyl properties of rootstocks and scions are of great importance for ensuring graft compatibility. This study aimed to select superior rootstock genotypes based on certain selection criteria, such as hypocotyl morphology, grafted seedling visual evaluation, and graft compatibility properties, of 42 pumpkin (Cucurbita moschata) genotypes that are included in the cucumber rootstock breeding program and are resistant to Fusarium oxysporum f. sp. cucumerinum. Three C. moschata and three C. maxima × C. moschata rootstocks were used as control cultivars. All genotypes were grafted with the Gordion 1F cucumber cultivar using the splice grafting method. In the pumpkin rootstock genotypes, the cross-sectional area of the hypocotyls varied between 3.47 mm2 and 10.42 mm2, pith cavity area varied between 0.59 mm2 and 4.14 mm2, and pith cavity rate varied between 10.3% and 65.2%. Vascular bundle numbers of the genotypes were between 6 and 12. In the rootstocks, success rates of the grafts ranged from 56% to 100%. Among the rootstock/scion combinations, 100% success rate was obtained in 19 inbred pumpkin rootstocks. Scion rooting never occurred in 11 of the pumpkin rootstocks. The correlation analysis of the investigated properties showed that there was a statistically nonsignificant relationship between the cross- sectional area of the hypocotyl, vascular bundle number, and graft success rate (P > 0.05). The parameter with the greatest negative effect on graft success rate was scion rooting (r: –0.49, P < 0.05). The results showed that scion rooting depended on the pith cavity rate and decreased as the pith cavity rate decreased (r: 0.56, P < 0.05). The weighted-ranking method was employed to select the most promising pumpkin rootstock genotypes for grafted seedling production. According to the evaluations, 12 pumpkin rootstocks were selected as the most superior rootstock candidates. We aim to develop new hybrid rootstocks for cucumber in the near future by carrying out special combinations of tools from test-based hybridization studies among the selected pumpkin lines.

Key words: Cucurbit, grafting, pith cavity, selection

1. Introduction rates around the world: 94% in Japan (Bie et al., 2017), Cucumber cultivation is a widespread practice throughout 95% in Korea (Davis et al., 2008), 30% in China (Huang the world. In the last decade, global cucumber production et al., 2014), 11% in Taiwan (Lee et al., 2010), 10% in increased by 58% and reached 75 million tons (http:// Greece (Diánez et al., 2007), and 3% in France (Diánez et www.fao.org/faostat/en/#data/QC). A contributing factor al., 2007). In Turkey, 8% of the total cucumber seedling to the increase in production is the now widespread use production is composed of grafted seedlings. In 2015, of grafted cucumber seedlings, which enables tolerance grafted cucumber seedling production reached 9 million to biotic and abiotic stress conditions (Louws et al., 2010; seedlings (Yetişir, 2017). Today, rootstocks derived from Schwarz et al., 2010; Bekar et al., 2017). The first studies C. moschata (pumpkin), C. maxima × C. moschata, and on grafting in cucumber were carried out in the 1960s Cucurbita ficifolia species are used as rootstocks for to increase the tolerance to Fusarium wilting disease cucumber (Balkaya, 2014). The use of hybrid C. moschata and low temperatures (Marukawa and Takatsu, 1969). rootstock is more common due to its positive effect on the Today, grafted cucumber seedlings are used at varying yield and quality of cucumber, cheaper rootstock seeds, * Correspondence: [email protected] 124 KARAAĞAÇ et al. / Turk J Agric For and higher graft compatibility of the rootstocks (Sakata et quality. Numerous studies have focused on the effects al., 2007, 2008a, 2008b). The hole insertion, splice, tongue of hypocotyl thickness and vascular bundle number on approach, cleft, excised-root, and pin grafting methods are the success of grafting (Oda et al., 1993; Traka-Mavrona used in grafted cucumber seedling production (Davis et al., et al., 2000; Edelstein et al., 2004; Yetişir and Sarı, 2004; 2008; Lee et al., 2010; Balliu and Sallaku, 2017). However, Farhadi et al., 2016). However, no detailed studies on in recent years, hole insertion and splice grafting methods the effect of the pith cavity on graft incompatibility were are preferred in graft production facilities for providing found in the relevant literature. This study aims to select higher labor productivity (Oda et al., 2001; Lee and Oda, promising rootstocks based on their hypocotyl properties, 2010). Surface area is also higher for healing after grafting seedling visual evaluation, and graft compatibility among in these methods (Oda, 2007). the pumpkin rootstock genotypes included in the squash Incompatibility can emerge when squash rootstocks rootstock breeding programs. Furthermore, we aim are grafted with cucumber and melon (Oda et al., 1994; to provide insight into the future use of the selected Aloni et al., 2008). Graft incompatibility occurs as a result rootstocks in the development of hybrid rootstocks for of hormonal (Aloni et al., 2010), genetic (Edelstein et grafted cucumber seedling production. al., 2004), and physiological (Aloni et al., 2008) factors. Hypocotyl properties of rootstocks are important factors in 2. Materials and methods the success of grafted seedling production. An excessively 2.1. Plant materials long hypocotyl obstructs the attachment of the graft clip to In this study, 25 pumpkin rootstock lines in the S7 the dissection point, while an excessively short hypocotyl generation obtained from studies by Balkaya et al. (2010) complicates grafting (Yıldız and Balkaya, 2016). Therefore, and Karaağaç and Balkaya (2013) and 27 pumpkin (C. hypocotyl properties of candidate genotypes in rootstock moschata) lines obtained from the USDA-ARS Seed breeding programs and their graft compatibilities with Gene Bank (USA) were used as rootstocks (Table 1). different scions are among the important selection criteria These lines were tested against Fusarium oxysporum f. sp. (Cousins, 2005; King et al., 2010; Yıldız and Balkaya, cucumerinum, and resistant materials were selected for 2016). Today, rootstock breeding programs for vegetables the hybrid rootstock breeding program (Göçmen et al., are mostly carried out to establish disease resistance 2014). In this study, C. moschata (Early type F1, Kitora and increase yield. A review of the relevant literature F1, Bodyguard F1) and C. maxima × C. moschata hybrid showed that there were few studies concerning hypocotyl cultivars (TZ-148 F1, S. Shintosa F1, Ultra F1) were used morphology-based selection in rootstock breeding. This as the control rootstock cultivars (Table 1). The study study will provide a detailed investigation of the effects was carried out from 15 November 2012 to 15 March of hypocotyl morphology, seedling visual evaluation, 2013 in the seedling production facility of Antalya Tarım and graft compatibility-based selection on new cultivar Hybrid Seeds Inc., Antalya, Turkey, and the greenhouse development for the breeding of new cucumber rootstocks. and laboratories of Ondokuz Mayıs University, Faculty of Graft compatibility depends on the rate of scar tissue Agriculture, Samsun, Turkey. repair and the reestablishment of transfer in the phloem 2.2. Seedling experiments and xylem on the dissection surface (Martínez-Ballesta Cucumber seeds were sown on 15 November 2012 and 15 et al., 2010). This necessitates similar cambial regions in January 2013 in a peat + perlite mixture at a 3:1 rate 4 days the dissection point both in the rootstock and in the scion before the rootstock seeds. Seedlings were grown at 25 °C (Pina and Errea, 2005). In some cases, some morphological with 70% humidity and 300 µmol m–2 s–1 light intensity differences emerging in later periods of rootstock and for 16 h and at 20 °C at 70% humidity and 10 µmol m–2 scion development can negatively affect seedling quality s–1 light intensity for 8 h for 15 days (Karaağaç, 2013). in spite of the previously established graft compatibility. The experiment was carried out in three replications The structural state of the pith cavity is one of the factors and designed in accordance with a randomized block that can cause this problem. Differences due to pith cavity experimental design such that each replication comprised characteristics can even occur in genotypes that have 30 plants. The Gordion F cucumber cultivar (Monsanto similar vascular number and hypocotyl thickness. Having 1 Seed Company, USA) was grafted onto the genotypes an excessive pith cavity area can hinder the reconnection used as the rootstocks by using the splice grafting between the present vascular bundles (Tiedemann, method 15 days after seed sowing at Antalya Tarım Inc. 1989). Notably, the roots forming on the scion of grafted (Lee and Oda, 2010). Grafted seedlings were placed in seedlings can elongate and form roots with the rootstock a 0.02-mm-thick transparent polyethylene greenhouse through the pith cavity, thereby reducing the resistance of plastic and a controlled 0.05-mm black mulch plastic- cucumber to soil-borne factors (Lee et al., 2010). This also covered low tunnel in a greenhouse at 20 µmol m–2 s–1 negatively affects the success rate of grafting and seedling light intensity, 95% relative humidity, and 27 °C for 5

125 KARAAĞAÇ et al. / Turk J Agric For

Table 1. Locations of the providers and origins of the investigated pumpkin rootstock genotypes.

Genotype Origin Introducing institution Fruit groups Fruit size MO-1 Turkey OMUFA Butternut Small MO-2 Turkey OMUFA Butternut Small MO-3 Turkey OMUFA Butternut Small MO-4 Turkey OMUFA Cheese Big MO-5 Turkey OMUFA Cheese Medium MO-6 Turkey OMUFA Butternut Small MO-7 Turkey OMUFA Calabaza Small MO-8 Turkey OMUFA Calabaza Small MO-9 Turkey OMUFA Butternut Small MO-10 Turkey OMUFA Calabaza Small MO-11 Turkey OMUFA Cheese Small MO-12 Turkey OMUFA Cheese Small MO-13 Turkey OMUFA Calabaza Small MO-14 Turkey OMUFA Calabaza Small MO-15 Turkey OMUFA Calabaza Small MO-16 Turkey OMUFA Calabaza Small MO-17 Turkey OMUFA Crookneck Medium MO-18 Turkey OMUFA Calabaza Small MO-19 Turkey OMUFA Cheese Big MO-20 Turkey OMUFA Cheese Big MO-21 Japan BSARI Calabaza Small MO-22 Japan BSARI Calabaza Small MO-23 Japan BSARI Calabaza Small MO-24 S. Korea BSARI Kabocha Small MO-25 China BSARI Calabaza Small MO-26 India USDA (163230) Calabaza Small MO-27 Guatemala USDA (195312) Acorn Medium MO-28 India USDA (381820) Butternut Small MO-29 India USDA (381818) Cheese Medium MO-30 China USDA (418965) Cheese Small MO-31 China USDA (419083) Cheese Small MO-32 China USDA (419137) Cheese Small MO-33 Guatemala USDA (438579) Calabaza Small MO-34 Mexico USDA (438731) Cheese Small MO-35 Mexico USDA (438826) Calabaza Small MO-36 Mexico USDA (442251) Butternut Small MO-37 Mexico USDA (442257) Calabaza Small MO-38 Guatemala USDA (449348) Cushaw Medium MO-39 USA USDA (550692) Crookneck Medium MO-40 USA USDA (550693) Butternut Small MO-41 India USDA (634693) Calabaza Small MO-42 China USDA (634695) Calabaza Small

TZ 148 F1 Cultivar Clause Seed, France Calabaza Small

S. Shintosa F1 Cultivar Zeraim Seed, Israel Calabaza Small

Bodyguard F1 Cultivar Hild Samen, Germany Cheese Small

Early type F1 Cultivar Saitama Seed, Japan Cheese Small

Kitora F1 Cultivar Kaneko Seed, Japan Cheese Small

Ultra F1 Cultivar Saitama Seed, Japan Calabaza Small MO: C. moschata; OMUFA: Ondokuz Mayıs University, Faculty of Agriculture, Samsun, Turkey; BSARI: Black Sea Agricultural Research Institute, Samsun, Turkey; USDA: (PI) Seed Genebank, Maryland, USA.

126 KARAAĞAÇ et al. / Turk J Agric For days. Then the black much was removed. Grafted plants (MO-36) (Table 2). Yıldız and Balkaya (2016) reported that were periodically disinfected against fungal diseases with the hypocotyl length varied between 2.9 cm and 4.8 cm fungicides containing 8% metalaxyl + 64% mancozeb and in grafted cucumbers. In other studies of the same family 25 g fludioxonil + 10 g mefenoxam as active substances. for watermelon grafting, Yetişir and Sarı (2004) reported a Grafted seedlings were also periodically sprayed with 0.5% hypocotyl length between 2.0 cm and 6.9 cm and Karaağaç Cu-, 1.1% Mn-, and 0.5% Zn-containing 1 mL L–1 solution (2013) reported between 2.7 cm and 5.9 cm. Hypocotyl during 10 days (Karaağaç, 2013). length is an important criterion for ease of grafting, and Hypocotyl length was measured with a ruler at the time an excessively long or an excessively short hypocotyl is not of grafting. For cross-sectional area (mm2), pith cavity area preferred for grafting (Yıldız and Balkaya, 2016). (mm2), and vascular bundle measurements, a section of The largest hypocotyl cross-sectional areas were 0.25 mm in thickness was taken using a scalpel. Images of observed in the MO-4 (10.42 mm2), MO-8 (10.38 mm2), the cross-section were photographed using a Zeiss Axio MO-12 (10.20 mm2), and MO-20 (9.62 mm2) pumpkin Lab A1 microscope at 100× magnification. Then length rootstocks (Table 2). Eleven pumpkin genotypes (MO- and area values were determined using the Zen imaging 33, MO-26, MO-34, MO-13, MO-36, MO-21, MO-5, system (Pu et al., 2016). The pith cavity rate (%) was MO-37, MO-35, MO-30, MO-42) were found in the same determined by proportioning the pith cavity area to the statistical group in terms of thin hypocotyl (3.37–5.50 cross-sectional area of the hypocotyl. Graft compatibility mm2). The average size of the cross-sectional areas was and seedling visual evaluation were evaluated 20 days after 6.84 mm2. The rootstocks with the hypocotyl thickness grafting. The viable grafted seedling number was divided nearest that of cucumber (5.47 mm2) were the MO- by the grafted seedling number to determine the % graft 30 (5.50 mm2) and MO-42 (5.59 mm2) genotypes. The compatibility rate (Yetişir et al., 2007; Karaağaç, 2013). hypocotyl cross-sectional area rate between rootstock and Seedling quality was visually evaluated on a scale of 1 scion varied between 0.63 and 1.91. Yetişir and Sarı (2004) (very weak) to 5 (very good) by a research team composed determined that this value varied between 0.96 and 1.53 in of technical staff with expertise in commercial grafted watermelon rootstocks and reported that it did not affect seedling production. The research team also carried out a graft compatibility. However, since it facilitates grafting separate evaluation based on the rooting level and general and allows better callus bridge formation by minimizing appearance of the grafted seedlings. Rooting levels of the slipping on the dissection surface, it is advantageous to scions in the grafted seedlings were categorized as high, have similar thickness values. intermediate, few, and absent by visually evaluating both In the pumpkin rootstocks, the lowest pith cavity the inner part using the vertical section dissected from the areas were measured in the MO-34 (0.59 mm2), MO-36 2 2 2 hypocotyl and the outer part. (0.85 mm ), MO-22 (0.88 mm ), Gordion F1 (0.92 mm ), 2 2.3. Data analysis and MO-35 (0.94 mm ) genotypes, while the highest The results of the study were first tested for normality stem cavity areas were measured in the MO-23 (4.14 2 to determine their conformity to a normal distribution. mm ) genotype (Table 2). The greatest variation among Variance analysis was carried out using the SAS-JMP 5.01 pumpkin rootstock genotypes was determined in pith statistical package program to determine the statistical cavity rate, which varied between 10.3% (MO-22) and significance of the investigated criteria and differences 65.2% (MO-23) (Table 2; Figure 1). The results showed among genotypes. Angle transformation was applied significant differences among the genotypes with similar for pith cavity rate and graft compatibility rate. The data hypocotyl thicknesses in terms of cavity rate (Figure 1). were also evaluated using the weighted-ranking method For example, although the hypocotyl thicknesses of the (Balkaya and Yanmaz, 1999; Balkaya et al., 2008). Class MO-28 and MO-23 genotypes were the same, their pith values of the selection criteria, class scores, and relative cavity values were 18.3% and 65.2%, respectively (Table 2). scores were assigned. The total points of the genotypes Yetişir and Sarı (2004), Edelstein et al. (2004), and Yıldız were calculated by summing the class scores multiplied by and Balkaya (2016) reported that there was no relationship the relative scores. Moreover, correlation analysis was used between hypocotyl thickness and grafting success. This is to determine whether there was a statistical relationship attributable to the differences in pith cavity area, which was between the investigated physical characteristics. not investigated in those studies. In this study, the MO-22 and MO-20 genotypes stood out with their low pith cavity 3. Results and discussion rates. In general, grafting success and seedling quality were The statistical significance level of the differences among higher in rootstocks with low pith cavity rates. the pumpkin rootstock genotypes was 1% in terms of all Of the genotypes tested, 78% had 6 vascular bundles investigated properties. Hypocotyl lengths of the pumpkin (Table 3). The number of vascular bundles in 11 pumpkin rootstocks varied between 8 and 12. All small-fruited rootstocks varied between 3.06 cm (Kitora F1) and 4.78 cm

127 KARAAĞAÇ et al. / Turk J Agric For

Table 2. Hypocotyl properties of the genotypes.

Genotype Hypocotyl length (cm) Hypocotyl cross-sectional area (mm2) Pith cavity area (mm2) Pith cavity rate (%) MO-1 4.19 c–g 9.11 b–d 2.50 ef 27.73 i–o MO-2 4.39 b–e 8.64 b–h 3.25 c 38.11 d–h MO-3 3.59 j–p 6.16 i–p 1.34 r–u 22.68 l–r MO-4 4.24 c–g 10.42 a 1.50 o–t 14.74 q–s MO-5 3.60 i–o 5.15 m–r 2.30 e–g 44.60 c–e MO-6 3.29 l–s 6.54 f–o 1.49 o–t 23.29 l–q MO-7 3.39 k–s 6.56 f–o 1.82 j–p 28.16 h–o MO-8 4.28 b–f 10.38 a 1.94 h–m 19.07 n–s MO-9 4.11 d–h 7.03 c–m 1.48 p–t 21.80 l–r MO-10 3.34 l–s 7.10 c–m 1.81 j–p 25.54 k–p MO-11 3.74 h–k 8.62 b–h 2.19 f–i 26.73 j–p MO-12 3.97 f–i 10.20 a 2.00 g–k 19.58 m–s MO-13 3.46 k–q 4.50 o–r 1.82 j–p 41.11 d–f MO-14 4.15 c–g 7.64 b–k 1.96 g–l 26.00 k–p MO-15 3.59 i–p 6.82 d–n 1.56 n–s 23.00 l–q MO-16 4.20 c–g 8.96 b–e 2.28 e–g 25.81 k–p MO-17 4.24 c–g 8.30 b–j 1.74 j–q 21.20 l–r MO-18 3.65 i–l 6.61 f–o 1.10 u–w 17.54 p–s MO-19 3.26 m–s 8.84 b–f 1.61 m–s 18.40 o–s MO-20 3.23 o–s 9.62 ab 1.16 t–w 12.77 rs MO-21 3.22 p–s 5.04 m–r 0.95 vw 18.86 n–s MO-22 4.48 a–d 9.30 b–c 0.88 wx 10.33 s MO-23 4.52 a–c 6.44 h–o 4.14 a 65.15 a MO-24 3.32 l–s 6.69 e–o 1.57 n–s 29.40 g–m MO-25 3.62 i–m 7.26 c–m 3.90 ab 53.89 bc MO-26 3.42 k–s 3.77 qr 2.06 g–j 55.83 ab MO-27 3.13 q–s 5.87 k–q 1.71 j–q 29.45 g–m MO-28 3.45 k–r 6.44 h–o 1.17 t–w 18.30 o–s MO-29 4.08 e–h 8.39 b–i 1.87 i–n 23.10 l–q MO-30 4.37 b–e 5.50 k–r 1.62 l–s 30.23 g–l MO-31 3.61 i–n 6.05 j–q 2.01 g–k 35.07 e–k MO-32 4.22 c–g 6.48 g–o 3.61 b 55.93 ab MO-33 3.56 j–p 3.47 r 0.96 vw 27.59 i–p MO-34 3.24 n–s 4.10 p–r 0.59 x 14.63 q–s MO-35 4.64 ab 5.26 l–r 0.94 wx 18.33 o–s MO-36 4.78 a 4.68 n–r 0.85 wx 18.37 o–s MO-37 3.64 i–l 5.18 m–r 1.48 o–t 28.92 h–n MO-38 3.08 rs 7.55 b–l 1.44 q–u 19.70 m–s MO-39 3.40 k–s 6.03 j–q 1.52 n–s 25.82 k–p MO-40 3.08 rs 5.95 k–q 1.93 h–m 34.22 f–k MO-41 4.34 b–e 8.31 b–j 1.51 o–t 19.10 n–s MO-42 3.61 i–n 5.59 k–r 2.61 de 47.66 b–d

Early-type F1 3.12 q–s 5.95 k–q 1.66 k–r 28.18 h–o

Kitora F1 3.06 s 4.54 n–r 1.83 j–o 40.49 d–f

Bodyguard F1 3.32 l–s 5.86 k–q 1.29 s–v 22.53 l–r

Ultra F1 3.40 k–s 6.79 d–o 2.45 e–f 36.19 e–j

S. Shintosa F1 3.87 g–j 8.79 b–g 3.22 c 37.43 e–i

TZ 148 F1 3.12 q–s 7.34 b–m 2.87 d 39.27 d–g

Gordion F1 3.03 s 5.47 k–r 0.92 wx 16.80 p–s Average 3.71 ± 0.49 6.84 ± 1.74 1.85 ± 0.78 28.5 ± 12.1 CV (%) 5.3 13.4 10.7 12.8 P <0.01 <0.01 <0.01 <0.01

128 KARAAĞAÇ et al. / Turk J Agric For

Figure 1. Images of the hypocotyl pith cavity of some genotypes (top- promising genotypes; bottom left and bottom middle- genotypes with the highest pith cavity rate; bottom right- genotype with the lowest pith cavity rate). rootstock genotypes in the Calabaza and Butternut group No scion rooting was observed in 12 rootstocks, while had 6 vascular bundles. Vascular bundle numbers of the remaining rootstocks showed moderate or high levels most of the large-fruited pumpkin rootstock genotypes of rooting (Table 3). It was determined that larger pith were higher than the average of 6. The average vascular cavity volume had a negative effect on the establishment of bundle number of all investigated C. moschata rootstock connections between the callus bridge and vascular bundle, genotypes was 6.77. Similarly, Oda et al. (1994) reported resulting in grafting failure. A review of the relevant literature that C. moschata rootstock cultivars (Unryuh and Kongo) showed that there was no study on the effects of pith cavity on had 6 vascular bundles. Yetişir and Sarı (2004) reported grafting properties. However, Tiedemann (1989) and Traka- that the local pumpkin cultivar had 8.1 vascular bundles Mavrona et al. (2000) reported that pith cavity may have a and that the interspecies hybrid Cucurbita rootstock negative effect on graft compatibility. In the graft union, the cultivars had 6.4 vascular bundles. Edelstein et al. (2004) scion surface transforms into stem cells and forms outward reported that this value varied between 6 and 10.6 in roots from within the rootstock through the unfilled pith Cucurbita spp. rootstocks. The results showed that there cavity (Figure 2). This also increases graft incompatibility was no relationship between vascular bundle number and and facilitates root elongation from the outer parts of the graft success rate, which agrees with previous studies (Oda scion. One of the higher pith cavity rates was determined in et al., 1993; Edelstein et al., 2004; Yetişir and Sarı, 2004). the MO-32 genotype, which also showed both external and Graft success rates varied between 56% and 100% internal scion rooting (Figure 2). Reexamination of the graft in pumpkin genotypes (Table 3). According to grafting union on this genotype 20 days after grafting showed that its trials, a 100% success rate was obtained in 25 rootstock pith cavity was poorly filled with callus tissue and there was genotypes including the commercial rootstocks. The a 1.33-mm-wide cavity on the union (Figure 2). lowest success rate (56%) was recorded for the MO-10 Seedling visual evaluations showed that the scale values and MO-13 pumpkin genotypes. The graft success rates in of the grafted cucumber combinations varied between 1.5 cucumber for different rootstock scion combinations were and 5.0 (Table 3). In addition to the TZ-148 and Early type between 50% and 87% in a study by Yıldız and Balkaya control genotypes, the MO-4, MO-6, MO-21, MO-22, (2016) and between 78% and 100% in a study by Cansev MO-31, MO-37, and MO-38 rootstock genotypes received and Ozgur (2010). the highest scores in terms of seedling quality.

129 KARAAĞAÇ et al. / Turk J Agric For

Table 3. Hypocotyl properties of the genotypes and graft success rates.

Genotype Number of vascular bundles Grafting success rate (%) Visual based seedling quality (1–5) Scion rooting status MO-1 6 e 83.3 a–d 3.4 d–f Absent MO-2 6 e 91.7 ab 2.2 h–j High MO-3 6 e 73.3 c–e 3.0 e–g High MO-4 11 b 100.0 a 5.0 a Absent MO-5 6 e 72.33 c–f 1.8 jk Intermediate MO-6 6 e 100.0 a 5.0 a High MO-7 6 e 83.3 a–d 1.9 i–k Intermediate MO-8 6 e 79.9 b–d 2.3 g–j Intermediate MO-9 6 e 100.0 a 2.8 f–h Few MO-10 6 e 55.7 f 3.7 c–e Intermediate MO-11 9 d 89.0 a–c 2.6 g–i Absent MO-12 8 e 100.0 a 3.0 e–g Absent MO-13 6 e 55.7 f 1.5 k High MO-14 6 e 72.3 c–f 2.1 h–k High MO-15 6 e 89.0 a–c 4.1 b–d Absent MO-16 6 e 67.0 d–f 1.9 i–k Intermediate MO-17 9 d 100.0 a 2.6 g–i Intermediate MO-18 6 e 100.0 a 3.9 b–d High MO-19 9 d 100.0 a 3.0 e–g High MO-20 11 b 100.0 a 2.8 f–h Few MO-21 6 e 100.0 a 4.3 a–c Absent MO-22 6 e 100.0 a 5.0 a Few MO-23 6 e 69.3 d–f 1.5 k High MO-24 8 e 100.0 a 3.1 e–g Absent MO-25 6 e 100.0 a 1.9 i–k High MO-26 6 e 100.0 a 2.8 f–h Intermediate MO-27 12 a 69.7 d–f 1.5 k Intermediate MO-28 6 e 100.0 a 3.0 e–g Few MO-29 10 c 100.0 a 3.0 e–g Absent MO-30 6 e 91.7 ab 2.3 g–j Intermediate MO-31 6 e 100.0 a 4.5 ab Intermediate MO-32 9 d 61.3 ef 1.9 i–k Intermediate MO-33 6 e 91.7 ab 2.3 g–j High MO-34 6 e 61.3 ef 2.1 h–k Intermediate MO-35 6 e 78.0 b–e 2.1 h–k Intermediate MO-36 6 e 100.0 a 3.5 d–f Intermediate MO-37 6 e 100.0 a 4.5 ab Absent MO-38 6 e 100.0 a 5.0 a Absent MO-39 8 e 69.7 d–f 2.3 g–j Intermediate MO-40 6 e 91.7 ab 2.6 g–i Intermediate MO-41 6 e 89.0 a–c 3.7 c–e Intermediate MO-42 6 e 75.7 b–e 1.9 i–k Intermediate

Early-type F1 6 e 100.0 a 5.0 a Absent

Kitora F1 6 e 100.0 a 3.5 cd High

Bodyguard F1 6 e 100.0 a 3.0 de Absent

Ultra F1 6 e 100.0 a 4.0 b–d High

S. Shintosa F1 6 e 100.0 a 4.0 b–d High

TZ 148 F1 6 e 100.0 a 5.0 a Intermediate

Gordion F1 6 e 2.7 f–h Average 6.77 ± 1.57 89.4 ± 14.0 3.1 ± 1.1 CV (%) 5.9 11.7 14.9 P <0.01 <0.01 <0.01

130 KARAAĞAÇ et al. / Turk J Agric For

Figure 2. Rooting of the scion in the MO-32 genotype (top left) and high-quality seedling of MO-21 genotype (top right) 20 days after grafting and view from the pith cavity in the graft union of these genotypes (bottom).

The statistical significance of the relationships Analysis of variance was applied to 42 inbred lines from between the investigated properties was determined the cucumber rootstock breeding program considering using correlation analysis (Table 4). There was a positive hypocotyl morphology, seedling visual quality, scion correlation between hypocotyl length and hypocotyl cross- rooting tendency, and graft success rate, and a criteria-based sectional area (r: 0.64, P < 0.01) (Table 4). However, there interpretation of the genotypes was performed. However, was no statistically significant relationship between graft the rankings of most genotypes varied depending on the success rate and rooting and hypocotyl length. There was investigated properties; therefore, the weighted ranking also no distinct relationship between the hypocotyl cross- method was employed to select the promising pumpkin sectional area and graft success rate and scion rooting. Oda genotypes. The selection criteria and class and relative et al. (1993) and Farhadi et al. (2016) reported that there scores are given in Table 5. The total scores of the genotypes was a negative correlation between differences in hypocotyl varied between 120 and 460 (Table 6). The average score thickness and graft compatibility rates in cucumber/squash of the control rootstocks was 360. In terms of hypocotyl rootstock combinations, while Traka-Mavrona et al. (2000) properties, pumpkin genotypes performed better compared reported that similar hypocotyl thickness increased the to interspecies hybrid pumpkin rootstocks. The MO-38 success rate of grafting. Yetişir and Sarı (2004), Edelstein (460), MO-21(450), and MO-37 (440) pumpkin rootstocks et al. (2004), and Yıldız and Balkaya (2016) reported that outranked the Early type F1 (420) and Bodyguard F1 (420) there was no significant relationship between hypocotyl cultivars in the control group, which received the highest length and graft success rate. These differences among scores among the control group. These rootstocks are the studies are attributable to the differences in cultivars, within the desired limits in terms of hypocotyl properties, grafting methods, and environmental conditions after graft compatibility, and scion rooting tendency. According grafting. There was a positive correlation between graft to weighted ranking scores, the 12 pumpkin genotypes success rate and seedling strength. One of the most that scored 400 or higher were selected as promising important factors negatively affecting graft success rate was rootstocks candidates (Table 6). The results collectively led scion rooting (Table 4). It was also determined that scion to the conclusion that with good graft compatibility, a low rooting depended on the pith cavity rate rather than the permissiveness to scion rooting, a low pith cavity area and pith cavity area. Scion rooting decreased with a decreasing rate, and high scores in visual based seedling quality, the pith cavity rate. The results showed that seedling quality MO-38, MO-21, MO-37, MO-4, MO-9, MO-22, and MO- was higher in rootstock/scion combinations with a low 28 pumpkin genotypes were the most suitable parents for pith cavity rate (r: 0.56, P < 0.05) and no scion rooting (r: hybridization programs aiming to develop hybrid cultivars –0.59, P < 0.01). for rootstock breeding programs.

131 KARAAĞAÇ et al. / Turk J Agric For

Table 4. Correlation values between the investigated properties and their significance levels.

VB HL PCA HCSA PCR GSQ SRS

HL –0.021

PCA –0.031 0.370

HCSA 0.309 0.635** 0.182

PCR –0.164 –0.006 0.779** –0.310

GSQ –0.102 –0.128 –0.077 0.145 –0.144

SRS 0.244 –0.013 –0.160 0.101 0.560* –0.591**

GSR 0.005 –0.020 –0.036 0.144 –0.093 0.610** –0.487*

VB: Vascular bundles; HL: hypocotyl length (cm), PCA: pith cavity area (mm2), HCSA: hypocotyl cross-sectional area (mm2), PCR: pith cavity rate (%), GSQ: grafted seedling quality, SRS: scion rooting status, GSR: grafting success rate (%). **: Significance ≤ 0.01, *: significance ≤ 0.05.

Table 5. Scores given to characteristics based on evaluations.

Selection criteria Class value Class scores Relative scores

4.21–4.78 (long) 1

Hypocotyl length (cm) 3.63–4.21 (medium) 5 10

3.06–3.63 (short) 3

2.96–4.14 1

Pith cavity area (mm2) 1.77–2.95 3 10

0.59–1.76 5

8.10–10.42 1

Hypocotyl cross-section area (mm2) 5.78–8.09 5 10

3.47–5.77 2

46.88–65.15 (much) 1

Pith cavity rate (%) 28.61–46.87 (intermediate) 3 15

10.33–28.60 (little) 5

<70 1

Grafting success rate (%) 71–90 3 20

91–100 5

1.45–2.15 (very weak) 1

2.16–2.86 (weak) 2

Grafted seedling quality (1–5) 2.87–3.57 (medium) 3 20

3.58–4.28 (good) 4

4.29–5.00 (very good) 5

High 1

Medium 3 Scion rooting status 15 Few 4

Absent 5

132 KARAAĞAÇ et al. / Turk J Agric For

Table 6. Weighted ranking points of pumpkin rootstocks (types that scored 400 or higher are shown in bold).

Genotype HL PCA HCSA PCR GSR GSQ SRS Total MO-38 30 50 30 75 100 100 75 460 MO-21 30 50 20 75 100 100 75 450 MO-37 50 50 20 45 100 100 75 440 MO-4 10 50 10 75 100 100 75 420 MO-9 50 50 30 75 100 40 60 405 MO-22 10 50 10 75 100 100 60 405 MO-28 30 50 30 75 100 60 60 405 MO-6 30 50 30 75 100 100 15 400 MO-12 50 30 10 75 100 60 75 400 MO-15 30 50 30 75 60 80 75 400 MO-18 50 50 30 75 100 80 15 400 MO-29 50 30 10 75 100 60 75 400 MO-31 30 30 30 45 100 100 45 380 MO-20 30 50 10 75 100 40 60 365 MO-1 50 30 10 75 60 60 75 360 MO-36 10 50 20 75 100 60 45 360 MO-24 30 50 30 45 100 20 75 350 MO-11 50 30 10 75 60 40 75 340 MO-19 30 50 10 75 100 60 15 340 MO-17 10 50 10 75 100 40 45 330 MO-33 30 50 20 75 100 40 15 330 MO-41 10 50 10 75 60 80 45 330 MO-3 30 50 30 75 60 60 15 320 MO-40 30 30 30 45 100 40 45 320 MO-10 30 30 30 75 20 80 45 310 MO-30 10 50 20 45 100 40 45 310 MO-7 30 30 30 75 60 20 45 290 MO-39 30 50 30 75 20 40 45 290 MO-14 50 30 30 75 60 20 15 280 MO-26 30 30 20 15 100 40 45 280 MO-35 10 50 20 75 60 20 45 280 MO-8 10 30 10 75 60 40 45 270 MO-34 30 50 20 75 20 20 45 260 MO-5 30 30 20 45 60 20 45 250 MO-16 50 30 10 75 20 20 45 250 MO-27 30 50 30 45 20 20 45 240 MO-2 10 10 10 45 100 40 15 230 MO-25 30 10 30 15 100 20 15 220 MO-42 30 30 20 15 60 20 45 220 MO-13 30 30 20 45 20 20 15 180 MO-32 10 10 30 15 20 20 45 150 MO-23 10 10 30 15 20 20 15 120

Early type F1 30 50 30 75 100 60 75 420

Bodyguard F1 30 50 30 75 100 60 75 420

TZ 148 F1 30 30 30 45 100 100 45 380

Ultra F1 30 30 30 45 100 80 15 330

S. Shintosa F1 50 10 10 45 100 80 15 310

Kitora F1 30 30 20 45 100 60 15 300 HL: Hypocotyl length (cm), PCA: pith cavity area (mm2), HCSA: hypocotyl cross-section area (mm2), PCR: pith cavity rate (%), GSR: grafting success rate (%), GSQ: grafted seedling quality, SRS: scion rooting status.

133 KARAAĞAÇ et al. / Turk J Agric For

In conclusion, the cost of grafted seedling production revealed that the pith cavity properties of genotypes should is 3- to 4-fold higher than the cost of nongrafted seedling also be investigated in rootstock breeding programs. production. To reduce the costs, grafted seedling Pumpkin genotypes for the cucumber rootstock breeding producers should achieve higher graft success rates and program were selected based on the weighted ranking produce higher-quality seedlings. Although the factors method, which considered the hypocotyl and pith cavity causing graft incompatibility are not yet fully elucidated, properties, graft compatibility, and seedling quality criteria. there are numerous fundamental studies investigating According to the results, the MO-38, MO-21, MO-37, MO- the hormonal, anatomical, and physiological properties. 4, MO-9, MO-22, and MO-28 genotypes were selected as However, there are no morphological indicators for the promising pumpkin parents for hybrid rootstock breeding effective evaluation and selection of the currently available programs due to their high graft compatibility and lack of genotypes to provide high graft compatibility and seedling rooting problems. Thus, the first stage of developing local quality in rootstock breeding programs. The selection hybrid cucumber rootstocks in Turkey was accomplished. criteria proposed in this study can greatly contribute to the literature concerning grafted cucumber production. Acknowledgment This study showed that there was a statistically We gratefully acknowledge the support of the Scientific significant and negative relationship between high graft and Technological Research Council of Turkey (TÜBİTAK compatibility and high-quality seedling production and Project No: TEYDEB 311O194). pith cavity area/hypocotyl cross-sectional area. The study

References

Aloni B, Karni L, Deventurero G, Levin Z, Cohen R, Katzir N, Lotan- Cansev A, Ozgur M (2010). Grafting cucumber seedlings on Pompan M, Edelstein M, Aktas H, Turhan E et al. (2008). Cucurbita spp.: comparison of different grafting methods, Physiological and biochemical changes at the rootstock-scion scions and their performance. J Food Agric Environ 8: 804- interface in graft combinations between Cucurbita rootstocks 809. and a melon scion. J Hortic Sci Biotech 83: 777-783. Cousins P (2005). Rootstock breeding: an analysis of intractability. Aloni B, Cohen R, Karni L, Aktas H, Edelstein M (2010). Hormonal HortScience 40: 1945-1946. signaling in rootstock–scion interactions. Sci Hortic 127: 119- Davis AR, Perkins-Veazie P, Sakata Y, López-Galarza S, Maroto 126. JV, Lee SG, Huh YC, Sun Z, Miguel A, King SR et al. (2008). Balkaya A (2014). Aşılı sebze üretiminde kullanılan anaçlar. Cucurbit grafting. Crit Rev Plant Sci 27: 50-74. TÜRKTOB Dergisi 3: 4-7 (in Turkish). Diánez F, Díaz M, Santos M, Huitrón V, Ricárdez M, Camacho F Balkaya A, Özbakır M, Karaağaç O (2010). Karadeniz Bölgesi’nden (2007). The use of grafting in Spain. In: Workshop on Non- toplanan bal kabağı (Cucurbita moschata Duch.) Chemical Alternatives to Replace Methyl Bromide as a Soil populasyonlarındaki meyve özelliklerinin karakterizasyonu ve Fumigant, 26–28 June, Budapest, Hungary, pp. 87-97. varyasyonun değerlendirilmesi. Tarım Bilimleri Dergisi 16: 17- Edelstein M, Burger Y, Horev C, Porat A, Meir A, Cohen R (2004). 25 (in Turkish). Assessing the effect of genetic and anatomic variation of Balkaya A, Yanmaz R (1999). Karadeniz Bölgesi taze fasulye Cucurbita rootstocks on vigour, survival and yield of grafted (Phaseolus vulgaris L.) populasyonlarından teksel seleksiyon melons. J Hortic Sci Biotech 79: 370-374. yoluyla geliştirilen çeşit adayları. In: III. Ulusal Bahçe Bitkileri Farhadi A, Aroeii H, Nemati H, Salehi R, Giuffrida F (2016). The Kongresi, 14–17 September, Ankara, Turkey, pp. 504-508 (in effectiveness of different rootstocks for improving yield Turkish). and growth of cucumber cultivated hydroponically in a Balliu A, Sallaku G (2017). Exogenous auxin improves root greenhouse. Horticulturae 2: 1-7. morphology and restores growth of grafted cucumber Göçmen M, Balkaya A, Kurtar ES, Karaağaç O (2014). Kabak seedlings. Hort Sci 44: 82-90. (Cucurbita spp.) Genetik Kaynaklarının Hıyar (Cucumis Bekar NK, Balkaya A, Göçmen M (2017). Kabak anaçlarının aşılı sativus L.) Anaç Islah Programında Değerlendirilmesi ve Yerli hıyar yetiştiriciliğinde vejetatif büyüme üzerine etkilerinin Hibrit Anaçların Geliştirilmesi. TÜBİTAK-TEYDEB Proje No: belirlenmesi. Selçuk Tarım Bilimleri Dergisi 3: 280-290 (in 3110194. Ankara, Turkey: TÜBİTAK (in Turkish). Turkish). Huang Y, Kong QS, Chen F, Bie ZL (2014). The history, current status Bie Z, Nawaz MA, Huang Y, Lee JM, Colla G (2017). Introduction to and future prospects of vegetable grafting in China. In: 1st vegetable grafting. In: Colla G, Alfocea FP, Schwarz D, editors. International Symposium on Vegetable Grafting, 17–21 March, Vegetable Grafting: Principles and Practices. Wallingford, UK: Hubei, China, pp. 31-39. CABI Publishing, pp. 1-21.

134 KARAAĞAÇ et al. / Turk J Agric For

Karaağaç O (2013). Karadeniz Bölgesi’nden toplanan kestane kabağı Pina A, Errea P (2005). A review of new advances in mechanism of (C. maxima Duchesne) ve bal kabağı (C. moschata Duchesne) graft compatibility- incompatibility. Sci Hortic 106: 1-11. genotiplerinin karpuza anaçlık potansiyellerinin belirlenmesi. Pu Y, Diane C, Bassham DC (2016). Detection of autophagy in PhD, Ondokuz Mayıs University, Samsun, Turkey (in Turkish). plants by fluorescence microscopy. In: Lois LM, Matthiesen Karaağaç O, Balkaya A (2013). Interspecific hybridization and R, editors. Plant Proteostasis: Methods and Protocols. Berlin, hybrid seed yield of winter squash (Cucurbita maxima Duch.) Germany: Springer, pp. 161-172. and pumpkin (Cucurbita moschata Duch.) lines for rootstock Sakata Y, Horie H, Ohara T, Kawasaki Y, Sugiyama M (2008a). breeding. Sci Hortic 149: 9-12. Influences of rootstock cultivar and storage on the texture of King SR, Davis AR, Zhang X, Crosby K (2010). Genetics, breeding cucumber fruits. J Jpn Soc Hortic Sci 77: 47-53. and selection of rootstocks for Solanaceae and Cucurbitaceae. Sakata Y, Ohara T, Sugiyama M (2007). The history and present state Sci Hortic 127: 106-111. of the grafting of Cucurbitaceous vegetables in Japan. Acta Lee JM, Kubota C, Tsao SJ, Bie Z, Hoyos Echevarria P, Morra L, Hortic 731: 159-170. Oda M (2010). Current status of vegetable grafting: diffusion, Sakata Y, Ohara T, Sugiyama M (2008b). The history of melon and grafting techniques, automation. Sci Hortic 127: 93-105. cucumber grafting in Japan. Acta Hortic 767: 217-228. Lee JM, Oda M (2010). Grafting of herbaceous vegetable and Schwarz D, Rouphael Y, Colla G, Venema JH (2010). Grafting as ornamental crops. Hortic Rev 28: 61-124. a tool to improve tolerance of vegetables to abiotic stresses: Louws FJ, Rivard CL, Kubota C (2010). Grafting fruiting vegetables thermal stress, water stress and organic pollutants. Sci Hortic to manage soilborne pathogens, foliar pathogens, arthropods 127: 162-171. and weeds. Sci Hortic 127: 127-146. Tiedemann R (1989). Graft union development and symplastic Martínez-Ballesta MC, Alcaraz-López C, Muries B, Mota-Cadenas phloem contact in the heterograftCucumis sativus on Cucurbita C, Carvajal M (2010). Physiological aspects of rootstock–scion ficifolia. J Plant Physiol 134: 427-440. interactions. Sci Hortic 127: 112-118. Traka-Mavrona E, Koutsika-Sotiriou M, Pritsa T (2000). Response of Marukawa S, Takatsu I (1969). Studies on the selection of Cucurbita squash (Cucurbita spp.) as rootstock for melon (Cucumis melo spp. as cucumber stock. 1. Compatibility, ability to tolerate low- L.). Sci Hortic 83: 353-362. temperature conditions and yield of black prickly cucumber. Yetişir H (2017). History and current status of grafted vegetables in Bulletin of the Ibaraki Horticultural Experiment Station 3: 11- Turkey. Chron Horticult 57: 13-18. 18. Yetişir H, Kurt Ş, Sarı N, Tok FM (2007). Rootstock potential of Oda M (2007). Vegetable seedling grafting in Japan. Acta Hortic 759: Turkish Lagenaria siceraria germplasm for watermelon: plant 175-180. growth, graft compatibility, and resistance to Fusarium. Turk J Oda M, Dosai M, Ikeda H, Furukowa H (2001). Causes of Low Agric For 31: 381-388. Survival in Cucumber (Cucumis sativus L.) Plants Grafted Yetişir H, Sarı N (2004). Effect of hypocotyl morphology on onto Pumpkin (Cucurbita moschata Duch.) Rootstocks by survival rate and growth of watermelon seedlings grafted on Horizontal-Cut Grafting at the Center of the Hypocotyl. rootstocks with different emergence performance at various Scientific Report of the Graduate School of Agriculture and temperatures. Turk J Agric For 28: 231-237. Biological Sciences. Osaka, Japan: Osaka Prefecture University. Yıldız S, Balkaya A (2016). Tuza tolerant kabak anaçlarının hipokotil Oda M, Tsuji K, Ichmura K, Sasaki H (1994). Factors affecting the özellikleri ve hıyarla aşı uyuşum durumlarının belirlenmesi. survival of cucumber plants grafted on pumpkin plants by Yüzüncü Yıl Üniversitesi Tarım Bilimleri Dergisi 26: 538-546 horizontal grafting at the hypocotyl level. Bulletin of the (in Turkish). National Research Institute of Vegetables, Ornamental Plants and Tea 9: 51-60. Oda M, Tsuji K, Sasaki H (1993). Effect of hypocotyl morphology on survival rate and growth of cucumber seedlings grafted on Cucurbita spp. Jpn Agric Res Q 26: 259-263.

135