Improved Induction of Somatic Embryo in Watermelon [Citrullus lanatus (Thunb.) Matsum. et Nakai]

In order to optimize somatic embryo induction of watermelon, three different explants from three different varieties were used to induce somatic embryo in medium contained BA in combination with either NAA or 2,4-D. ‘Kangbingjingxin’ cotyledon explant with all of PGR combination treatments has a high callus inductivity (75.6%-98.9%), but inductivity of somatic embryo was varied among different PGR combination, the highest inductivity (54.7%) was obtained in medium with BA 3.0 mg/L and NAA 0.05 mg/L. Somatic embryogenesis was different among ‘Kangbingjingxin’, ‘Zaojia 8424’ and ‘Heimeiren’, the inductivity was 46.2%, 34.1% and 15.7% respectively. Callus inductivity of three explants was 88.9%-98.9%, in which callus induced from cotyledon and cotyledon node was mostly embryonic callus while callus from hypocotyl was non-embryonic callus. Somatic embryo inductivity of cotyledon node (62.0%) was significantly higher than cotyledon (46.1%), there is no somatic embryo induced from hypocotyl. However, somatic embryos number per explant of cotyledon node (2.3) was significantly lower than cotyledon (6.0).


Introduction
Watermelon [Citrullus lanatus (Thunb.)Matsum.et Nakai] is an important fruit crop in Cucurbitaceae family and it is widely planted as an economically important crop (Yang, 1991;Zhang et al., 1996).Conventional breeding of watermelon has the shortage such as long breeding cycle, great difficulty and genetic instability, while plant regeneration via organogenesis also has the defects of low frequency in regeneration and huge difference in gene expression.Somatic embryo could be mass-produced in a short time, the reproduction rate of somatic embryo is high, and it's easy to develop into plantlet (Song et al., 2004).Somatic embryo provides perfect material for artificial seed, haploid breeding and asexual propagation.Moreover, somatic embryo has little somaclonal variation while has high capability to accept exogenous DNA, so it is used as receptor system of genetic modification in many plant species (Yan, 2003).
Somatic embryogenesis in watermelon was first reported by Compton and Gray (1993), cotyledon from immature embryos was used as explants, and the highest induction rate was 30% while the PGR combination was 10 μM 2,4-D and 0.5 μM TDZ.After that Zhang (2004), Niu (2006) and Song et al. (2007) tried to induce somatic embryo using mature cotyledon and hypocotyl et al as explants.However, both the induction rate of somatic embryo and seedling regeneration rate were low, the quality of somatic embryo and the amount of somatic embryos per explant were as well.
In this study, three different explants from three different varieties were used to induce somatic embryo, while the PGR combination was researched.germinated somatic embryos appeared on some explants.The third non-embryonic callus was mostly induced by BA and 2,4-D (Figure 3 b).The quality of callus and inductivity of somatic embryo decreased as the increase of 2,4-D concentration.All of the callus induced is non-embryonic callus when 2,4-D concentration is higher than 1.0 mg/L.The cotyledon explant without PGR just slightly rolled and enlarged, there isn't callus on them while there is adventitious root on some explants.This reflects PGR is necessary for induction of callus and somatic embryo, which was the same as the report on melon (Yuan, 2007).
The inductivity of callus was high in all PGR combinations, the lowest inductivity is 75.6% when BA 4.0 mg/L with NAA 0.05 mg/L (Table 1).The highest inductivity of somatic embryo (54.7%) was gained in medium supplemented with BA 3.0 mg/L and NAA 0.05 mg/L, the second (45.6%) with BA 3.0 mg/L and NAA 0.1 mg/L, these two treatments were significantly higher than others.The inductivity of somatic embryo and number of somatic embryos per explant were first increased and then decreased as the increase of BA concentration, the inductivity of somatic embryo gained the highest when the concentration of BA is 3.0 mg/L, while the number of somatic embryos per explant gained the highest (7.8) in the same concentration.These show that 3.0 mg/L was the best BA concentration for induction of somatic embryo.The inductivity of somatic embryo and number of embryos per explant were decreased as the increase of NAA concentration, 0.05 mg/L was the best NAA concentration for induction of somatic embryo.When the concentration of NAA added to 1.0 mg/L, callus induced was non-embryonic callus which couldn't produce somatic embryo, and a large number of adventitious roots occurred on it (Figure 3 c).Inductivity of callus was slightly decreased as the increase of BA concentration and was increased as the increase of 2,4-D concentration (Table 2).It shows that low concentration BA in combination with high concentration 2,4-D was good for callus induction.When the concentration of BA increased, the inductivity of somatic embryo increased but there is no significant difference, while the number of somatic embryos per explant decreased and there is significant difference between low and high concentration.The inductivity of somatic embryo and number of somatic embryos per explant decreased as the increase of 2,4-D concentration.The inductivity and number gained maximum (12.6% and 6.9) when 2,4-D concentration is 0.1 mg/L.When 2,4-D concentration reach 1.0 mg/L, most of induced callus was non-embryonic callus.There is no somatic embryo induced when 2,4-D concentration is 2.0 mg/L.It proves that high 2,4-D concentration was not advantageous to induce somatic embryo from cotyledon explant.Somatic embryogenesis is a complex procedure which was affected and regulated by multiple factors, and the composition of medium is the most important factor (Guis et al., 1998;Guan et al., 2009).The inductivity of somatic embryo obtained maximum in medium with BA 3.0 mg/L and NAA 0.1 mg/L.In the process of in vitro culture, some explants were browned, while there is abnormality and premature germination in somatic embryogenesis, these problems restricted the induction of high frequency and high quality somatic embryo.Data represented mean ± SE of three replicates.
Means having the same letter in a column were not significantly different by Least Significant Differences test (P = 0.05).

Effect of Genotypes on Induction of Callus and Somatic Embryo
The callus inductivity of 'Zaojia 8424' and 'Kangbingjingxin' was 94.4% and 88.9% respectively, there is no significant difference between these two varieties (Table 3).However, the explant enlargement degree, callus initiation time and callus inductivity of 'Heimeiren' was weaker than the previous two varieties significantly.Inductivity of somatic embryo among different genotypes has significant difference, the highest inductivity was gained from 'Kangbingjingxin' (46.2%) cotyledon, then 'Zaojia 8424' (34.1%) and 'Heimeiren' (15.7%).However, the number of somatic embryos per responding explant was similar for all the genotypes tested.
The different ability of somatic embryogenesis induction among different genotypes was reported in other Cucurbitaceae plants.Nadolska-Orczyk and Malepszy (1989) suggested a genetic determinism in the ability to regenerate somatic embryo derived plants from cucumber (Cucumis sativus L.) leaf explants.Carol et al. (1995) studied the somatic embryogenesis of six squash cultivars, all cotyledons produced somatic embryos after 11 to 17 weeks on induction medium.However, the optimal culture time and the rate of plant regeneration were significantly different between the six cultivars.Significant differences of somatic embryogenesis were also observed among different cultivars in melon (Toshiro et al., 1992).It proved that genotype was a important factor to affect somatic embryo induction.Most plants have potential to induce somatic embryo, but the sensitivity and difficulty of inducing is different; and the frequency of somatic embryogenesis is widely different as the difference of genotype in the same species.Means having the same letter in a column were not significantly different by Least Significant Differences test (P = 0.05).

Effect of Explants on Induction of Callus and Somatic Embryo
Callus inductivity of cotyledon node and hypocotyl was higher than cotyledon in MS medium with BA 3.0 mg/L and NAA 0.1 mg/L (Table 4).Callus induced from cotyledon and cotyledon node was mostly embryonic callus while callus from hypocotyl was non-embryonic callus which cannot induce somatic embryo (Figure 3 d-f).Somatic embryo inductivity of cotyledon node (62.0%) was significantly higher than cotyledon (46.1%) and hypocotyl (0%).However, somatic embryos number per explant of cotyledon node (2.3) was significantly lower than cotyledon (6.0).It pointed that explant type was an extremely important factor in somatic embryo induction.
Means having the same letter in a column were not significantly different by Least Significant Differences test (P = 0.05).

Table 1 .
Effect of BA and NAA combination on induction of callus and somatic embryo

Table 2 .
Effect of BA and 2,4-D combination on induction of callus and somatic embryo

Table 3 .
Effect of genotypes on induction of callus and somatic embryo

Table 4 .
Effect of explants on induction of callus and somatic embryo