Vigor for In Vitro Culture Traits in S. melongena   ×   S. aethiopicum Hybrids with Potential as Rootstocks for Eggplant

Hybrids of Solanum melongena and S. aethiopicum are of interest as rootstocks of eggplant, as they are highly vigorous and can incorporate resistance to several diseases. However, hybridization between both species is difficult. Therefore, protocols for in vitro culture are of great interest for their micropropagation and biotechnological breeding. We assessed the organogenesis response from leaf explants in four interspecific hybrids and in their parents testing two organogenic media: SIM-A, containing 6-benzylaminopurine and kinetin, and SIM-B, which contains thidiazuron. A higher regeneration capacity in the hybrids compared to their parents was observed. Whereas in interspecific hybrids and in one accession of S. melongena similar regeneration rates were observed for SIM-A and SIM-B, higher regeneration was found in the rest of genotypes when thidiazuron was used. Rooting ability in the interspecific hybrids was lower in in vitro micropropagated plants (35–60%) than in plants regenerated from explants (100%). The addition of indolbutiric acid (1 mg L−1) induced roots in nonrooted genotypes. In summary, we have adjusted in vitro culture conditions for regenerating and rooting S. melongena × S. aethiopicum hybrids. We have also demonstrated that these hybrids are heterotic for regeneration, which may be of interest for basic science studies.


Introduction
Common eggplant (Solanum melongena L.), also known as brinjal or aubergine, is an important vegetable crop widely consumed worldwide. Often, this crop displays insufficient levels of resistance to soil pests and diseases [1]. Therefore, the development of new rootstocks providing a combination of high levels of tolerance or resistance to soil stresses and high vigor can be useful for improving the yield and resilience of modern eggplant cultivars [2]. Interspecific hybrids between S. melongena and related species S. incanum L. or S. aethiopicum L. are highly vigorous [3,4] and confer productive advantages for eggplant production when used as eggplant rootstocks [2].
In S. aethiopicum, resistance to Fusarium oxysporum f. sp. melongenae, Ralstonia solanacearum, and root-knot nematodes has been reported [5][6][7][8][9][10][11]. Furthermore, contrarily to the wild S. incanum, which presents high levels of glycoalkaloids [12], S. aethiopicum is a cultivated species known as scarlet eggplant [13] that presents low levels of -solasonine and -solamargine [14] and therefore presents no risk of translocation of these glycoalkaloids to the fruits. This is a very important issue, as translocation of alkaloids from the rootstock to the scion may produce undesirable results. For instance, eggplant fruits from plants grafted onto Datura inoxia P. Mill. accumulated scopolamine and atropine at levels high enough to cause poisoning [15]; nicotine was also obtained in tomato fruits from plants grafted onto Nicotiana tabacum L. [16]. Therefore, interspecific hybrids S. melongena × S. aethiopicum may be of interest, not only for increasing vigor of the scion and for conferring resistance to some important eggplant diseases, but also because they are safe from translocation of undesirable compounds from the rootstock to the scion and fruit.
Although interspecific hybrids between S. melongena and S. aethiopicum can be obtained by sexual hybridization, 2 The Scientific World Journal on many occasions, fruit set is low and fruits may be parthenocarpic or present very few seeds [7,17]. Therefore, it would be desirable to have protocols available for efficient micropropagation. Also, the induction of regeneration from explants cultured in vitro, which is required for application of biotechnological techniques such as the genetic transformation [18], is of great interest for further improvement of these hybrids. Genetic transformation has been used in order to improve rootstocks in different crops [19][20][21][22].
The induction of regeneration in eggplant is achieved via organogenesis [23][24][25][26][27][28] or embryogenesis [25,26,29,30] and some interesting traits related to abiotic [31,32] and biotic [33,34] stresses have been introduced by genetic transformation. Although eggplant tissues showed a high morphogenetic potential some drawbacks were found in different culture conditions, mainly, buds which fail to develop into shoots [23,24,27] and shoots which fail to develop roots [26,27]. Research in in vitro culture of both S. aethiopicum and interspecific hybrids between S. melongena and S. aethiopicum is scarce [7,35]. In S. aethiopicum, regeneration from cotyledons and leaf explants was described in Gisbert et al. [35] and, as what occurred in other species, it was genotype dependent. On the other hand, Collonnier et al. [7] reported regeneration of plants from calli resulting from protoplast fusions between S. melongena and S. aethiopicum.
The aim of this study is to evaluate the organogenic response of four S. melongena × S. aethiopicum hybrids which present potential interest as rootstocks, as well as their ability for rooting, which is a prerequisite for micropropagation. It was also to compare their organogenic response with that of their parents. For organogenesis induction, we have tested two shoot induction media (SIM) which differ in the growth regulators. The effect of genotype and culture medium on bud induction, shoot elongation, and the subsequent development of roots is examined.

Material and Methods
A schema of all the assays and parameters noted along the work of our study is shown in Figure 1.

Plant Material, Germination, and Culture Conditions.
Materials consisted of two accessions of S. melongena (coded as M1 and M2), two accessions of S. aethiopicum (coded as A1 and A2), and the four S. melongena × S. aethiopicum interspecific hybrids (Table 1). All seeds of the M1, M2, A1, and A2 parents as well as those of the interspecific hybrids were provided by Dr. John R. Stommel (ARS-USDA, Beltsville, MD, USA). Seeds came from the same harvest and were conserved under the same conditions (stored in a nofrost refrigerator at 4 ∘ C).
Seeds were surface-sterilized by immersion for 15 min in a solution of 25% commercial bleach (containing 40 g L −1 of active chloride) followed by three rinses in sterile distilled water and cultured in plastic Petri dishes (90 × 25 mm) sealed with parafilm containing 40 mL of basal medium (BM), which consisted of Murashige and Skoog salts [36] including vitamins (DUCHEFA, Haarlem, The Netherlands), 1.5% sucrose, and 0.7% plant agar ( Figure 1). The root system of germinated plantlets and the ends of cotyledons were cut and removed, and shoots were transferred to tubes (15 cm in length and 22 mm in diameter) containing 15 mL of BM medium for rooting.
After 30 days of culture, the percentage of rooting, callus formation (presence/absence), and fresh weight (FW) and dry weigh (DW) of interspecific hybrids roots was noted in 10 plants of each genotype. DW of roots was obtained after oven drying at 70 ∘ C for 24 h.
All genotypes were micropropagated and maintained in in vitro culture by transferring nodes every 3-4 weeks to fresh BM. These plants were used as source of explants.
The pH of all the media was adjusted to 5.8 before sterilization at 121 ∘ C for 20 min, and cultures were incubated in a growth chamber at 26 ∘ C ± 2 ∘ C under a 16 h photoperiod with cool white light provided by Sylvania cool white F37T8/CW fluorescent lamps (90 mol m 2 s −1 ).

Organogenesis Induction.
Leaf explants (0.6-0.8 cm 2 ) were obtained from in vitro cultured plants and placed with the abaxial side in contact with the shoot induction media (SIM) containing Murashige and Skoog's salts [36], 3% sucrose, and 0.7% plant agar supplemented with either 2 mg L −1 BA plus 0.5 mg L −1 Kin (SIM-A) or 0.05 mg L −1 TDZ (SIM-B). Growth regulators were filtered (0.22 m Millipore filters) and then added to sterilized medium. The media were plated in Petri dishes (90 × 25 mm) with 40 mL of culture medium per plate. For each combination of genotype and treatment, five repetitions (plates), with five explants per plate, were evaluated.
After 20 days of culture on SIM media, the induction of buds, necrosis appearance, and callus formation was visually assessed using these three indexes: bud index (BI), necrosis index (NI), and callus index (CI) in a scale from 0 (absence) to 4 (throughout the explant). Subsequently, the explants were transferred to BM supplemented with GA 3 (1 mg L −1 ) for elongation. Explants producing many buds were divided into portions and tagged in order to trace their origin. The frequency of explants with organogenic buds (B), frequency of shoot regeneration (R), and mean number of shoots per explant (PR) were measured 20 days after the explants had been transferred to the elongation medium (i.e., 40 days since the beginning of cultivation of the explants in SIM media) ( Figure 1).
Ten shoots isolated from regenerating explants (cultivated in either SIM-A or SIM-B and subsequently cultivated in elongation media) were transferred to tubes with BM for rooting ( Figure 1). At 40 days of culture percentage of rooting was note and those without roots were subcultured to BM supplemented with indolbutiric acid (IBA) at 1 mg L −1 in order to induce rooting.   the first week, each plant was protected from dehydration by covering it with an inverted transparent plastic vessel.

Statistical Analysis.
Data was subjected to factorial analysis of variance (ANOVA). For mean separation, the Duncan multiple-range test was used. The Statgraphics Centurion XVI software (StatPoint Technologies, Warrenton, VA, USA) was used for the statistical analyses.

Results and Discussion
Interspecific hybrids between S. melongena and S. aethiopicum are of interest as rootstocks for improving the yield of eggplant [2]. These hybrids are heterotic for plant vigor, can incorporate resistance to several diseases from the S. aethiopicum parent, have a good grafting compatibility with eggplant, and, given their low levels of glycoalkaloids, do not 4 The Scientific World Journal present problems associated with the possible translocation of these compounds, which may be harmful for human health [3,7,10,11,14,17]. However, up to now, no protocols for in vitro culture and regeneration, which are necessary for multiple biotechnological improvements, have been established for this new innovation in the field of eggplant grafting. Micropropagation, somaclonal variation, and genetic transformation are some of the techniques that can benefit from the development of regeneration protocols [18].
This study was initiated with the establishment in vitro and micropropagation of four interspecific S. melongena × S. aethiopicum genotypes and their respective parents. These plants were used as source of leaf explants that were cultured onto two shoot induction media (SIM): the first (SIM-A) is supplemented with 6-benzylaminopurine (BA) (the most common growth regulator used in eggplant) plus kinetin (Kin), a combination which resulted in high frequency of developed shoots per explant in the work published by Shivaraj and Rao [37]; the second (SIM-B) is supplemented with thidiazuron (TDZ), which is a good inducer of shoot regeneration in several species including eggplant and S. aethiopicum [26,35].
After 20 days of culture, bud induction was observed in all the treatments with the exception of explants from genotype A2 cultured on SIM-A (Table 2). ANOVA revealed significant effects of genotype and medium for BI, NI, and CI, as well as a significant genotype × medium interaction for NI (Table 2). Among genotypes, the lowest BI values were obtained in S. aethiopicum genotypes (A1 and A2). In general, a higher proportion of explants with buds were obtained in SIM-B than in SIM-A, with average values of 2.27 and 1.05, respectively ( Table 2). Among explants from parent (M1, M2, and A1) or hybrid (M1 × A1, M1 × A2, M2 × A1, and M2 × A2) genotypes, none or small differences were obtained for BI in SIM-B. However, in SIM-A a higher response was obtained in explants from hybrids compared to those from parents. Necrosis was present in many explants and it was higher in medium SIM-A than in medium SIM-B (Table 2, Figures 2(a) and 2(b)). The lowest levels of necrosis (NI < 1) were observed in explants of S. melongena M1 and M2 and in the hybrids derived from M1 (M1 × A1 and M1 × A2) cultured on SIM-B. Differences for CI were also observed among genotypes. Among parents, A1 cultured on SIM-A medium showed the lowest callus formation. In hybrids, the lowest CI values were obtained in M2 × A1 cultured on SIM-A medium (Table 2). B, R, and PR were measured 20 days after transferring explants to the elongation medium (Table 3). ANOVA revealed significant effects of genotype and medium for B, R, and PR, as well as a significant genotype × medium interaction for the three indexes. The highest percentages of explants with buds (B = 100) were obtained in the interspecific hybrids previously induced in either SIM-A or SIM-B. The lowest values were obtained in S. aethiopicum genotypes A1 (B = 58) and A2, which was excluded from this analysis due to the low percentage of responding explants (BI = 0.36 on SIM-B and BI = 0.00 on SIM-A). As for the frequency of explants with organogenic buds, B was higher in SIM-B than in SIM-A. The lowest B values were obtained in explants of M2 (60%) and A1 (16%) cultured on SIM-A. Thus, TDZ resulted in a higher bud induction than BA combined with Kin (Table 3, Figure 2(a)) as it is observed with BI ( Table 2). The frequency of shoot regeneration (R) was, in general, lower than B and higher in explants from hybrids than in explants from parents. In genotype A2, no explants produced shoots in SIM-A medium (Figure 2(a)). PR greatly differed among parents and hybrids (Table 3). Whereas a few number of explants regenerated more than 1 shoot per explant in parents (average PR > 1 values were only obtained in M1 and M2 cultured on SIM-A and SIM-B media, resp.), values of PR between 2 and 9 were obtained in interspecific hybrids ( Table 3). The highest PR value (8.68) was obtained in M2 × A1 cultured on SIM-B. In some explants previously cultured on SIM-B vitrification was observed (Figure 2(b)). Despite this, PR values indicate that the number of healthy developed isolable shoots from explants was either similar in both media or higher in SIM-B than in SIM-A. The results obtained for the BI and B (80% to 100%) are in agreement with the reported great morphogenetic potential of eggplant tissues for responding to organogenesis [23][24][25][26][27][28]. However, the accessions of S. aethiopicum (A1 and A2) could be considered of low potential or even recalcitrant for regeneration (A2). These accessions had lower organogenesis response than other S. aethiopicum genotypes (BBS107 and BBS116) assayed in a medium similar to SIM-B [35]. Regarding the interspecific hybrids S. melongena × S. aethiopicum, regeneration was higher than in their respective parents as it is reflected for B, R, and PR values. Thus, they are heterotic for regeneration. This effect was also observed in hybrids of S. lycopersicum L. × S. pennellii Correll, in which populations derived from this cross allowed the detection of six QTLs involved in the regeneration capacity [38]. Although in S. melongena the genetics of regeneration has not been studied, several genes are probably implicated. In general, higher BI, B, R, and PR were obtained in SIM-B compared to SIM-A. Thus, at the concentrations used in our work, TDZ (SIM-B) was more effective than BA plus Kin (SIM-A) in inducing adventitious shoot regeneration from leaf explants. The effectiveness of TDZ in inducing adventitious shoot regeneration with respect to other growth regulators has been reported in several works [39][40][41][42][43].
Necrosis of explants and callus formation are observed in explants cultured in both SIM-A and SIM-B media, with higher NI in SIM-A and higher CI in SIM-B. Although necrosis may reduce regeneration, genotypes with similar NI have showed different regeneration response. Thus, similar NI values were observed on SIM-A cultured explants of the low responding genotype A1 and in other high responding genotypes like M1, M1 × A1, or M2 × A1. Necrosis may be related to ethylene production or accumulation in in vitro culture conditions [44][45][46]. The higher callus formation in explants cultured on SIM-B does not make the isolation of shoots difficult. Although some shoots were vitrified in this medium, similar or higher PR was observed in SIM-B versus SIM-A. It is interesting to take into account that callus formation may increase the appearance of somaclonal variation, which may be an advantage or a drawback depending on the goal of 6 The Scientific World Journal regeneration. Thus, in the case of interspecific hybrids and genotype M1, which had high regeneration in both media, depending on the aim of the induction of regeneration, the more adequate one of these two SIM can be used. A successful rooting of in vitro cultured plants is a prerequisite for micropropagation or transference to field conditions [18]. In eggplant, fail in rooting has been previously reported [25][26][27]37]. This problem is also commonly reported in other species in shoot induction media containing TDZ [39][40][41][42][43]. Thus, the rooting ability of interspecific hybrids was tested, first in the starting micropropagated plants and, secondly, in plants regenerated from leaf explants.
Roots of plantlets were excised and shoots were transferred to individual tubes containing BM. At 30 days of culture, 100% of rooting was observed in the four tested genotypes although callus formation at the base of the plants was present in some plants. The presence of calli differed among genotypes: it is observed in 37.5% of plantlets of interspecific hybrid M1 × A1, in 6.7% of M1 × A2, and in 4% of M2 × A2 (Table 4). FW and DW of roots (without calli) were measured in order to quantify putative differences for root development in the micropropagated interspecific hybrids. Among tested genotypes, a greater root development was observed in M1 × A1 when compared to the other interspecific hybrids (Table 4).
In plants isolated from explants, the frequency of rooting after 30 days of culture in MB (Table 5) was lower than that obtained in micropropagated plants (in the range between 10% and 70% versus 100%). Thus, the cytokinin used in both SIM media may be diminishing rooting capacity. Callus formation was also appreciated at the base of the shoots in all treatments although with low frequencies (between 10% and 20% in six out of the eight treatments). The differences for rooting and callus formation among genotypes could be due to different concentrations or profiles of endogenous growth regulators [40,41]. Regarding differences from the origin of shoots, similar or higher rooting frequencies were observed in shoots from SIM-B when compared to SIM-A ( Table 5). The transfer of shoots without roots to BM medium supplemented with IBA at 1 mg L −1 induced roots in few days. Thus, rooting is not a limiting step for regeneration of the genotypes assayed.
Standard acclimatization procedures were applied to regenerated plants with a 95% of survival.
Overall, results show that TDZ, at the low concentration used in our work (medium SIM-B), is adequate to induce buds in all tested genotypes. These buds develop into shoots able to root on BM or BM supplemented with IBA. Thus, a protocol has been established for the suitable regeneration of S. melongena × S. aethiopicum interspecific hybrids and their parents. We have also demonstrated that these interspecific hybrids are heterotic for regeneration, which may be of interest for basic science studies.

Conclusions
Taking into account all the results obtained it is concluded that a higher capacity for regeneration is observed in interspecific hybrids compared to their parents. Thus, hybrid vigor The Scientific World Journal 7 is manifested for in vitro culture traits. Complementation of parental genes positively influencing regeneration may be taking place in these hybrids. SIM-B seems to be better than SIM-A for bud induction as frequencies of 100% or near 100% were obtained in all tested genotypes. Development of buds into shoots was also higher in SIM-B versus SIM-A, as regenerated shoots were isolated in all genotypes. Elongation of shoots from SIM-A medium may be better, although the higher amount of buds in SIM-B medium cultured explants gives a higher proportion of isolated shoots. Although necrosis is visible in some explants and may contribute to the lack of development of buds into shoots, genotype has a great influence as genotypes with similar NI differ in R. Although rooting may fail in the regenerated shoots due to cytokinins type and/or concentration, those used in our work do not limit rooting capacity of the S. melongena × S. aethiopicum interspecific hybrids. The results are of interest for the development of S. melongena × S. aethiopicum interspecific hybrids as rootstocks for eggplant.