J Sci.Univ.Kelaniya 6 (2011) : 13-34 PRODUCING INTER-SPECIFIC HYBRIDS BETWEEN BRASSICA JUNCEA (L.) CZERN & COSS AND B. OLERACEA (L.) TO SYNTHESIZE TRIGENOMIC (ABC) BRASSICA

Polyploidy is recognized as a major mechanism in plant evolution. Polyploid crops often have wider adaptation, better quality and higher yielding capacity than their diploid counterparts. Although many successful natural and man-made hexaploid crops are existing, hexaploid Brassica are still not available. So far, relatively a limited work has been conducted to synthesize hexaploid Brassica with A, B and C genomes, which will provide a very good potential to create new crops for domestication. An investigation was conducted to evaluate the possibility of synthesizing trigenomic (AABBCC) hexaploid Brassica by crossing Brassica juncea (L.) Czern & Coss and B. oleracea (L.). Five genotypes of B. juncea (AC 0747, 0790, 1099, 2180 and 7700) and five genotypes of B. oleracea (Chinese Broccoli, Broccoli-var. Shogun, Cauliflower-var. Snowball and var., Phenomenon Early and Cabbage-var. Sweet Eureka) were selected for the study. Hand pollination was done by emasculating buds of one species and pollination using another species in both directions. Success of pod formation of the crosses of B. juncea (♀) x B. oleracea (♂) was 25%. Totally 893 putative hybrid seeds were harvested. Although 9% pod formation was observed in reciprocal crosses, no seeds were developed. Evaluation of 80 putative hybrids by molecular markers and agro-morphological characterization confirmed four true hybrids resulting crosses between AC 0747 x Chinese Broccoli, AC 0790 x Chinese Broccoli and AC 2180 x Broccoli-var. Shogun. The present investigation confirms that hybridization of tetraploid B. juncea (4x, AABB) with diploid B. oleracea (2x, CC) is a potential approach to produce hexaploid Brassica (6x, AABBCC) genotypes.


INTRODUCTION
The genus Brassica has been subjected to a great deal of scientific attention because it contains very important agricultural and horticultural crops. Brassica oilseed species together constitute the third most important oilseed crop in the world after oil palm and soybean (Ahuja et al., 2010). Brassica oilseed species are an important source of oil for human consumption, for protein rich meal for animal consumption and for bio-diesel production. Among many Brassica species the most common oilseed crops cultivated on a commercial scale are rapeseed (B. napus L. and B. rapa L.) and mustard (B. juncea [L.] Czern & Coss).
Brassica oleracea and B. rapa are species commonly consumed as vegetables in many countries. Brassica vegetables are rich in different forms of vitamins, calcium, iron, magnesium, phosphorus, potassium, zinc and soluble fiber (Zhang et al., 1999). They also contain multiple nutrients with potent anti-cancer properties such as diindolylmethane, sulforaphane and selenium (Lampe and Peterson, 2002;Brandi et al., 2005 ). In Sri Lanka, B. oleracea varieties such as cabbage, broccoli, cauliflower are widely grown as vegetables in the up-country.
Flowers in the genus Brassica are hypogynous, mostly actinomorphic. Sepals 4, in 2 decussate pairs and free. Petals 4, alternate with sepals, arranged in the form of a cross. Stamens 6, in 2 whorls, tetradynamous (lateral (outer) pair shorter than median (inner) 2 pairs). There are four nectar glands which are median and lateral. Anthers are dithecal, dehiscing by longitudinal slits. Pollen grains 3-colpate, trinucleate. Nectar glands receptacular and disposition around base of filaments, always present opposite bases of lateral filaments, median glands present or absent. Pistil 2-carpelled; ovary superior, sessile or borne on a distinct gynophore, mostly 2-locular and with a false septum connecting 2 placentae (Erbar and Leins, 1997). Brassica juncea is mainly a self pollinated crop, but 8-18% out crossing is also observed (Labana and Banga, 1984). The primary pollinating mechanism is by wind.
However, insect pollination substantially enhances the outcrossing (Delaplane and Mayer, 2000). B. oleracea is primarily a cross pollinated plant and the self fertility is prevented by a complex system of self incompatibility (Stephenson et al., 1997). B. oleracea flowers produce nectar and pollen in abundance and they are extremely attractive to the bees. It was observed that when bees were present, the plants produced more seeds for síliquas, besides the same ones were larger and viable (Mussury and Fernandes, 2000). In B. juncea the maximum stigma receptivity was recorded one day before flower opening (Labana and Banga, 1984), whereas in B. oleracea the stigma receptivity is maximum 2-4 days before flower opening and it is poor during the evening (Delaplane and Mayer, 2000).
In Brassica pollen viability varies with environmental conditions, particularly temperature and humidity. Under controlled conditions in the laboratory, Brassica pollen can remain viable for between 24 hours and one week under natural conditions pollen viability gradually decreases over 4 -5 days (Ranito-Lehtimäki, 1995). Therefore, as a practice Brassica breeders cross pollinate Brassica in early mornings when the environment temperature is low and the relative humidity is high to maximize successful hybridization (Meng et al., 1998;Li et al., 2005).
Polyploidy, a change whereby the entire chromosome set is multiplied, is recognized as a major mechanism in plant evolution. Allopolyploidy often occurs in association with inter-specific hybridization (Leitch & Leitch, 2008). In terms of plant breeding, induction of polyploidy could initiate new genetic combinations which provide the breeders with more variability. Further, the induction of polyploidy is a common technique to overcome the sterility of a hybrid species during plant breeding (Leitch & Leitch, 2008). Angiosperms are remarkable in their ability to tolerate the considerable genomic impact of polyploidy arising from the accommodation of divergent genomes in the same nucleus, which results in an instantaneous multiplication in DNA content (Leitch & Leitch, 2008).
There are many successful hexaploids (6x) in crops such as bread wheat, triticale, oat and kiwifruit. Common wheat (Triticum aestivum L.) is a fine example of hexaploid plants having wider adaptation, better quality and higher yielding capacity than its tetraploid counterpart, durum wheat (Triticum durum Desf.) (Gooding & Davis, 1997). Even though canola and mustard are important crops as a major source of plant-based oil, the highest ploidy level in the Brassica is tetraploid (4x).
The cytology and relationships between six of the Brassica species are well understood (U, 1935). The six species have three types of diploid genomes designated as A, B, and C either singly or in pairs ( Figure 1). Brassica napus, B. juncea and B. carinata are the naturally occurring tetraploids. Amphidiploid species, B. napus, B. juncea and B. carinata have been re-synthesized by hybridizing diploid species followed by doubling the chromosomes (Song et al., 1995). This resulted in completely homozygous polyploid lines. So far relatively limited work has been done on the synthesis of hexaploid Brassica species. Hexaploids of Brassica (AABBCC) have been synthesized from reciprocal inter-specific crosses between yellow-seeded B. campestris (AA) and B. carinata (BBCC) to transfer the genes for yellow seed coat from both genomes A and C to B. napus (AACC) (Meng et al., 1998). Li et al (2005) and Jiang et al. (2007) were able to produce trigenomic hexaploid Brassicas successfully by inter-specific crosses between B. rapa (AA) and B. carinata (BBCC). Takeda and Takahata (1996) succeeded in producing alloplasmic Chinese cabbage (B. oleracea) using synthesized trigenomic hexaploid Brassica. Extensive work on molecular mechanisms regarding the cytology in inter-generic hybrids between synthetic Brassica allohexaploids (2n = 54, AABBCC) and another crucifer Orychophragmus violaceus has also been carried out by Ge & Li (2006) and Li & Ge (2007) as a further step towards understanding the genome-specific chromosome behavior in wide hybrids. Hexaploid Brassica plants of the genomic constitution AABBCC were also synthesized by crosses between B. rapa (AA), and B. alboglabra/B. oleracea (CC) and B. carinata (BBCC) (Rahman, 2002). Double haploid technology was used by Nelson et al. (2006) and Nelson et al. (2009) to further advance the inter-specific hybridization of Brassica spp. and rapidly developed homozygous populations for cultivar development. In the genus Brassica, production of hexaploid somatic hybrids (AABBCC) were also reported. Yamagishi et al. (1989) and Arumugam et al. (1996) were able to produce hexaploid  Although somatic hybridization is still an important tool in plant breeding, sexual polyploids are gaining increasing importance (Ramanna & Jacobsen, 2003).
Alfalfa sexual polyploids are more productive than somatic ones (McCoy & Rowe, 1986). In red clover polyploid plants can be produced more consistently by sexual means than through somatic hybridization (Simioni et al., 2006). There are at least three possible approaches to produce hexaploid Brassica (Yan & Weerakoon, 2007).
ie. The use of one tetraploid and one diploid as parents (4x -2x) followed by the chromosome doubling of triploid hybrids; the use of three tetraploids as parents (4x -4x -4x) and the use of three diploids as parents (2x -2x -2x).
Hybridization between a tetraploid and a diploid species is difficult and failures occur at many stages starting from pollination incompatibility to pre/postgermination barriers. Most inter-specific crosses do not produce mature seeds due to failure of endosperm development (Nishiyama et al., 1991). Ovule culture was used to overcome postzygotic inter-specific incompatibility in reciprocal crosses between B.
rapa and B. oleracea (Diederichsen & Sacristan, 1994). Similarly, the cross between B. napus and B. oleracea is normally unsuccessful, but the use of embryo culture techniques can produce hybrids (Gowers & Christey, 1999 Therefore, this study would provide a significant contribution to produce a new crop for domestication with a number of potential properties. The present study attempted to examine the success rate of crosses between tetraploid B. juncea and diploid B. oleracea and to obtain inter-specific F 1 hybrid seedlings of triploid nature (ABC). Sweet Eureka) were used in the experiment. Ten replicates for each parental genotype were grown in a glasshouse at the Open University of Sri Lanka, Nawala and the crosses were made between B. juncea (♀) and B. oleracea (♂) varieties. The flowers of the female parent were opened and emasculated using a fine forcep and fresh pollen from the male parent was transferred to the stigma. There were 44 different cross combinations (Table 1). Depending on the availability of flower buds, 30-60 crosses were made for each cross combination. The pollination was carried out between 6.00 -7.30 am to ensure successful pollination. The pollinated flowers were covered with perforated plastic bags and tagged. Reciprocal crosses (B. oleracea (♀) x B. juncea (♂)) were undertaken for each of the crosses. All hybridizations were performed without the aid of embryo rescue. The number of flowers crossed, mature pods formed and the seed set for each cross was recorded.

Agro-morphological evaluation
About 10% of the hybrids seeds were germinated and 80 F 1 hybrid seedlings were obtained. They were transplanted in pots and agro-morphological traits such as flowering time, stem height at flowering, number of leaf nodes at flowering, leaf, petiole, stem colour and hairiness, flower colour and seed coat colour were recorded for each F 1 plant.
Lyophilised oligonucleotide primers were reconstitute by addition of Mili Q water to bring the concentration to 500 µM. Then 50 µM and 10 µM dilutions were prepared from 500 µM stock solutions. Deoxyribo nucleotide mixture (dNTP), was prepared by diluting stock solutions of (100 mM) dATP, dCTP, dGTP and dTTP in an eppendrof tube to the final concentration of 2 mM. The standard PCR protocol for one sample is given in Based on the results, the SSR marker sN12353 was selected for confirming the true hybridity of 80 F 1 hybrid plants. Cross progeny were considered true inter-specific hybrids when they possessed sN12353 alleles from both parents.

Pollen viability
Pollen viability of the parental plants, B. juncea and B. oleracea was tested with 1% acetocarmine.
The same procedure was followed to test the pollen viability of the four confirmed true hybrids.  cm) parents. The average number of seeds in B. juncea was 12.9 ± 1.37 and in B.
However, no seeds were developed in mature pods (Table 1).
About 10% of the F 1 hybrid seeds were germinated to obtain 80 F 1 seedlings.
Although usually B. juncea and B. oleracea seeds take 2-3 days to germinate, the F 1 seeds took 5-20 days to germinate.

Agro-morphological evaluation
The 80 resulted F 1 hybrid plants were evaluated based on their agromorphological characters. A range of variation in agro-morphological traits was observed in the F 1 hybrids of each cross combination. The agro-morphological characters observed were leaf colour, colour of petiole, hairiness of leaves, colour of stem, colour of flowers, colour of stem, flowering time, stem height-and number of leaf nodes at flowering. In F 1 hybrid plants mainly the leaf morphology (shape and size), hairiness of leaves and the colour of petioles varied in different degrees compared to their respective parents. The agro-morphological variation among some selected putative F 1 hybrids with their parents were presented in Table 3. There was a wide range of morphological differences among the F 1 individuals resulted from crosses of B. juncea (♀) x B. oleracea (♂) genotypes. Figure 2A and Figure 2B show morphological differences between the leaves and pods of B. juncea (AC 0747) and B.
oleracea (var. Chinese Broccoli) parents and their F 1 hybrids. True F 1 hybrid plants, which were confirmed by molecular markers, are more vigorous than none-truehybrids ( Figure 3). Compared to none-true-hybrids, true hybrids have broader and bigger leaves, taller and stronger stems, many branches, flowers and pods. However, since the pollen viability was very low, although many pods were formed, no seeds were developed in any of the true hybrid triploid (ABC) plants.

Molecular confirmation
Out of 80 putative hybrids, four F 1 hybrids, i.e. crosses of one AC 0747 x Chinese Broccoli, two AC 0790 x Chinese Broccoli and one AC 2180 x Broccoli (var. Shogun) were confirmed to be true hybrids with the SSR marker sN12353. Figure   A (x 1/5) B (x 1/3)     Figure 4B shows the DNA profile of parents and two hybrids in 4% Agarose 1000 (Invitrogen) gel using the SSR marker sN12353.
However, the stainability in 1% acetocarmine of four confirmed true hybrids was very poor (5%) indicating that they are sterile.

DISCUSSION
Inter-specific reproductive isolation is the main mechanism for speciation and specific maintenance (Griffiths et al., 2002). However, genetics leaks in the isolation system have occasionally enabled inter-specific gene flow and promoted the evolution of species (Jiang et al., 2007). Polyploidy often occurs in association with interspecific hybridization. It is long been established as a key mechanism in plant evolution and adaptation (Leitch & Leitch, 2008). Increase in genome size (polyploids) of plant species is associated with an enhanced cell size and dry matter production (Arnold, 1997) (Meng et al., 1992;Meng & Lu, 1993;Liu & Meng, 1995). Many Brassica inter-specific crosses were successful when the female parent is with a higher ploidy level than the male parent (Schelfhout et al., 2006). Consistent with the present results, Schelfhout et al. (2006) also found that the success rate was better in inter-specific crosses of Brassica only when tetraploids were used as female parents.
Hybridization between allotetraploid species, B. napus, B. juncea, B. carinata and the diploids, B. nigra, B. oleracea and B. rapa are naturally highly incompatible (Diederichson & Sacristan 1994). Fertilization may take place, but abortion occurs early in the development of the embryo. In the present study as well the crosses in both directions a very high rate of seed abortion occurred early in the development. Total seed set of the crosses between B. juncea (♀) and B. oleracea (♂) was 793 ie. only 3.2±0.63 seeds per pod, whereas for reciprocal crosses seed set was zero.
Natural crossing does not occur between B. juncea and B. oleracea under field condition (Bing et al., 1996) and also a successful artificial pollinalion has not been recorded until now except obtaining hexaploid Brassica somatic hybrids by protoplast fusion between B. oleracea (CC) and B. juncea (AABB) by Arumugam et al., (1996).
In an investigation of inter-specific hybridization among several Brassica species, Takeda (1983) reported that hybridization between B. juncea and B. oleracea was particularly difficult. This demonstrates the high level of sexual incompatibility between B. juncea and B. oleracea. The barriers might overcome by selecting suitable cross combinations involving different genotypes. Therefore, in our study, a wide range of genotypes were selected to maximize the success rate of crossing.
To confirm the true hybrid nature of putative F 1 plants, agro-morphological traits and molecular markers (SSR) were used. The agro-morphological characterization of putative F 1 hybrids demonstrated that there was a range of agromorphological variation compared to their respective parents. The molecular confirmation with the SSR marker sN12353 proved that four F 1 plants were true hybrids. B. juncea (♀) accessions 0747, 0790 and 2180 were successfully hybridized with B. oleracea (♂) genotypes Chinese Broccoli and Broccoli (var. Shogun) to produce true hybrids. The true F 1 hybrids which were confirmed by molecular analysis clearly showed that the agro-morphologically they are different from the respective parents. However, some of the putative F 1 hybrids were also agro-morphologically different from their respective parents, whereas some were more or less agromorphologically similar to either of their respective parents.
In the present study, B. juncea accessions 0747, 0790 and 2180 appear to have allowed foreign pollen from B. oleracea to germinate on their stigmas and enabled pollen tubes to penetrate their styles. Only varieties with a high crossability have the mechanism(s) to enable the hybrid embryos to develop fully. Evaluation and characterization of the remaining putative hybrid seeds will be carried out and the triploid true hybrid seedlings will be treated with chromosome doubling technique to synthesize hexaploid (6x, AABBCC) population. been carried out (Scholze & Hammer, 1998) and obtained very promising results.

Besides commercially important
Thus, these species offer prospective additional sources of genes/traits for resistance to a number of biotic and abiotic stresses as well as genes/traits for yield, oil content, nutrition value and medicinal properties, when hybridized with six Brassica species in the "U" triangle to produce hexaploid Brassica cultivars.

CONCLUSIONS
Polyploidy is accepted as a key mechanism in plant evolution and adaptation (Leitch & Leitch, 2008). Many successful natural hexaploid plants such as wheat, oat and kiwifruit and man-made hexaploids such as hybrid (Triticale) of Triticum and Secale exist. However, naturally occurring hexaloid Brassica are not available. There are at least three possible approaches to produce hexaploid Brassica. In the present investigation, one tetraploid (4x) B. juncea and diploid (2x) B. oleracea were intercrossed. Evaluation of 80 putative hybrids by molecular markers and agromorphological characterization confirmed four true hybrids resulting crosses between AC 0747 x Chinese Broccoli, AC 0790 x Chinese Broccoli and AC 2180 x Broccolivar. Shogun, ) which is a potential approach to artificially produce hexaploid Brassica (6x, AABBCC) genotypes in future.