Towards onions and shallots (Allium cepa L.) resistant to beet armyworm (Spodoptera exigua Hübner) by transgenesis and conventional breeding

Onion ( <em>Allium cepa</em> L. group Common Onion) and shallot ( <em>A. cepa</em> L. group Aggregatum) are two subspecies of <em>A. cepa</em> . Both onion and shallot together with other <em>Allium</em> species like garlic ( <em>A. sativum</em> ), leek ( <em>A. porrum</em> ) and bunching onion ( <em>A. fistulosum</em> ) are very important vegetable crops on a worldwide scale. <em>A. cepa</em> is cultivated mainly as a biennial but some types are treated as perennials. It is propagated by seeds, bulbs, or sets (small bulbs). TThe bulbs of common onion are large, normally single, and plants are reproduced from seeds or from seed-grown sets. By contrast the bulbs of shallot are smaller than common onions, they form aggregated clusters as a result of the rapid formation of lateral bulbs or shoots. Reproduction of shallot is almost only vegetative via daughter bulbs although seed production is possible.The cultivation of onion and shallot is sometimes severely limited due to the occurrence of diseases and pests. The most important pest in <em>A. cepa</em> cultivation for (sub) tropical zones is the beet armyworm ( <em>Spodoptera exigua</em> Hübner). The beet armyworm is an extremely polyphagous insect. In this research programme, two different strategies were followed to develop plant material, which is resistant to <em>S. exigua</em> , namely via marker-assisted breeding (MAB) and via genetic transformation. For marker-assisted breeding, first of all a reliable bio-assay needs to be developed. Secondly, suitable sources of resistance must be identified. Thirdly, the genetic basis (and the mechanism of resistance) needs to be uncovered and fourthly, molecular markers must be linked to the resistance gene(s), and finallyourthly an efficient means of screening large populations for the molecular markers should be available and the screening technique should have high reproducibility.In marker-assisted breeding, we succeeded to develop an <em>in vivo</em> and <em>in vitro</em> bio-assay for the identification of resistance to beet armyworm in <em>A. cepa</em> and its wild relatives (Chapter 2). In the <em>in vivo</em> bio-assay the lowest larval survival (36 %) and the lowest fresh weight (10.3 mg per larva) were found on <em>A. roylei,</em> a wild relative of <em>Allium cepa</em> L.. This was not, however, significantly different from the resistance in the tropical shallot cultivar Bawang Bali. Furthermore, in the <em>in vitro</em> bio-assay no toxic insecticidal compound was present in <em>A. roylei</em> because no significant differences were found in mean fresh weight per larva and mean survival of larvae among different accessions. There were also no significant differences in pupal weight and developmental time. All larvae became pupae 10 days after inoculation. Therefore, a marker-assisted breeding approach to introduce beet armyworm resistance in <em>A. cepa</em> was abandoned.For genetic transformation, first of all a reliable plant regeneration system from callus cultures and suspension cultures must be established. Secondly, an efficient and stable transformation system should be developed. Thirdly, molecular characterization techniques of transgenic plants should be available and fourthly, a construct carrying specific <em>B. thuringiensis cCry</em> or other gene sequences effective against <em>S. exigua</em> should be available. We started with a systematic study on the effects of subspecies, cultivar, basal medium, sucrose concentration and 2,4-dichlorophenoxyacetic acid concentration on callus induction, propagation and subsequent plant regeneration in <em>Allium cepa.</em> A reliable regeneration system was developed based on mature zygotic embryo-derived callus. It was shown that regeneration in 45.4% of the callus lines using an optimal combination of factors influencing tissue culture response in <em>Allium</em> could be obtained (Chapter 3).The development of a reliable regeneration system is thought to be of vital importance for the next step in our research, which is the development of a reliable transformation system. The aim was to identify callus lines with a high regeneration potential and to set up a suspension culture system for later transformation studies by means of particle bombardment. Suspension cultures were initiated from callus cultures of <em>Allium cepa</em> , which had been precultured on a solidified medium for seven months. For another three months the 83 callus lines were kept in suspension culture. Of these, 20 suspension lines showed adequate growth and were used to test the effect of cytokinins on plant regeneration. However, it was found that the plant regeneration capacity of selected lines significantly decreased in a relatively short time-period. Plant regeneration also proved to be highly dependent upon the line used. Contrary to expectations, the type and concentration of cytokinins could not increase the shoot regeneration capacity (Chapter 4). On the basis of these results it was the concludedsion that only relatively young callus (less 3 months) could be used for <em>Allium</em> transformation.In Chapter 5 the development of a reliable transformation system is described using <em>Agrobacterium tumefaciens</em> as a vector. A number of parameters, such as callus age, callus induction medium, co-cultivation period, osmotic treatment, cultivars, selection method and <em>Agrobacterium</em> strains were used to evaluate transient expression of the <em>uidA</em> reporter gene in a series of experiments followed by evaluating stable expression conditions. Eventually, an <em>Agrobacterium tumefaciens-</em> mediated transformation system both for onion and shallot was developed using three-week old callus derived from mature embryos using two different strains EHA105 (pCAMBIA 1301) and LBA4404 (pTOK233) carrying a binary vector coding for <em>uidA</em> and <em>hpt</em> . This transformation protocol for onion and shallot can be used year-round because mature zygotic embryos are used as starting material. A total of 11 independent callus lines regenerating transgenic plants have been produced in several independent experiments. In one of these experiments, starting with 154 embryos, transgenic plants were recovered from three independent callus lines, giving a transformation frequency of 1.9 %. The best line produced 90 transgenic plants. Subspecies (onion and shallot) and cultivar were important factors for successful transformation: shallot was better than onion and for shallot with cv. Kuning the best results were obtained. Also, it was found that constantly reducing the size of the calli during subculturing and selection by chopping, thus enhancing exposure to the selective agent hygromycin, improved the selection efficiency significantly. The transformants were genetically characterized by means of standard PCR, genomic DNA blot hybridization and FISH (fluorescence <em>in situ</em> hybridization).Genomic DNA blot hybridization is traditionally used to analyse integration of foreign genes into host genomes. . Although genomic DNA blot hybridization can be used to determine the presence of T-DNA and copy number, it becomes labour-intensive when many samples have to be assayed, and it consumes relatively large amounts of genomic DNA. Especially for large genome species crops, e.g. onion (15290 Mbp/1C), with a genome size which is 105 times larger than that of <em>Arabidopsis.</em> Adapator ligation PCR (AL-PCR) followed by the sequencing of the genomic DNA flanking the T-DNA borders was developed (Chapter 6). The AL-PCR patterns obtained were specific and reproducible for a given transgenic line. The results showed how T-DNA integration took place and it also provided insight into the number of T-DNA copies present. Comparison of AL-PCR and previously obtained Southern hybridization results pointed into the direction of rather complex T-DNA integration patterns in some of the transgenic plants. After cloning and sequencing of the AL-PCR products, the junctions between plant genomic DNA and the T-DNA inserts were analysed into great detail. For example, it was shown in one of the transgenic lines that upon T-DNA integration a 66 bp sequence was deleted, and no filler DNA was inserted. Primers located within the left and right flanking genomic DNA in transgenic shallot plants were used to recover this particular target site for T-DNA integration. The target site sequence proved to be repetitive and <em>Allium</em> specific because a similarly-sized PCR fragment was obtained in all <em>Allium</em> species tested while it was not detected in closely related genera such as <em>Lilium</em> and <em>Tulipa</em> .In Chapter 7, the general discussion, the possibilities for the development of onion and shallot resistant to <em>Spodoptera exigua</em> are discussed. It is argued that both MAB and genetic transformation methods are potentially very powerful and that in case of the introduction of resistance to beet armyworm into onion and shallot genetic transformation is the most promising.