Fusarium head blight resistance in wheat : using the in vitro androgenic approach

Wheat ( Triticum aestivum L.) belongs to the three most important food crops in the world. In certain years, the crop can suffer considerable damage as a result of Fusarium Head Blight (FHB), especially as no chemical control is effective against this disease. This disease is mainly caused by the fungi Fusarium culmorum and F. graminearum but in the cooler regions of North-Western Europe, F. culmorum predominates. The pathogen causes a wide range of different damage, of which toxin contamination of the seeds is among the most threatening. These toxins, of which deoxynivalenol (DON), nivalenol (NIV) and acetyldeoxynivalenol (ADON), with the isomers 3-ADON and 15-ADON are most notorious, are capable of inhibiting protein synthesis and are, therefore, extremely hazardous to man and animal. For economic and environmental reasons, host plant resistance is the most appropriate and sustainable disease control method and should be given a high priority in any wheat breeding programme.Androgenesis is the outgrowth of the male reproductive cell into a haploid plant. Up until now, for more than 250 plant species haploid plants have been produced via in vitro androgenesis. However, when using the in vitro androgenesis technique there are still specific problems to be solved for the individual crops. The applications and advantages of in vitro androgenesis are for example rapid production of haploid plants evoking a shorter breeding regime, easier genetic analyses both at crossings and at the DNA level and possibilities for genetic modification and in vitro selection. In vitro selection has been used efficiently to find agronomically altered traits and to produce new cultivars. With the use of toxins as selective agent new resistances have been found in wheat, e.g. against Pseudomonas syringae pv. syringae or Helminthosporium sativum . A general overview on Fusarium Head Blight (FHB), on toxicity of the toxins produced by this pathogen, on in vitro androgenesis and in vitro selection is presented in Chapter 1.In Chapter 2, the results of a 7x7 full diallel on the inheritance of androgenic ability in wheat anther culture are presented. Seven parental cultivars, differing in both androgenic response and FHB-resistance, together with the 42 F 1 -combinations of the complete diallel were evaluated for several androgenic traits in five replicates. In total 130,000 anthers were cultured, of which 14% responded. Diallel data were analysed by the model of Gardner and Eberhart and it appeared that most of the genetic variation could be explained by additive genetic effects. A total of 17,819 embryos were transferred to MS regeneration medium, of which on average 30% regenerated into plantlets. Of them 11% was green. Except for two combinations, green plants were recovered from all 42 F 1 -combinations. Significant genetic differences were found and genetic effects explained 38%, 48% and 21% of the total variation for the percentage of green regenerants, the percentage of albino regenerants and the percentage of embryos that formed only roots, respectively. Additive effects explained 30%, 65% and 37% of the genetic variation and narrow sense heritabilities were 0.11, 0.32 and 0.08, respectively. Replicate effects were highly significant for the factors percentage albino regenerants and percentage embryos with only root formation. No significant General Combining Ability (GCA) effects were found. Variety heterosis was only significant for the percentage green regenerants and specific heterosis was significant for percentage green- and percentage albino regenerants. No reciprocal effects were found. Large significant differences in Specific Combining Ability (SCA) values were observed, with 13.5% of the F 1 -combinations outyielding the best parent.About 2,000 plants were doubled with colchicine and 84% of the doubled haploid (DH) plants could be grown to seed set. For seed set, genetic effects explained 78% of the total variation, but additive effects where responsible for only 4% of the genetic variation and, therefore, the narrow sense heritability was low (0.01). According to heritabilities, for embryo production progress can be rapid, for green plant regeneration it will be important to choose the parents very carefully and for seed set, progress is more hard to make. No correlation between embryo production, plant regeneration or seed set could be found. Over 200,000 seeds were formed on the DH-genotypes.A low green plant regeneration is considered to be one of the main bottlenecks for efficient use of the in vitro androgenesis technique in wheat. To study the inheritance of anther culture response and green plant regeneration more specifically, reciprocal crosses were made between the wheat cultivars Ringo Sztar, Ciano 067 and Benoist H77022, each of which had both a good response in anther culture and a high frequency of green plant regeneration (Chapter 3). It was found that, averaged for all genotypes, 23.0% of the anthers responded and a callus induction frequency of 77.8% was observed.Furthermore it appeared that of all the embryos, 43.0% developed into plantlets, 25.6% of the regenerants being green, resulting in 3.3 green plants per 100 anthers. It was also found that genotypic effects accounted for 57.7%, 86.3% and 77.5% of the total variance of anther culture response, callus induction frequency and embryo induction frequency, respectively. Additive and dominant gene actions were detected for all androgenesis and regeneration characteristics and no reciprocal differences were found, indicating the absence of cytoplasmic effects. It was concluded that embryo production was primarily correlated with anther culture response and not with the number of embryos produced per plated anther or per responding anther.Advantages of in vitro selection compared to in vivo selection are that a larger number of genotypes can be screened under controlled conditions and that a limited amount of space is needed to screen all genotypes. For wheat, resistance against several diseases was reported through in vitro selection with the help of toxins as selective agent. In order to elucidate the phytotoxicity of FHB-produced toxins, effects were studied on four types of wheat plant material i.e. seedlings, coleoptile segments, anther derived callus and anther derived embryos, using different concentrations of DON and 3-ADON (Chapter 4). It appeared that DON inhibited growth of all types of plant material and that the seedling growth response to 4´10 -5M DON of a large set of genotypes did not differentiate between tolerant and sensitive genotypes according to the observed FHB-resistance level in the field. In general, coleoptile segments showed a growth reduction at 10 -5M DON, whereas a concentration of 10 -4M DON appeared to be the optimum concentration to differentiate between haploid wheat calli for DON-tolerance. However, growth analysis data of 40 callus clones did not show any correlation with the known FHB-resistance levels of the original donor genotypes and populations. Regeneration of the anther derived embryos in the embryo selection experiment was decreased 100-fold on DON-containing medium. Averaged across the callus and embryo selection experiments, green plant regeneration showed a decrease of approximately 20-fold on medium containing the toxin.Most of the resistance genes against FHB that are known up to now, are located in for European standards considered exotic wheat genotypes and introduction of these genes into varieties requires extensive backcrossing. A haploid step could accelerate the transfer of the genes to cultivars. The final aim of our research was selection for high FHB-resistance in the field and, therefore, crosses were made between resistant and susceptible genotypes for four consecutive years (Chapter 5). Parents, F 1 - or F 2 -populations were used as donor material for anther culture and were, together with the anther culture derived doubled haploid (DH) wheat lines, tested in the field for their FHB-resistance. Percentage infection was measured three and four weeks after artificial inoculation.Besides infection, also date of flowering and, during two years, straw length was scored. The observed plant traits within the various DH-lines were stable, homogeneous and no visible segregation occurred. In most cases, the F 1 -, F 2 -populations and the DH-lines were for infection level intermediate between the two parental infection levels, indicating an additive inheritance. However, the infection levels of some of the doubled haploids were significantly lower than the levels of the most resistant parent. In cases where the F 2 -populations were significantly more resistant than the two parents, it was concluded that accumulation of resistance genes of the partially resistant parents had occurred. In cases where the DH-lines, derived from cultivars were significantly more resistant than the cultivars, it was attributed to gametoclonal variation. No evidence was found that a longer callus phase might lead to a higher level of variation. In 1994 infection levels were substantially higher than in the three previous years, probably due to higher maximum temperatures in the inoculation period.Microspore culture has several advantages over anther culture, e.g. for in vitro selection experiments, single cells are preferred to multicellular structures as a more uniform selection pressure is secured. The possibility of using isolated microspore culture of wheat for in vitro selection experiments are described in Chapter 6. Experiments were carried out to optimise the isolation and culture of isolated microspores of this recalcitrant crop. It was found that the viability of the microspores was better when co-cultured with wheat ovaries from one cultivar as compared to co-culture with a mixture of wheat ovaries from four cultivars. Furthermore it appeared that co-culture of the microspores with ovaries in culture plate inserts had no significant effect on viability of the microspores, but had, in comparison with culture in agarose rings, a large positive effect on the percentage of swollen microspores. A significant correlation between the number of swollen microspores and the number of multicellular structures was found.For future research it will be necessary to analyse the progenies of the diallel-derived DH-regenerants for their FHB-resistance for several generations. It has to be elucidated whether or not the resistance levels of the highly resistant DH-genotypes will also be found in their offspring. Because in some DH-lines accumulation of resistance genes appears to have occurred, these lines will be very useful in breeding for introgression of this trait into commercial varieties.Wheat and maize, two of the three most important food crops in the world, are affected by Fusarium head blight. Research on FHB resistance and resistant cultivars are of the utmost importance to ensure an adequate food supply around the world.