Fusarium head blight (FHB) of wheat is a widespread and destructive disease which occurs in humid and semi-humid areas. FHB epidemics can cause serious yield and quality losses under favorable climatic conditions, but the major concern is the contamination of grains with mycotoxins. Resistance to FHB is quantitatively inherited and greatly influenced by the environment. Its evaluation is costly and time-consuming. The genetic basis of FHB resistance has mainly been studied in spring wheat. The objective of this study was to map quantitative trait loci (QTLs) for resistance to FHB in a population of 240 recombinant inbred lines (RILs) derived from a cross between the two Swiss winter wheat cultivars Arina (resistant) and Forno (susceptible). The RILs were genotyped with microsatellite and RFLP markers. The resulting genetic map comprises 380 loci and spans 3,086 cM. The 240 RILs were evaluated for resistance to FHB in six field trials over 3 years. Composite interval mapping (CIM) analyses carried out on FHB AUDPC (i.e. mean values across six environments) revealed eight QTLs which altogether explained 47% of the phenotypic variance. The three main QTLs were mapped on the long arms of chromosomes 6D ( R(2)=22%), 5B ( R(2)=14%) and 4A ( R(2)=10%). The QTL detected on 5B originated from the susceptible parent Forno. Other QTLs with smaller effects on FHB resistance were detected on chromosomes 2AL, 3AL, 3BL, 3DS and 5AL.

SummaryPlant resistance (R) genes are highly effective in protecting plants against diseases, but pathogens can overcome such genes relatively easily by adaptation. Consequently, in many cases R genes do not confer durable resistance in agricultural environments. One possible strategy to make the use of R genes more sustainable depends on the modification of R genes followed by transformation. To test a possible transgenic use of R genes, we overexpressed in wheat the Pm3b resistance gene against powdery mildew under control of the maize ubiquitin promoter. Four independent transgenic lines were tested in the greenhouse and the field during 3 years. The four lines showed a five-to 600-fold transgene overexpression compared with the expression of the endogenous Pm3b gene in the landrace 'Chul'. Powdery mildew resistance was significantly improved in all lines in the greenhouse and the field, both with naturally occurring infection or after artificial inoculation. Under controlled environmental conditions, the line with the strongest overexpression of the Pm3b gene showed a dramatic increase in resistance to powdery mildew isolates that are virulent on the endogenous Pm3b. Under a variety of field conditions, but never in the greenhouse, three of the four transgenic lines showed pleiotropic effects on spike and leaf morphology. The highest overexpressing line had the strongest side effects, suggesting a correlation between expression level and phenotypic changes. These results demonstrate that the successful transgenic use of R genes critically depends on achieving an optimal level of their expression, possibly in a tissue-specific way.

We constructed a genetic linkage map based on a cross between two Swiss winter wheat ( Triticum aestivum L.) varieties, Arina and Forno. Two-hundred and forty F(5) single-seed descent (SSD)-derived lines were analysed with 112 restriction fragment length polymorphism (RFLP) anonymous probes, 18 wheat cDNA clones coding for putative stress or defence-related proteins and 179 simple-sequence repeat (SSR) primer-pairs. The 309 markers revealed 396 segregating loci. Linkage analysis defined 27 linkage groups that could all be assigned to chromosomes or chromosome arms. The resulting genetic map comprises 380 loci and spans 3,086 cM with 1,131 cM for the A genome, 920 cM for the B genome and 1,036 cM for the D genome. Seventeen percent of the loci showed a significant ( P < 0.05) deviation from a 1:1 ratio, most of them in favour of the Arina alleles. This map enabled the mapping of QTLs for resistance against several fungal diseases such as Stagonospora glume blotch, leaf rust and Fusarium head blight. It will also be very useful for wheat genetic mapping, as it combines RFLP and SSR markers that were previously located on separate maps.

Bread-making quality in wheat and spelt re¯ects the combination of several, mostly quantitatively inherited parameters. The aim was to ®nd molecular markers linked to quantitative trait loci (QTL) for quality parameters. Zeleny sedimentation values (Zel), protein (Prot), kernel hardness (KH) and 1000-kernel weight (TKW) of 226 F 5 recombinant inbred lines (RILs) from a cross between wheat and spelt were assessed in dierent environments. The dough properties of 204 RILs were assessed with an alveograph. Based on a genetic map of 187 loci, nine QTL were found for Zel and Prot, explaining 47% and 51% of the phenotypic variance, respectively. Fifty-four per cent of the variance was explained by 10 QTL for KH and eight for TKW. For the alveograph parameters 10 QTL were found for baking strength, nine for tenacity, seven for con®guration ratio, and four for elasticity index and extensibility. The phenotypic variance explained ranged from 25% to 48%. The population mean of the dough parameters was shifted towards the spelt parent. It is concluded that non-additive eects are crucial in the expression of high bread-making quality of wheat. The consequences for wheat and spelt breeding programmes are discussed.

A major concern related to the adoption of genetically modified (GM) crops in agricultural systems is the possibility of unwanted GM inputs into non-GM crop production systems. Given the increasing commercial cultivation of GM crops in the European Union (EU), there is an urgent need to define measures to prevent mixing of GM with non-GM products during crop production. Cross-fertilization is one of the various mechanisms that could lead to GM-inputs into non-GM crop systems. Isolation distances between GM and non-GM fields are widely accepted to be an effective measure to reduce these inputs. However, the question of adequate isolation distances between GM and non-GM maize is still subject of controversy both amongst scientists and regulators. As several European countries have proposed largely differing isolation distances for maize ranging from 25 to 800 m, there is a need for scientific criteria when using cross-fertilization data of maize to define isolation distances between GM and non-GM maize. We have reviewed existing cross-fertilization studies in maize, established relevant criteria for the evaluation of these studies and applied these criteria to define science-based isolation distances. To keep GM-inputs in the final product well below the 0.9% threshold defined by the EU, isolation distances of 20 m for silage and 50 m for grain maize, respectively, are proposed. An evaluation using statistical data on maize acreage and an aerial photographs assessment of a typical agricultural landscape by means of Geographic Information Systems (GIS) showed that spatial resources would allow applying the defined isolation distances for the cultivation of GM maize in the majority of the cases under actual Swiss agricultural conditions. The here developed approach, using defined criteria to consider the agricultural context of maize cultivation, may be of assistance for the analysis of cross-fertilization data in other countries.

No abstract

Breeding for wheat varieties resistant to Stagonospora nodorum blotch (SNB) is the most sustainable strategy for controlling the disease. In order to map quantitative trait loci (QTLs) for SNB resistance we analysed 204 recombinant inbred lines of the cross between the winter wheat (Triticum aestivum L.) variety Forno and the winter spelt (Triticum spelta L.) variety Oberkulmer. We determined the level of resistance of adult plants to leaf blotch (SNL) and glume blotch (SNG) as well as morphological traits for 2 years after artificial inoculation with S. nodorum. Using composite interval mapping and LOD > 3.7, we detected ten QTLs for SNG blotch resistance (six inherited from the susceptible parent Forno) and 11 QTLs for SNL resistance (four inherited from Forno) across 2 years. Both resistance traits were moderately correlated (r = 0.52) and had only one common QTL. For SNL resistance, seven QTLs were not associated with QTLs for morphological traits. Among them, QSnl.eth-2D, QSnl.eth-4B and QSnl.eth-7B3 had major effects (R(2) > 13%) and were potential candidates for marker-assisted selection. For SNG, the major QTL on chromosome 5A, explaining 36% of the phenotypic variance for resistance, was associated with the q locus conferring the spelt morphology (long lax ear, long culm and hard glumes). Only QSng.eth-1BS, which explained 7% of the variance for resistance to SNG blotch, was not associated with QTLs for morphological traits. The consequences for breeding programmes are discussed.

No abstract