Faba bean breeding for sustainable agriculture

During the European project EUFABA we have developed molecular maps in RILs populations of Vicia faba and located genes and QTLs controlling broomrape and ascochyta blight resistance. Besides, we have studied the stability of QTLs across diverse locations and genetic backgrounds to allow the identification and further development of specific markers (SCARs) for pyramiding and rapid screening. The first step involved two existing faba bean F6-7 populations segregating for broomrape and Ascochyta resistance. For the population Vf6 X Vf136 the linkage map was constructed with 278 RAPD, 6 SSR, 5 STS, 2 legumins, 1 SCAR and 21 ESTs. A total of 19 linkage groups were obtained, 16 of them consisting of three or more markers, covering 2811.23cM of the Vicia faba genome with a mean inter-marker distance of 13.51cM. A molecular map corresponding to the F6 population of cross 29H-Vf136 has been developed. The majority of the molecular markers already mapped in the F2 were genotyped in the corresponding F6 RILs to construct a skeletal genetic map of reference to be used for QTL analysis. The map consisted of 199 RAPD, microsatellites, seed protein genes and ESTs derived from Medicago truncatula. A total of 23 linkage groups were obtained, 18 of them consisting of three or more markers, covering 2204.7cM of the Vicia faba genome with a mean inter-marker distance of 14.2cM. Evaluation for resistance to broomrape performed in the RILs derived from cross Vf6 x Vf136 in two localities and years (Córdoba 2003, Cordoba 2004 and Mengíbar 2004) allowed the detection of 5 QTLs controlling the trait (Oc2 to Oc6). Four of these regions (Oc2, Oc3, Oc5 and Oc6) were found to be associated with O. crenata resistance in more than one trial and only one region (Oc4) in a single trial and two of them (Oc2 and Oc3) were consistent with those reported in the original F2. The Composite Interval Mapping (CIM) analysis in the same RILs population has revealed the existence of 2 QTLs associated to ascochyta blight resistance (Af1 and Af2). Af1, assigned to chromosome III, was detected in the RILs population explaining about 9% of the phenotypic variability of the trait. Additionally, Af2, located on chromosome II, was detected as well and explained about 14% of the trait variability. The analysis in the RILs population has revealed the conservation of this 2 QTLs (Af1 and Af2) previously reported (Roman et al. 2003). At these moment, the analysis for cross 29Hx136 is being completed Until now, we have considered the disease evaluation in growth chamber conditions with the isolate from Córdoba and the field evaluation in Córdoba 2006. More details in the article: Roman B, Satovic Z, Avila CM, Rubiales D, Moreno MT, Torres AM (2003) Locating genes associated with Ascochyta fabae resistance in Vicia faba L. Aust J Agric Res 54 (1): 85-90

Rapid methods for screening for drought tolerance. Drought is the major abiotic factor limiting faba bean yield in many environments. During the EU-Faba project, a range of germplasm was screened for drought tolerance. Plants were grown in pots and half of the replicates were subjected to increasing water stress, starting about 40 days after planting. Methods of evaluating transpiration response included stomatal conductance, relative water content, leaf temperature, carbon isotope discrimination, water use and transpiration efficiency. Biochemical response was measured by osmotic potential. Morphological markers for drought avoidance were specific leaf weight, tap root length, and root mass. Data were also compared with yield data from previous or concurrent field trials of the same materials. The measures of stomatal response were highly correlated with each other. Carbon isotope discrimination and stomatal conductance are robust methods for identifying drought-tolerant material when the plants are grown with adequate moisture supplies. Nevertheless, carbon isotope discrimination is expensive and stomatal conductance measurement is time-consuming. Leaf temperature, as measured by a non-contact infrared thermometer, provided an excellent balance of non-destructiveness, speed, low cost and high correlation with carbon isotope discrimination. For this method as well, only one plant treatment is necessary, namely growth at adequate moisture levels. In these conditions, the greater the leaf temperature, the more restricted the transpiration and the greater the conservation of water by the plant. Infrared thermometry can be used on large numbers of plants or plots and then the most interesting identified material can be verified by carbon isotope discrimination. In the experiments, the infrared thermometer was a Raynger ST60 ProPlus (Raytek, Santa Cruz, CA, USA) held at an appropriate angle about 10 cm above the leaf, focusing the laser point at the centre of the leaflet. This instrument is now the Fluke 66 (Fluke Corporation, Everett, WA, USA). The focus point may be achieved more precisely and the temperature determined more closely with a Fluke 570 series at greater cost. Infrared thermometry is already used to screen other crop species for drought tolerance and yield potential. This project has demonstrated its utility for faba beans.

Identification of sources of improved drought tolerance Drought is the major abiotic factor limiting faba bean yield in many environments. During the EU-Faba project, a range of germplasm was screened for drought tolerance. Plants were grown in tall, cylindrical pots 12 cm x 41 cm filled with 4 kg of potting mix (sand for root studies) and half of the replicates were subjected to increasing water stress for 12 days, starting about 40 days after planting. Methods of evaluating transpiration response included stomatal conductance, relative water content, leaf temperature, carbon isotope discrimination, water use and transpiration efficiency. Biochemical response was measured by osmotic potential. Morphological traits for drought avoidance were specific leaf weight, tap root length, and root mass. Data were also compared with yield data from previous or concurrent field trials of the same materials. Cultivar Mélodie and inbred line ILB 938/2 both presented good drought-tolerance attributes as measured by stomatal characteristics. They consistently demonstrated low stomatal conductance, water use and carbon isotope discrimination along with high leaf temperature and transpiration efficiency. These traits indicate that these two accessions maintain water in plant tissues through minimizing water loss under drought conditions, an established mechanism of dehydration avoidance. ILB938/2 was the only line which showed evidence of osmotic adjustment in response to moisture stress. Its osmotic potential was already more negative (i.e., more solutes in the cell) than other accessions in the well watered condition and it increased (i.e., still more solutes in the cell) during drought stress. Osmotic adjustment has not previously been found in faba bean and this demonstrates a further mechanism of drought tolerance in this line. Root length and total root dry matter varied widely in the germplasm set. Roots of line ILB938/2 were 20% longer than those of most other lines, and those of line L-7 and Apollo/1 were 15% longer. In addition, cultivar Enantia had almost twice the root dry matter of most other lines. Enantia and L-7 showed reasonably good stomatal traits as well. Thus ILB938/2 is a valuable source of dehydration avoidance by stomatal traits, drought response by osmotic adjustment and drought avoidance by deep rooting. Cultivar Mélodie also has good stomatal traits and much higher yield potential. Lines L-7 and Apollo/1 along with cultivar Enantia are sources of good rooting characteristics to provide drought avoidance.

RAPD markers were used to study variation among 19 species of the genus Orobanche: O. alba, O. amethystea, O. arenaria, O. ballotae, O. cernua, O. clausonis, O. cumana, O. crenata, O. densiflora, O. foetida, O. foetida var. broteri, O. gracilis, O. haenseleri, O. hederae, O. latisquama, O. mutelii, O. nana, O. ramosa, O. rapum-genistae and O. santolinae. A total of 202 amplification products generated with five arbitrary RAPD primers were obtained and species-specific markers were identified. The estimated Jaccard's differences between the species varied from 0 to 0.864. The pattern of interspecific variation obtained is in general agreement with previous taxonomic studies based on morphology, and the partition into two different sections (Trionychon and Orobanche) is generally clear. Nevertheless, the position in the dendrogram of O. clausonis did not fit the botanical classification since it clustered with members of section Trionychon. Within this section, our dendrogram shows greater differences between O. arenaria and the other members of the section: O. mutelii, O. nana and O. ramosa. Within section Orobanche all O. ramosa populations showed a similar amplification pattern, whereas differences among O. crenata populations growing on different hosts were found. Orobanche foetida and O. densiflora clustered together supporting the morphological and cytological similarities and the host preferences of these species. More detailed information in: Román B, Alfaro C, Torres AM, Moreno MT, Satovic Z, Pujadas A, Rubiales D (2003) Genetic relationships among Orobanche Species as Revealed by RAPD Analysis. Annals of Botany 91: 637-642

Artificial screening procedures were set up in glasshouses conditions which allowed reproducible ranking of susceptibility of genotypes: - an experimental procedure enabling to test the susceptibility of cultivars of faba beans towards Fusarium oxysporum inducing wilts. - an experimental procedure enabling to test the susceptibility of cultivars of faba beans towards Rhizoctonia solani. However, this level of susceptibility under well controlled conditions is opposite to the field behavior of these two varieties. We conclude the field damages should be explained by more complex pathogenic factors, still to be understood and that we have no final screening method available to breeders at the end of this work.

Brief description of recommended method for screening for Ascochyta fabae resistance 1.2.1 Field evaluations. Field screening is commonly used to evaluate faba bean germplasm against A. fabae. It is recommended to inoculate the field trial to assure a homogeneous distribution of the disease avoiding disease escape. This can be done by spraying plants with a conidial suspension of A. fabae (1-2 x 105 conidia per ml is enough) or by spreading infested barley seeds or debris at sowing. A susceptible control should be distributed throughout the trial. A first assessment of disease development should be made early, on young plants with about the fourth-fifth leaves completely expanded. A second can be done at flowering and a third on mature pod-bearing plants. Several scales can be used. Disease severity scale (DS) is based in the percentage of symptomatic area of the whole plant, while the qualitative scale Infection Type (IT) based in the lesion size and the presence or not of pycnidia. This scale is rather quick but not suitable for field observations when few lesions are present that could not easily be found. The evaluation could result in underestimation of rate-reducing resistance types resulting in low DS but where the few lesions visible are well developed. This scale might be very useful in growth chamber studies complemented with DS ratings. Other scales try to enclose both criteria. Thus, the 0-9 disease scoring recommended by ICARDA (International Center for Agricultural Research in the Dry Areas) is a combination of lesion type, lesion frequency and extent of damage, being more complete than IT alone, and also quick what makes it more adequate to be used in large scale breeding programs. However, we recommend to slightly modifying the ICARDA scale including among low ratings those representing relative little amount of well formed lesions to avoid discarding useful sources of resistance. 1.2.2 Growth chamber evaluation. Plants can be inoculated by spraying a suspension of conidia of the appropriate isolate of A. fabae (5x105 spores/ml). Plants should be incubated for 24 hours in darkness and 100% RH, and then kept at 20ºC with a 14/10 hour day/night photoperiod cycle in a growth chamber. Disease development can be evaluated fifteen days after inoculation with any of the scales indicated above. In addition to this, components of resistance such as lesion size, latency period and spore production can be measured for a more accurate description of partial resistance.

483 faba bean accessions were screened under field conditions for resistance against crenate broomrape (Orobanche crenata) scoring number of emerged broomrapes per host plant. Those accessions selected during the two first field season at Córdoba (Spain) were used to compose the Eufaba-broomrape-ringtest. These 41 accessions, together with a susceptible local check variety, were tested under field conditions at Córdoba, Spain, and Kafr-El Sheik, Egypt, in soil with a high and natural infestation with Orobanche crenata seeds. The same Eufaba-brommrape-ringtest, with two susceptible local checks, was also screened at Beja, Tunisia, under O. foetida infested field condition. Three complete repetition field design was used in all location. At crop maturity the number of emerged broomrape shoots per plant was recorded and referred to the average number of shoots of the four rows of the susceptible check surrounding each test row. Resistance against Orobanche resulted to be a multi-component factor that can act at any of the phases of the infection process: including induction of parasite germination, penetration/establishment and tubercle development. Resistant plants stop parasite infection and/or development in at least one of these steps. Contrary to previous results obtained with chickpea, low induction of germination does not seem to be a determinant factor in resistant faba bean. However, stoppage of parasite penetration through host tissues and hampering of tubercle development appear as the main available resistance responses found in faba bean. No escape due to low root density has been identified in faba bean at the moment. Histochemical studies have revealed more details about the nature of resistance to O. crenata in faba bean. It has been confirmed that the parasite intrusive cells are unable to penetrate into the host central cylinder and the parasite is stopped in both, the cortex or the endodermis. In the resistant faba bean, an accumulation of substances around the penetration pathway during incompatible interactions can be observed. Detailed studies with resistant and susceptible genotypes showed no significant differences inducing germination of O. crenata seeds (about 45% of germination). The percentage of germinated O. crenata seeds developing a tubercle was higher in the susceptible cultivar 'Prothabon' than in the resistant one 'Baraca' more than 30 times). In addition, the number of established tubercles per plant was 127.0 in the case of 'Prothabon' and only 3.4 in 'Baraca'. A darkening of the tissues around the attachment and penetration point was observed in some cases. The seedlings stopped their development, became dark and did not originate a tubercle. The percentage of these attached O. crenata seedlings unable to develop a tubercle was also determined for both genotypes, resulting in a higher percentage of aborted penetration attempts in 'Baraca' (68.5%) respect to 'Prothabon' (39.4%). Staining of historesin embeded samples with TBO revealed that the parasite intrusive cells are unable to penetrate into the host central cylinder and the parasite is stopped in both, the cortex or the endodermis. Moreover, an accumulation of dark blue stained substances around the penetration pathway during incompatible interactions was observed. Staining with AGS of paraffin embedded samples confirmed these results, and showed that the substances around the penetration pathway are of heterogeneous nature. No fluorescence from the substances or neighboring host cells and walls was detected. Neither observation under polarized light showed any special feature. Accumulation of callose around the penetration pathway of the parasite has been detected using aniline blue fluorochrome, and it appears as the mechanism responsible for stopping parasite penetration in the host cortex. The reinforcement of host cells walls by callose accumulation originates a characteristic bottleneck aspect of the parasite intrusive tissues. In case the parasite was not being stopped in the cortex, a lignification of endodermal cell walls in contact with parasitic tissues takes place. This barrier to O. crenata penetration into the central cylinder has been previously described during the incompatible interaction of O. crenata and vetch.

QLK5-2001-02307 (EUFABA) eTIP wlink@gwdg.de 03. 12. 2007 Title of the result: Assessment and analysis of the effect of hardening on the frost tolerance and the fatty acid composition of leaves and stems of a set of faba bean (Vicia faba L.) genotypes. This text contains data on the usefulness of the fatty acid composition of leaves and stems of juvenile faba bean plants to be used as auxiliary traits to genetically improve frost tolerance and eventually winter hardiness of this grain legume. With these results, breeding becomes more efficient. Category of the result is "A" Partners (result owners) involved: These results were acquired at the Göttingen Institute in the group of W. Link (WP3). These results are data. This group is the owner of the results. The results are by now under publication, a modified version of the text given here will be published in 2008 in the journal EUPHYTICA, thus all these results are available for free use by everybody. If any genetic material mentioned here is considered as "result", the owner is again the group of W. Link (representing the University at Göttingen). Small quantities of such material can be asked for and will be available for free use to everybody. PhD Mustapha Arbaoui, Prof. Dr. Wolfgang Link Actual Summary: Frost tolerance is a main component of winter-hardiness and improving it would promote faba bean (Vicia faba L.) cropping in cool-temperate regions. In many species, leaf fatty acid composition was found to be related to frost tolerance. The objective of this study was to determine, in a representative sample of genotypes, the effect of hardening on leaf and stem (1) frost tolerance and (2) fatty acid composition, and to seek correlations between them. First leaf, second leaf, and stem of 31 faba bean genotypes were analyzed after hardening and without hardening. High frost tolerance of known winter genotypes and several experimental lines was shown. Hardening had a significant, positive effect on frost tolerance of all three organs. Stems were on average more frost tolerant than leaves. Hardening induced significant changes in the fatty acid composition: oleic acid decreased significantly in leaves by 3.24% and in stems by 1.77%, whereas linolenic acid increased in leaves by 6.28% and in stems by 9.06%. In stems, correlations between frost tolerance and fatty acid composition were not significant. Correlation coefficients strongly indicated that non-hardened oleic acid content, changes in oleic acid and in linoleic plus linolenic acid content in leaves partly explained their frost tolerance ( 0.347; p<0.1 < |r| < 0.543; p<0.01). The results corroborate the importance of using genetic differences in the fatty acid metabolism in breeding grain legumes for frost tolerance.

1.1. Brief description of recommended method for screening for rust resistance (more extended description in a cooperative review paper published in Euphytica) 1.1.1. Field screenings To screen for resistance under field conditions, infection should be uniform and high enough to avoid escapes. In some locations, natural rust epidemics are so frequent that no artificial inoculations are needed. However, to ensure high and uniform disease distribution, artificial inoculations are highly recommended. Artificial inoculation might consist of spraying with aqueous suspension of rust spores or dusting mixture of spores in an inert carrier such as pure talc powder or Lycopodium spores. Plants should be inoculated after sunset to benefit both from the darkness and the high relative humidity of the night. Plots can be irrigated with microsprinklers to maintain high relative humidity. The assessment of quantitative rust resistance under field conditions can be accomplished in various ways. When rust development starts, disease severity can be assessed at two-week intervals, by a visual estimation of the leaf area covered with rust pustules (disease severity, DS). Resistance of accessions can be compared by the last measure of the disease severity (the final disease severity), by the area under the disease progress curve (AUDPC) or by epidemic growth rate (r). The means of the observed AUDPC values are frequently converted into relative values and expressed as a percentage of a susceptible check. In addition to the amount of infection, the infection type (IT) must also be assessed. The IT describes the reaction of the plant to the infection in terms of amount of necrosis or chlorosis at the infection sites and the rate of sporulation of the individual colonies. 1.1.2. Controlled conditions screenings Testing under controlled conditions can be carried out in either a growth chamber or a greenhouse, allowing screening in seedling and adult plant stages. This type of experiments allows effective control of the environmental conditions, and is fundamental for testing different isolates. Furthermore, for large scale screening, especially at the early stages of breeding programmes, controlled environment tests are likely to be more efficient than field testing. Rust inoculation can be done by dusting the plants with urediospores diluted in an inert carrier. For quantitative measurements the use of a spore setting tower is recommended in order to ensure a more uniform spore deposition. Plants are then incubated for 24 hours in an incubation chamber at 20oC in complete darkness at 100% relative humidity, and subsequently maintained in a growth chamber at 20oC with a 14-hour photoperiod. Several macroscopic components of resistance can be measured under controlled conditions: Latent Period (LP) is the period of time between inoculation and sporulation of 50% of the pustules. LP is determined by counting daily the number of pustules visible in a marked area on the leaves, using a pocket lens, until the number of pustules no longer increases. Infection Frequency (IF) is the number of pustules per unit area, which can be calculated in the same area in which LP was estimated. Infection Type (IT) describes the external appearance of the pustules, as have been described above. Colony Size (CS) can be measured microscopically with the help of a micrometer from four to ten days after inoculation, using epifluorescence or contrast phase microscope, or can also be estimated macroscopically in the basis of the pustules diameter. All these components of resistance contribute to the epidemic-retarding effect of partial resistance. The presence of host cell necrosis associated to the infection units is indicative of hypersensitive response. A prolonged LP seems to be the most influential component in the epidemic process. Reduced IF, spore production (number of spores produced per pustule and day) or duration of sporulation (number of days of spore production per pustule) are other important components of resistance. These parameters expressed in polycyclic situations increase the differences between susceptible and resistant lines. 1.1.3 Detached leaves assays A method for testing faba bean rust in detached leaves has been proposed by Herath et al. (2001), excising and maintaining them in Oasis water-retaining medium, with 5ppm gibberelic acid, in a enclosed box in a temperature-controlled glasshouse. A different method has been used in Petri dishes with tissue paper with distilled water to which benzimidazole (1%) was added. Faba bean leaflets are excised and carefully laid, adaxial surface up, on the tissue paper, and then inoculated. Petri dishes are incubated and maintained as described above.

QLK5-2001-02307 (EUFABA) eTIP wlink@gwdg.de 03. Dec. 2007 Title of the result: Study of faba bean (Vicia faba L.) winter-hardiness and development of screening methods. Results contain data on winter-hardiness of a set of faba bean genotypes and on the development of screening methods (frost resistance, winter-hardiness). Objective: Genetically improve frost tolerance and winter hardiness. In cooperation with NPZ Lembke company (Dr. Olaf Sass, Hohenlieth), these results were and are used to breed improved winter bean cultivars, the prospects of a release of such cultivars in the near future are good. With these data, breeding can be addressed to more appropriate material and gain from selection can be increased. Sources of superior winter hardiness and frost tolerance were the following genotypes: Hiverna/2, Hiverna, Karl, Bulldog/1, GöttWAB, F7-(Cor1xBPL)-95. Small quantities of the material can be asked for and will be available for free use to everybody. Category of the result is "A" Partners (result owners) involved: Results were acquired in cooperation with EUFABA-partners M. Hybl (Sumperk), L. Narits (Jogeva), G. Duc (Dijon), J. Winkler (Gleisdorf), O. Sass (Hohenlieth) at the Göttingen Institute in the group of W. Link (WP3). Results are data & seeds of genotypes. The Göttingen group is the owner of the results. Results are published (journal FIELD CROPS RESEARCH, 2008); all these results are available for free use by everybody. As to any genetic material identified here which may be considered as "result", the owner is the group of W. Link (representing the University Göttingen). Authors: Mustapha Arbaoui, Christiane Balko, and Wolfgang Link (C. Balko was not member of EUFABA) Abstract In cool-temperate regions, faba bean (Vicia faba L.) is mainly grown as a spring crop despite the higher yield potential of the winter type, because of the insufficient winter-hardiness of the present winter genotypes. The objective of this study was to assess winter-hardiness and frost tolerance, to quantify the hardening effect on physiological traits, and to identify auxiliary traits for winter-hardiness. To do so, 31 representative entries were tested in controlled frost tests for frost tolerance and in 12 European environments for winter-hardiness. Total fatty acid composition, proline content, and electrolyte leakage of leaves were analysed. Across all environments, five European winter genotypes were identified with superior winter-hardiness. Controlled frost tests indicated that frost tolerance is a significant, but not an exhaustive component of winter-hardiness (0.021<|r|<0.737**). These tests revealed the high frost tolerance of several poorly winter-hardy experimental lines and the limited frost tolerance of well known winter types. Fatty acid changes due to hardening, proline content, and electrolyte leakage were more strongly correlated with frost tolerance than with field based winter-hardiness. Although frost tolerance, fatty acid composition, proline content, and electrolyte leakage were significantly correlated with winter-hardiness, the rather low correlation values do not allow a general use of one of them alone to indirectly select for winter-hardiness.

Uromyces viciae-fabae (Pers.) J. Shört. is the causal agent of faba bean rust, a disease worldwide distributed that can cause up to 70% yield losses particularly when the infection starts early in the season. The development of resistant cultivars seems to be the most practical and cost-efficient method for the control of diseases affecting grain legumes. Several faba bean sources showing two different types of incomplete resistance to U. viciae-fabae have been reported. On one hand, resistance is expressed as a reduction of disease severity without any macroscopically visible necrosis. On the other hand, hypersensitive resistance to faba bean rust has been recently described as an incomplete resistance associated with late-acting necrosis of the host tissue resulting in a reduction of the infection type. Both types of incomplete resistance only differ in the presence or absence of macroscopically visible necrosis. Concerning the genetic basis of hypersensitive resistance Sillero et al. (2000) suggested the monogenic inheritance of the character. Bulk segregant analysis was used to identify RAPD markers linked to a gene determining hypersensitive resistance in Vicia faba line 2N52 against race 1 of the rust fungus Uromyces viciae-fabae. The monogenic nature of the resistance was determined by analyzing the F2 population from a cross between the resistant line 2N52 and the susceptible line VF-176, and further confirmed in the F2:3 derived families. Linkage of the RAPD markers was confirmed by screening 55 F2 plants segregating for resistance. Three RAPD markers (OPD13736, OPL181032 and OPI20900) were mapped in coupling phase to the resistance gene for race 1 (Uvf-1). No recombinants between OPI20900 and Uvf-1 were detected. Two additional markers (OPP021172 and OPR07930) were linked to the gene in repulsion phase at a distance of 9.9 and 11.5 cM, respectively. The RAPD marker OPI20900 was transformed into SCAR marker and we obtained a complete association between the presence of the SCAR I20 and the resistant genotypes. The manuscript describing the process and further applications is being written at present.

Faba beans (Vicia faba L.) have a great potential as a protein-rich fodder crop, but anti-nutritional factors such as condensed tannins reduce the biological value of their protein. Tannins can be removed from seeds by any of the two complementary genes, zt-1 and zt-2, which also determine white flowered plants. The less common gene, zt-2, is also associated with increased protein levels and energy values and reduced fiber content of the seeds. To identify a cost-effective marker linked to zt-2, we have analyzed a segregating F2 population derived from the cross between the colored flower and high tannin content genotype Vf6 and a zt-2 line. By using Bulked Segregant Analysis (BSA), 5 RAPD markers linked in coupling and repulsion phase to zt-2 were identified and their conversion into Sequence Characterized Amplified Regions (SCARs) was attempted. Amplification of the SCARS was more consistent, although the initial polymorphism was lost. Restriction digestion of SCAR SCAD16589 with AluI (SCAD16-A), Bsp120I (SCAD16-B) and HinfI (SCAD16-H) revealed clear differences due to the amplification of different loci. The consensus sequence of these CAPs (Cleavage Amplification Polymorphisms) markers allowed discrimination of three bands from which two new forward SCAR primers were developed based on specific sequences from zero tannin and high tannin content genotypes. To improve the efficiency of the marker screening, a multiplex PCR was developed that allowed the simultaneous amplification of the SCAR with the same advantages as a codominant marker. Marker validation was carried out with a new F2 population segregating for flower color and tannin content, underscoring the potential of these markers in breeding selection to introgress the zt-2 gene for the development of new tannin free faba bean cultivars. More detailed information in the publication: Gutierrez N, Avila CM, Moreno MT and Torres AM (2008). Development of SCAR and CAPS markers linked to zt-2, one of the genes controlling absence of tannins in faba bean. Australian Journal of Agricultural Research, in press.

In collaboration with the INRA Institute in Tours, France (led by M. Lessire), RBCs were isolated and ghosts prepared from 3 groups of chicken fed for three months with: a) 25% faba-bean rich in V/C; b) 25% faba-bean poor in V/C; and d) control standard diet without V/C (soybean meal). At the end of each month (for 3 months), blood was collected from each experimental group, ghosts immediately prepared, deep-frozen and stored at 80°C for further use. Ghost proteins from the retrieved ghosts were separated by mono- and bi-dimensional PAGE. The first dimension was run on strips with pH gradient from pH 3 to pH 10. The second dimension took advantage of molecular sieve separation. Gels were silver-stained and after re-run stained with colloidal Coomassie. After image analysis to identify those spots where changes in positioning/protein amount had occurred, relevant spots were excised and analyzed by MALDI-Tof (Micromass equipment). Taking advantage of the recently disclosed full genome sequence of Gallus gallus proteomic analysis of laying hen RBC membrane proteins was performed for the first time. Comparison of relative amounts of membrane proteins modified by diet has shown distinct changes in spectrin alpha chain (relative densitometric units: control, 695; high V/C, 1971; low V/C, 1187), actin (relative densitometric units: control, 228; high V/C, 7600; low V/C, 17900), heat shock protein 70 (relative densitometric units: control, 1249; high V/C, 6474; low V/C, 852). In separate experiments, RBC membranes were labelled with fluorescein 5-maleimide, a redox-sensitive label that allows discrimination of SH-oxidized membrane proteins. A characteristic pattern was evident in laying hens fed with high V/C diet.

Faba beans are inexpensive, nutrient-dense sources of plant protein, but anti-nutritional factors such as condensed tannins reduce the biological value of their protein. Two recessive genes, zt-1 and zt-2, control the absence of tannins in faba bean seeds and also determine a white flower character on the plant. However, crosses between them produce colored F1 plants with tannins that contaminate the crop. Therefore, it is important to identify the gene present in all tannin-free cultivars and gene bank accessions to enable breeders to choose appropriate genitors for their crosses. The aim of this study was the identification of markers linked to zt-1, one of the genes governing free tannin content in faba bean. A segregating F2 population derived from the cross between the colored flower and high tannin content genotype Vf6 and a zt-1 line was developed and characterized phenotypically. Bulked Segregant Analysis (BSA) was used to identify Random Amplified Polymorphic DNA (RAPD) markers linked to the zt-1 gene. Four RAPD loci (OPC5551, OPG15600, OPG111171 and OPAF20776) showed polymorphism between the contrasting bulks. The markers were sequenced to develop specific Sequence Characterized Amplified Regions (SCARs). Amplification of SCC5551 produced a single product which was only observed in the white flowered and zero tannin content genotypes, whereas SCAR SCG111171 only produced a band in F2 plants with colored flower and high tannin content. SCARs SCC5551 and SCG111171 were tested for their applicability for routine screening in 37 faba bean genotypes differing in flower color and tannin content. SCC5551, allowed the prediction of the zt-1 genotypes with a 95% of accuracy, underscoring the potential of this SCAR marker as a cost-effective tool for MAS More detailed information about these markers is available in the following publication: - Gutierrez N, Avila CM, Rodriguez Suarez C, Moreno MT, Torres AM (2007). Development of SCAR markers linked to a gene controlling absence of tannins in faba bean. Molecular Breeding, (2007) 19:305-314

Crosses between lines carrying desirable traits (abiotic and biotic stresses and nutritional quality) with lines of agronomic interest adapted to the different regions involved in the projects have been performed. Seeds of genotypes used in WP5, carrying favorable alleles for resistance to Ascochyta (29H), hypersensitive resistance to rust (2N52), quantitative resistance to rust (BPL261) and Botrytis (Sel. 97 Lat.97 135) were sent to P6 and crossed with 7 varieties of agronomic interest adapted to the different EU regions (Maya, Hobbit, Alameda, Wizzard, Marcel, Baraka and Target). Different F2 populations were obtained (Tables 1 and 2). The most abundant progenies have been distributed among partners interested in these pre-breeding materials (mainly breeding companies) to be used in their respective breeding programs.

Our study allowed having an overview on different parameters which could interfere in the quantitative analysis of vicine and convicine in a seed sample. We have demonstrated that the HPLC analysis is the most reliable method and relatively easy to perform. In order to give a value close to the real content of vicine in a sample, we recommend to start vicine extraction from flour and not by hot dialysis. Nevertheless, we showed a good correlation between both extraction methods. We determined the size of seed samples required. The written procedure is available on request from P. Dulieu RD Biotech.

The efficiency of SCARs and CAPs developed in the project and linked to genes, zt1, zt2 and zvc was validated in a collection of 37 accession provided by INRA-Dijon some of which accessions were carrying those genes. Table 1 List of inbred lines faba beans tested in this study. European inbred lines Flower color Tannina Markersb SCC5551 SCG111171 AD23 MAINTENEUR 2300 Colored T - + G58 MAINTENEUR 2302 Colored T -+ GLORIA 2308 White ZT1 + + 19 TB -T 2316 White ZT1 + + DIVINE 44.2 2391 Colored T - + BLANDINE 2073 White ZT1 + + POLLEN 2074 White ZT1 + + 19 TB T 2317 Colored T - + FABIOLA -T 2318 White ZT1 + + FABIOLA T 2319 Colored T - + POUILLY 2327 Colored T - + DIVA 2366 Colored T - + DISCO 2390 White ZT2 - + MAXIME 196 Colored T - + AQUITAINE 267 Colored T - + MARAIS POITEVIN 276 Colored T - + GERS 277 Colored T - + LORRAINE 279 Colored T - + PICARDIE 437 Colored T - + DIANA 455 Colored T - + WIERBOON 704 Colored T - + MIKKO 1179 Colored T - + TICOL 1191 Colored T - + STRUBE 1216 Colored T - + OPTICA 1482 White ZT1 + - BOURDON 1505 Colored T - + WEBO 1508 Colored T - + TROY 1579 Colored T - - CÔTE D'OR 1626 Colored T - + HG115 1757 Colored T - + ASCOTT 1777 Colored T - + TALO 1795 Colored T - + SORAVI 2070 Colored T - - MELODIE M 2393 Colored T + + LADY 2401 Colored T + + MAYA 2077 Colored T - + PUNCH 2067 Colored T - + a T: genotype with tannin. ZT1: genotype with zero tannin content carrying the zt-1 gene. ZT2: genotype with zero tannin content carrying the zt-2 gene. b + and - indicate presence and absence of the SCAR bands, respectively. The key for the efficiency of MAS lies in the close linkage between the marker and the target gene. In the case of SCC5551, the lower genetic distance to the flower color (9.7 cM) leads to an efficient selection for zero tannin content, since only in very few cases (2 out of 37) the marker allele was separated from the desired trait by a recombination event. Thus, SCC5551, allowed the prediction of the zt-1genotypes with a 95% of accuracy, underscoring the potential of this SCAR marker as a cost- effective tool for MAS in large faba bean breeding populations. In case of zt-2, the identification and further development of markers linked to this gene vas performed in the F2 progeny derived from the cross Vf6 x zt-2. To asses the ability of the new SCAR markers to distinguish contrasting phenotypes in different genetic backgrounds, a new F2 population was further analyzed. The progeny (n = 21) was derived from a cross between a white flowered faba bean variety (Disco 2390), carrying the gene zt-2, and the colored flower variety Wizzard. The segregation for the trait gave a good fit to a 3:1 ratio (15 colored white flowered plants). Five of the six tannin-free individuals were unambiguously identified by the SCARs, which underscores the high selection efficiency of these markers (83%) in predicting the trait in related families.

In the first year of EUFABA project, 484 Vicia faba accessions randomly selected were screened for resistance to Asochyta fabae in the field at Cordoba. Seventy-two accessions were selected with a damage area on leaves lower than 15% and a value in the pod scale lower or equal to 2 (considered as the limit for resistance in these organ plant). Out of these, the best 31 were selected and multiplied to constitute the Eufaba-ascochyta-ring test to be distributed to other partners for multi-location field trials during year 2. The best 20 genotypes selected during the first and second year were evaluated in different environmental conditions and against different isolates of the pathogen. Out of these 20 genotypes, 15 maintained good level of resistance with disease scores lower than 10% on leaves. Considering all results, we conclude that the search of new sources of resistance against the fungus A. fabae is needed in each crop area since differences in isolate aggressiveness has been detected among different countries (for example infection levels in the Czech Republic were higher than in Spain during the last two years). In addition, a degree of interaction between the genotype resistance and the isolate has been detected what suggest the existence of pathogenic variability in the pathosystem. We have optimized a method to visualize the development of the fungus within the plant using fluorescent microscopy. This would allow studying the first infection phases. At present we are performing a study to investigate the differences in this process between a susceptible and two different resistant entries. We selected two resistant entries since they seem to show different mechanisms of resistance (genotypes 29H and L-8).

In the first year (2002/03) 288 accessions were screened for resistance to B. fabae in the field at Cordoba. Based on Disease Severity scores, the 39 most resistant accessions were selected, multiplied to compose the Eufaba-Botrytis-ringtest, and distributed for multilocation field testing during 2003/04. The trial was also established at Kafr El-Sheik, Egypt. Reliable scored were obtained only in Spain and Egypt. The ringtest was also established in year 3. 50 accessions were distributed to partners P1, P3, P5, P11, and P13 for the multi-location trials in the season 2004/05. Two additional sites were located in Egypt and Tunisia. Multi-location analysis showed a significant interaction between location and genotype. This could be explained by the different levels of disease reached in each trial: chocolate spot is very sensitive to weather conditions which differed greatly across trial sites; moreover, it may be assumed that B. fabae populations in each location might differ in virulence. It could be expected that the less resistant lines have performed better than they have done under more severe pathogen infestation, which would account for this. However, when results from France were removed from the analysis, the interaction disappeared for the other three locations. As the French trial had a moderate level of disease and not the lowest one, it appears that the interaction cannot be explained solely on the grounds of disease pressure, and that the influence of the pathogen isolate should be taken into account. This possible isolate-accession interaction ought to be further studied. In any case, the lack of interaction between location and genotype in three of the trials indicates that resistance against B. fabae in this collection is quite stable. As for the multi-year analysis in Spain, a significant interaction between season and genotype was found. In this case, it can be explained by the different level of disease in both trials. Season 2004/05 was particularly dry in the south of Spain, which resulted in low infection by B. fabae. Under these conditions, differences in levels of resistance among accessions were smaller. Since significant differences were detected among accessions in both analyses, a selection of the best genotypes is proposed. Though the response of some accessions may vary from one location to another, it is possible to suggest a list of 10 accessions which are found in all trials among the top 20 performers. They can provide good sources of durable resistance for breeding programs. Immunofluorescence microscopy was carried out to study the interactions between Botrytis fabae and broad bean (Vicia faba) at the microscopic level. Three different genotypes of V. faba were used: BPL 710, BPL 261, and ILB 365, their responses to the pathogen being, respectively, resistant, partially resistant and susceptible. The detached-leaf test was used for these experiments. Antibodies raised against antigen BC-12.CA4 of Botrytis cinerea (kindly provided by Dr. Frances M. Dewey, Oxford University) labeled successfully hyphae of Botrytis fabae. Subsequent immunofluorescence studies showed the development of the pathogen on the surface of the leaves as well as beyond the penetration point. According to the results, the different reactions of the genotypes to B. fabae might be explained by the differential ability of the pathogen to grow in them. The amount of hyphae present in the resistant genotype 24 hours after inoculation is very small. In many cases, no hypha was found at all in this genotype. This amount increases in the partially resistant genotype, and reaches its maximum in the susceptible one. In this case, big clusters of hyphae cover an important portion of the inoculated surface.

Identified goals of the faba bean breeding programs Yield potential for current faba bean cultivars in Europe is not far from 8.0 to 9.0 t.ha-1. It is not a limiting factor anymore like it used to be in the 70's. At the moment, breeders are making efforts on yield stability and seeds quality. 1) Seed characteristics: High variability exists for seed size and protein content depending on winter or spring types. Breeders try to introduce the zero tannin character or both zero tannin character and a low level of vicine-convicine into high yielding genotypes. 2) Plant characteristics: all attempts at architecture modification are directed toward a similar genetic ideotype: greater compactness, synchrony in reproductive development and reduced vegetative growth. 3) Biotic stresses: Several lines of resistance have been observed in the past in different programs for the main faba bean diseases. They are still used in the breeding programs. 4) Abiotic stresses: Breeders put principally the accent on frost hardiness and drought tolerance through earliness at flowering. D Breeding schemes Varieties currently sold in Europe are only synthetics populations. Marker Assisted Selection is used in EUFABA project for hastening faba bean breeding and satisfy both users and producers expectations. Conclusion In many countries, unstable yields result in unstable areas grown and this in turn results in inadequate grower experience in relation to the husbandry of the crop. This leads to crops being grown with little or no knowledge of optimum techniques. Genotypes adapted to specific environments, when grown with appropriate husbandry and care, show an impressive yield potential. Hence, it is clears that synergies exist between EUFABA and GL-Pro1 projects. These two programs should combine their efforts in order to improve farmers and users perception of faba bean crop in Europe.

Brief description of recommended method for screening for broomrape resistance 1.4.1 Field testing It is crucial to ensure uniform distribution of the parasite seeds in the soil to prevent selection of genotypes that merely remain unchallenged. It is recommended to sow a susceptible cultivar the year before the experimentation. The upper layer of the soil should be cultivated several times after harvest in order to ensure a more uniform distribution of seeds across the whole field. For small scale tests, the plots can be artificially inoculated mixing parasite seeds with sand and applying them to the row with the crop seeds when sowing. However, escapes can not be precluded, and each test row should be surrounded by rows of a susceptible and vigorous check to be used as a reference for each accession. 1.4.2 Pot testing Despite these safeguards field screens are probably most efficient for discarding susceptible accessions. Carefully conducted pot trials are needed to confirm that accessions remaining uninfected in the field are truly resistant. Pot methods allow control over the environment, the inoculum density and its origin. Several substrates can be used mixed 30-40 mg O. crenata seeds/kg substrate to screen chickpea for broomrape resistance. Pot trials are usually located in a glasshouse. 1.4.3 In vitro testing Various techniques have been used to allow close observation of the germination, attachment and early development of parasitic weeds. However none of these methods unfortunately worked nicely for faba bean, perhaps due to susceptibility of the roots to oxygen depletion. Thus, an alternative screening method for broomrape (Orobanche crenata) resistance under growth chamber conditions has been developed by P1. It involves growing host and parasite in sand or vermiculite held in the gap between two glass sheets. Two cork strips of 0.5 cm thickness are placed between both glasses in left and right sides, and the lower side is sealed with a porous material (foam rubber, sponge) that allows nutrient solution to penetrate. Seeds are placed on the sand in the upper side of the plates. This method allows to grow plants of big size like faba bean or chickpea using glasses of 50 x 30cm, and also to follow the roots spatial distribution and the development of the broomrapes at different depths.

Calibration for protein content was established on NIRS 6500 (Foss, France) equipment and with a range of variability of genotype of 19.65% to 38.66% of seed DM: Calibration equation had following parameters R²= 0.932, 1-VR=0.880, SEC=0.687, SECV=0.920, SLOPE= 0.986 The validation of these calibrations on 100 distinct samples gave the following performance: slope=0.958; R²= 0.870; SEP=0.778; (Bias)= -0.026; SEPC=0.781 Conclusion: This calibration of adequate precision for breeding activity is available on request from M. Huart , INRA Dijon To use the calibrations in future, all the apparatus should be standardised for whole faba bean seeds by Gembloux Agronomy Research Centre (Belgium).

QLK5-2001-02307 (EUFABA) eTIP wlink@gwdg.de 03. December 2007 Title of the result: Quantitative trait loci of frost tolerance and physiologically related trait in faba bean (Vicia faba L.) These results contains data on frost tolerance and physiological traits of a set of so-called recombinant inbred lines of one cross of two faba bean lines and on the several resultant QTL detected herein. Furthermore, the prospects of the use of these results in a marker-assisted selection program are elucidated. Marker-assisted selection renders breeding less dependent from appropriate natural frost and winter events. The objective is to genetically improve frost tolerance and winter hardiness of this grain legume, and to do so more efficiently. Category of the result is "A" Partners (result owners) involved: These results were acquired in cooperation with EUFABA-partners A. Torres, jointly at Córdoba and at Göttingen (WP3). These results are data and seeds of genotypes. The Cordoba group and the Göttingen group are the owners of the results. The resulting data are by now under publication, a modified version of the text given here will presumably be published in 2008 in the journal EUFPHYTICA, and a short abstract has been published in the AEP Congress 2007 at Lisboa, thus all these results are available for free use by everybody. As to any genetic material identified here which may be considered as "result", the owners are still the groups of A. Torres and of W. Link (representing the Universities of Córdoba and of Göttingen). Small quantities of such material can be asked for at Göttingen and will be available for free use to everybody. Authors were: Mustapha Arbaoui, Wolfgang Link, Zlatko Satovic & Ana-Maria Torres. Satovic was not member of EUFABA Actual Summary In faba bean, field based winter-hardiness is a complex trait that is significantly correlated to frost tolerance. Frost tolerance could be used to indirectly select for faba bean winter-hardiness. The aim of this study was to identify putative QTL associated with frost tolerance and auxiliary traits and to quantify the efficiency of marker assisted selection (MAS) as compared to classical phenotypic selection (CPS). To do so, 101 recombinant inbred lines (RIL) derived from the cross between two frost tolerant lines (the European line Côte d Or/1 and the exotic line BPL 4628) were tested for their hardened and unhardened frost tolerance in controlled conditions and for their fatty acid content before and after hardening. The LOD threshold to declare putative QTL was fixed according to the Bonferroni correction. Cross-validation (CV) was additionally performed in order to assess the unbiased genotypic proportion explained. Significant differences among the RIL were observed for all studied traits. For frost tolerance, five putative QTL were detected; three for unhardened frost tolerance that explained 40.7% (8.6% after CV) of its genotypic variance and two for hardened frost tolerance that explained 21.8% (1.0% after CV). For fatty acid content, three QTL were detected for oleic acid content in unhardened leaves that explained 62.9% (40.6% after CV) of its genotypic variance. This fatty acid was significantly correlated with unhardened frost tolerance. The unbiased genotypic variance explained (after CV) enabled to draw realistic prospects of MAS for frost tolerance. Thus, combined MAS was more efficient than CPS alone. Such efficiency was expected to be even higher on large populations at early generations. Moreover, favourable alleles inherited from the exotic line BPL 4628 could be introgressed to European winter-hardy beans for further improvement.

484 faba bean accessions were screened for rust resistance in year 1 under field conditions at Córdoba. In addition to this, the whole collection (484 accessions) was tested under field conditions at Kafr-El Sheik, Egypt. In Córdoba 17 extra accessions were tested in addition to the ringtest. Those accessions selected during the two first field season at Cordoba (Spain) were used to compose the Eufaba-rust-ringtest. These 49 accessions were distributed for multi-location trials with 3 replications. In addition, this ringtest was kindly tested under field conditions at Kafr-El Sheik, Egypt, and at Beja, Tunisia. Lack of durability of resistance is especially a problem for airborne fungal pathogens like rusts. These pathogens share a rather similar infection process that can be analyzed systematically. The process can be subdivided into spore deposition, spore germination, appressorium formation, penetration of stomata, formation of the first haustoria or intracellular hyphae, colonization, spore formation and spore release. We found limited varietal differences in germination and germtube directional growth suggesting that reduction of urediospore germination and fungal development on the leaf surface are of marginal importance. Faba bean rust appeared to be rather inefficient in finding stomata, but genotypic variation for degree of stoma recognition was not found. Formation of misplaced appressoria by the faba bean rust was usually high, but rather uniform among genotypes, about 20% of the germlings, thus offering little opportunities for breeding. We found some variation for failed stoma penetration ("stomatal exclusion"). In some accessions a proportion of substomatal vesicles were formed outside the leaves instead of inside the substomatal cavity. However, these differences, although significant, were too small to expect substantial epidemiological effects. We found incomplete resistance against rust to be common in faba bean, but also found hypersensitive resistance. This form of incomplete hypersensitive resistance is associated with late-acting necrosis of the host tissue resulting in a reduction of the infection type. Both types of incomplete resistance are associated with an increased latent period, a reduction in colony size and a decreased infection frequency. They only differ in the presence or absence of macroscopically visible necrosis. Six faba bean genotypes differing in the response to rust were selected for cellular studies. The experiment was carried out on detached faba bean second leaves. Inhibition of CAD reduced penetration resistance of lines 2N-34, 2N-52, V-1271 and V-1272 showing an important role for the phenylpropanoid pathway in resistance in these lines. This was further supported by the inhibition of the hypersensitivity response that characterized these lines. On the other hand D-mannose had no or few effect in reducing resistance. By contrast CAD inhibition did not affect the penetration resistance of the line BPL-261 characterized with partial resistance not associated with hypersensitivity. Interestingly, application of D-mannose inhibited the penetration resistance of this line. Biochemical characterization of resistance against U. viciae-fabae was also approached by analyzing polyphenoles and related enzymes (peroxidases) in lines BPL-261 (with partial resistance), V-1273 (with hypersensitivity resistance) and VFM-176 (highly susceptible), 2, 6 and 12 days after inoculation with U. viciae-fabae (with non-inoculated plant as control). There were no differences in the content of total phenolic compounds before infection between resistant and susceptible lines. The amount of phenolic compounds increased in resistant lines but not in the susceptible one after infection. Thin layer chromography (TLC) and high performance liquid chromatography (HPLC) analysis were used to detect changes in the phenolic profile. Three unidentified fluroscent compounds were only detected in resistant lines, showing Rf values of 0.11 (UN1), 0.17 (UN2) and 0.23 (UN3) respectively. The content of these compounds increased after inoculation. We suggest that UN1, UN2 and UN3 could be used as markers to select rust resistance lines from big collection of faba bean. Soluble as well as ionically bound peroxidase activities were extracted and assayed by using scopoletine, guayacol and O-dianisidine as substrates. Ionically bound peroxidase, increased in resistance lines after infection, particulary in line V-1273 (hypersensitivity resistance). These results indicate that phenolic compounds and peroxidases activities may play an important role in the mecanism of faba bean defence against U. viciae-fabae.

Two experiments were performed in laying hens using diets with 200g/kg and 250g/kg incorporation rate of Faba bean which contained high or low amounts of vicine and convicine. Significant difference was observed for the mean egg weight in the two experiments. Birds fed the V+C+ diet produced lighter eggs (60.4g) than those fed the control and the V-C- diet (62.1 and 61.6g). A kinetic study (third experiment) detected a depressive effect of vicine-convicine as soon as one week after giving the birds' fababean containing diet.

It was expected that high V/C eliciting increased redox sensitivity (see above results) could also induce modifications in RBC membrane proteins. A method to isolate chicken RBCs and prepare suitable ghosts was worked out. Laying hens RBCs are nucleated and isolation of ghosts without nuclei contamination was not trivial. A satisfactory protocol that allowed separation of Hb-free and DNA-free ghosts was established. A method to separate RBC membrane protein bands by mono- and bi-dimensional PAGE was set up. The presence of contaminating DNA in the RBC ghosts made electrophoretic separation of bands unsatisfactory. Further addition of DNAase to the samples proved important and allowed good separation of bands. The pattern was quite different from that of human RBCs and was characterized by distinctly low amounts of spectrin and band 3 (confirmed by Western blot with anti-spectrin and anti-band 3 mAbs), and by the appearance of new, still not identified bands in the low-molecular weight region of the gel. In collaboration with the INRA Institute in Tours, France (led by M. Lessire), RBCs were isolated and ghosts prepared from 3 groups of chicken fed for three months with: a) 25% faba-bean rich in V/C; b) 25% faba-bean poor in V/C; and d) control standard diet without V/C (soybean meal). At the end of each month (for 3 months), blood was collected from each experimental group, ghosts immediately prepared, deep-frozen and stored at 80°C for further use. Ghost proteins from the retrieved ghosts were separated by mono- and bi-dimensional PAGE. The first dimension was run on strips with pH gradient from pH 3 to pH 10. The second dimension took advantage of molecular sieve separation. Gels were silver-stained and after re-run stained with colloidal Coomassie. After image analysis to identify those spots where changes in positioning/protein amount had occurred, relevant spots were excised and analyzed by MALDI-Tof (Micromass equipment). Taking advantage of the recently disclosed full genome sequence of Gallus gallus proteomic analysis of laying hen RBC membrane proteins was performed for the first time.

The antinutritional factors (ANFs) present in Vicia spp. seeds are a major constraint to the wider utilization of these crops as grain legumes. In the case of faba bean (Vicia faba L.), a breeding priority is the absence vicine and convicine (v-c); responsible for favism in humans and for the reduced animal performance or low egg production in laying hens. The discovery of a spontaneous mutant allele named vc-, which induces a 10-20 fold reduction of v-c contents, may facilitate the process. However, the high cost and difficulty of the chemical detection of v-c seriously restricts the advances in breeding-selection. To identify random applied polymorphic DNA (RAPD) markers linked to this gene, we have analyzed an F2 population derived from a cross between a line with high v-c content (Vf6) and the vc- genotype (line 1268). Quantifiation of v-c was done by spectrophotometry on the parents and the F2 population (n = 136). By using bulked segregant analysis (BSA), two RAPD markers linked in coupling and repulsion phase to the allele vc- were identified and further converted into sequence characterized amplified regions (SCARs). Amplification of SCARS was more consistent, although the initial polymorphism between pools was lost. To recover the polymorphisms several approaches were explored. Restriction digestion with HhaI (for SCAR SCH01620) and RsaI (for SCAR SCAB12850) revealed clear differences between the parental lines. The simultaneous use of the two cleavage amplified polymorphism (CAP) markers will allow the correct fingerprinting of faba bean plants and can be efficiently used in breeding selection to track the introgression of the vc- allele to develop cultivars with low v-c content and improved nutritional value. More detailed information about these markers is available in the following publication: - Gutiérrez N, Avila CM, Duc G, Marget P, Suso MJ, Moreno MT and Torres (2006) CAPs markers to assist selection for low vicine and convicine contents in faba bean (Vicia faba L.). Theoretical and Applied Genetics, 114 (1):59-66

Brief description of recommended method for screening for Botrytis fabae resistance 1.3.1 Field screening. In general, chocolate spot is favored by early planting. A local susceptible check should be grown to help spread the disease and develop a uniform disease pattern throughout the nursery. Best results are obtained when artificial inoculations are done on cloudy rainy days, particularly at sunset. To produce and multiply inoculum stored sclerotia or fresh leaves or stems showing symptoms of the aggressive stage of chocolate spot (dark brown blackish lesions) should be surface sterilized, plated on PDA medium, and incubated at room temperature (20-25°C). Conidia are then harvested and diluted with sterile tap water until a spore suspension containing about 4-5×105 spores/ml is obtained that is sprayed over plants about two months after sowing. High humidity can be maintained by sprinkling the plants with a fine vapor of water several times a day, until the symptoms appeared extensively on the susceptible lines. Disease symptoms can be scored weekly on the basis of a 1-5 or a 1-9 visual scales, being a combination of lesion type, lesion frequency and extent of damage. A single assessment at about 3 week after inoculation in the field can be sufficient to rank the genotypes for their reaction to the disease, but sequential observations and calculation of AUDPC values is recommended to study the disease evolution. 1.3.2 Detached Leaf Test in the Laboratory. Fully-expanded leaflets of similar physiological age should be collected from about the eighth node position and laid flat on a moistened filter paper laid on sterile benches. The upper side of the leaves is inoculated with 1.5 ml of a spore suspension containing about 5×105 spores of B. fabae/ml. The benches are then covered with polythene sheets and leave at room temperature (20 ± 2°C) for 5 - 6 days till disease is assessed. A 1-4 scale can be used, where 1 = no infection or very small flecks (1-25% necrosis); 2 = necrotic flecks with few small lesions (26-50% necrosis), and very poor sporulation; 3 = medium coalesced lesions (51-75% necrosis) with intermediate sporulation; 4 = large coalesced lesions (76-100% necrosis) with abundant sporulation.

Rationale It has been observed that toxic compounds contained in faba beans included in laying hens diets negatively modify body mass development and egg laying activity. Likely culprits of faba bean toxicity are the beta-glycosidic pyrimidine derivatives vicine (V) and convicine (C) contained in large amounts in faba beans. In vivo, inactive V and C are activated to divicine (D) and isouramil (I). D and I react over-stoiochiometrically with reduced glutathione (GSH) and generate oxidized glutathione (GSSG). After faba bean consumption, humans and animals experience a transient decrease of the GSH level in RBCs. As a consequence of GSH oxidation, RBCs become more sensitive to oxidant insult (hemolysis). Laying hen RBCs reflect the redox status of other organs and are suitable for follow-up studies due to their easy availability in sufficient amounts, and possibility to repeat sampling at intermediate time points without animal sacrifice. The working hypothesis of the present study was that the level of V and C in diet would reflect in the circulating levels of D and I in the blood and in the redox state of laying hen RBCs in vivo. The usage of cultivars low in V/C developed by Dr. Duc's group should be free of the negative effects of feeding with traditional faba bean cultivars.

Time course study of diet effects on red blood cells (RBC) of laying hen produced by P3 in trial 1 (20% faba bean in diet) Present results confirm that a diet containing high levels of V/C indeed significantly increased the redox sensitivity of laying hen RBC. Blood was taken on day 0, 7, 14 and 21 after start of diet. Assay of the RBC sensitivity to oxidants was performed with a H2O2 challenge at 0.02mM, 0.2mM and 2mM. Results indicate an increased oxidant sensitivity of RBCs from laying hens fed a high vicine/convicine (V/C) diet. The effect was evident at 2 mM H2O2 after 12, 24 and 48 h incubation where percent hemolysis was significantly higher compared to control RBCs. GSH levels were significantly lower (p<0.002) in the RBCs from laying hens fed a high V/C diet examined on day 7, and were tendentially also lower at other time points without attaining a constant significance compared to control values, though. Membrane ghosts were also isolated. Two-dimensional electrophoretic analysis of membrane proteins and proteomic analysis were performed.

The variability for vicine contents ranged from 0.036 to 0.469 % of seed DM, for convicine content from 0.004 to 0.241 % of seed DM. The variation in protein content ranged from 19.7 % to 38.7% of seed DM: A large variability for tannin and vicine-convicine content was made available for chemistry and animal tests. This material is available on request from Centre of genetic resources of grain legumes at INRA Dijon.

Along this period, Uromyces spp. isolates collected on different legume species from 4 continents (Europe, North America, Africa and Australia) have been multiplied. The rust collection has been enlarged along this period and include at present 154 isolates from 18 countries. Part of these materials have been used for DNA extraction in order to amplify and sequence the ribosomal DNA internal transcribed spacer (ITS) region. This region has been useful to investigate how the geographical origin, the host specificity or both factors affect the filogenetic relationships at the species level. Maximum parsimony An unweighted maximum parsimony (MP) analysis was conducted using PAUP* 4.0b10 (Swofford, 2003) with the following settings: random addition sequence (addseq=random) with 1000 replicates (nreps=1000) and no more than 1000 trees saved per replicate (nchuck=1000), tree bisection reconnection branch swapping (swap=tbr). Gaps were treated as missing data. Bootstrap support values (Felsenstein, 1985) from 1000 pseudoreplicates were calculated using the heuristic search options as above except random addition sequence with 10 replicates (Figure 4). Maximum likelihood The model of nucleotide evolution that best fitted the data was determined by using Akaike Information Criterion (AIC; Akaike, 1974) as implemented in Modeltest 3.7 (Posada and Crandall, 1998). Parameters of the chosen models were used in PAUP* to find the best trees under the maximum likelihood (ML) criterion Accelerated maximum likelihood explorations were performed using likelihood ratchets in conjunction with PAUPRat (Sikes and Lewis, 2001) for more efficient searching of the likelihood surface. We ran 200 iterations with GTR+G likelihood model of evolution and uniformly reweighing 15% of the dataset per iteration. Bootstrap support values from 30 pseudoreplicates were calculated using the heuristic search with random addition sequence with 10 replicates limited to 10000 tree rearrangements (branch swaps) imposed separately for each addition-sequence replicate (rearlimit=10000; limitperrep=yes). Results derived from this study are being summarized at present in an article that will be will submitted shortly for publication. The outcome will be applied to diagnose the disease in absence of fungal reproductive structures. References Akaike H. 1974. A new look at the statistical model identification. IEEE Transactions on Automatic Control 19: 716-723. Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-793. Posada D. y Crandall K.A. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:917-818. Sikes D.S. y Lewis P.O. 2001. PAUPRat: PAUP* implementation of the parsimony ratchet (beta software, version 1). Distributed by the authors. Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, USA http://www.ucalgary.ca/~dsikes/software2.htm. Swofford D. L. 2003. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4. Sinauer, sunderland, MA.

Deliverables

Share this page

Download