Crop Science Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pilet, M.L.
Right arrow Articles by Delourme, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Pilet, M.L.
Right arrow Articles by Delourme, R.
Agricola
Right arrow Articles by Pilet, M.L.
Right arrow Articles by Delourme, R.
Related Collections
Right arrow Other Oil Crops
Right arrow Cell Biology & Molecular Genetics
Right arrow Plant Disease
Right arrow Crop Genetics
Crop Science 41:197-205 (2001)
© 2001 Crop Science Society of America

CELL BIOLOGY & MOLECULAR GENETICS

Stability of QTL for Field Resistance to Blackleg across Two Genetic Backgrounds in Oilseed Rape

M.L. Pilet, G. Duplan, M. Archipiano, P. Barret, C. Baron, R. Horvais, X. Tanguy, M.O. Lucas, M. Renard and R. Delourme

UMR INRA-ENSAR, Amélioration des Plantes et Biotechnologies Végétales, BP 35327, 35653 Le Rheu Cedex, France

Corresponding author (rdelourm{at}rennes.inra.fr)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION-PERSPECTIVES
 REFERENCES
 
Blackleg, caused by Leptosphaeria maculans (Desm.) Ces. et de Not., is a major disease of oilseed rape (Brassica napus L.) worldwide. Molecular markers would be useful tools to assist breeding for blackleg resistance. The objective of this study was (i) to map and characterize quantitative trait loci (QTL) for field blackleg resistance in doubled haploid (DH) and F2:3 populations from the cross `Darmor' (resistant) x `Samourai' (susceptible) and (ii) to compare QTL with those previously identified in the cross `Darmor-bzh' x `Yudal'. A total of 134 DH lines and 185 F2:3 families were genotyped with random amplified polymorphic DNA (RAPD) and restriction fragment length polymorphism (RFLP) markers and assessed for a disease index of resistance in 1998 and/or 1999 in one location. Genetic maps derived from the two populations included a total of 257 and 81 markers, respectively. Up to 30% of these markers were common to the Darmor-bzh x Yudal map previously used. A total of six and four genomic regions were associated with resistance in the DH and F2:3 populations, respectively. They collectively explained 36 to 42% of the variation within each year and population. Three of them were consistent across the two populations derived from Darmor x Samourai cross and expressed dominant or overdominant effects. Four favorable alleles were derived from the susceptible parent. A total of 16 genomic regions were revealed for blackleg resistance in the two crosses Darmor-bzh x Yudal and Darmor x Samourai studied. Four of them were consistent over the two crosses. The inconsistencies observed between populations and crosses can be explained by different genetic backgrounds and disease infestation levels. For marker-assisted selection, these results suggest that QTL mapping must be carried out separately for each population.

Abbreviations: DH, doubled haploid • MAS, marker-assisted selection • PCR, polymerase chain reaction • QTL, quantitative trait locus(i) • RAPD, random amplified polymorphic DNA • RFLP, restriction fragment length polymorphism


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION-PERSPECTIVES
 REFERENCES
 
BLACKLEG (crown canker) of oilseed rape is a serious disease in most oilseed rape producing countries (Gugel and Petrie, 1992). In Europe, since the mid 1980s, the levels of stem canker have fluctuated because of interyear variations in climatic conditions, pathogen populations, and resistance level of cultivated varieties. Because chemical control is not efficient, efforts focus on research programs aiming at a genetic control of the disease.

In oilseed rape, several sources of resistance to blackleg are known (Rimmer et al., 1995; Rimmer and Van Den Berg, 1992; Salisbury et al., 1995). Genetic studies have allowed two main types of resistance to L. maculans to be distinguished: qualitative total resistance at the seedling and/or adult plant stage(s) (Ansan-Melayah et al., 1998; Delwiche, 1980; Dion et al., 1995; Ferreira et al., 1995; Hill, 1991; Mayerhofer et al., 1997; Sawatsky, 1989; Stringam et al., 1992) and quantitative adult partial resistance (Cargeeg and Thurling, 1979; Ferreira et al., 1995; Pang and Halloran, 1996a, b; Pilet et al., 1998a). In breeding for blackleg resistance, qualitative resistances are easier to manipulate than quantitative resistances; however, durability of the former has not been demonstrated yet. The quantitative resistance to L. maculans in `Jet Neuf' has been shown to be durable for about 10 yr in western and eastern Europe, so it is of great interest. In a previous study (Pilet et al., 1998a), we genetically dissected the quantitative resistance of the cultivar Darmor, derived from Jet Neuf, in a population of 152 DH lines from the cross Darmor-bzh (resistant) x Yudal (susceptible) (Foisset et al., 1996). `bzh' is a dwarf gene introduced into Darmor. Using two resistance criteria, we identified a total of 13 genomic regions for field blackleg resistance assessed for 2 yr (1995, 1996) in one location (Le Rheu, France). Four of them were stable across the 2 yr, including one from the susceptible parent.

In breeding, molecular markers offer the possibility to make easier the engineering of genetic factors (QTL) controlling blackleg quantitative resistance, but several questions have to be considered before resistance QTL are used in marker-assisted selection (MAS) schemes. Specifically, these questions concern the effects of QTL (additivity, dominance, epistasis) and QTL stability across different genetic backgrounds. In introgression schemes for creating oilseed rape hybrids, QTL with dominant effects independent of the genetic background (polymorphic in different genetic backgrounds and without epistatic effects) would be very useful.

In this paper, we carry out the genetic study of Darmor resistance to L. maculans in the Darmor x Samourai cross. Darmor has been used without the dwarf gene to avoid the gene's effect on blackleg resistance, which we observed previously (Pilet et al., 1998a). Samourai has been chosen because of its extensive use as a genitor in oilseed rape breeding programs. Samourai is moderately resistant to blackleg, whereas Yudal is very susceptible. In the Darmor x Samourai cross, the genetic analysis was performed in two populations, a DH and a F2:3 progenies. The DH population presents the main advantage of being indefinitely multipliable, thus permitting a better evaluation of complex traits such as resistance to L. maculans. The F2:3 population is not immortal but allows an estimate of dominance effects of QTL. The objectives of this study were (i) to detect, characterize, and compare QTL associated with field blackleg resistance in the DH and F2:3 populations from the Darmor x Samourai cross and (ii) to compare the QTL detected in this cross with those previously mapped in the Darmor-bzh x Yudal DH population and determine the consistency of the QTL detected across the two crosses.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSION-PERSPECTIVES
 REFERENCES
 
Mapping Populations
The two parents Darmor and Samourai used for the cross studied are French winter double-low lines, registered in 1983 and 1989, respectively. From 10 Darmor x Samourai F1 plants, a total of 134 DH lines were derived by in vitro androgenesis (Polsoni et al., 1988), then genotyped and phenotyped. From a single F1 hybrid plant, 185 F2:3 families were produced by self-pollination and used for QTL mapping. For field trials, the controls used were five winter B. napus cultivars with various levels of adult resistance to L. maculans (`Columbus': resistant; `Vivol': resistant; `Falcon': partially resistant; `Eurol': moderately resistant; `Shogun': susceptible).

Field Disease Trials
The Darmor x Samourai DH and F2:3 progenies were tested in field disease trials at one location (Le Rheu, France). In 1998, the two populations were grown in a randomized incomplete block design with three replicates and six blocks per replicate. Each replicate included five-row plots (2.5 m2, 25 cm between rows) of parental, DH or F2:3 and control lines. In 1999, experimentation was performed on the DH progeny in a similar design as in 1998, with two replicates. Reinforced conditions of contamination were set up in the field at the three-leaf development stage by scattering L. maculans infested oilseed rape stubble collected from the previous year at Le Rheu.

Blackleg severity was assessed when the plant were just beginning to ripen (June). Thirty plants per plot were uprooted and scored by the following scale which was based on the extent of external and internal symptoms at the crown level: 0 = healthy plant; 1 = external small necrotic spot; 3 = one-side external or small internal necrosis; 5 = external nearly complete girdling or one-side internal necrosis, resistant to stem bending; 7 = deep necrosis and complete girdling, plant yellow; 9 = broken crown, dead plant. For each plot, a disease index is calculated as the mean of the 30 plants uprooted.

Isozyme, RAPD, PCR-Specific, and RFLP Assays
Isozyme analyses were performed from the Darmor x Samourai DH population as described in Foisset et al. (1996).

RAPD, PCR-specific, and RFLP genotyping of Darmor x Samourai DH line or F2:3 family was realized from young leaves harvested on 20 pooled plants per family, then kept at -80°C until DNA extraction. DNA was extracted as described in Doyle and Doyle (1990).

RAPD assays and nomenclature of RAPD markers used have been previously described in detail by Foisset et al. (1996). PCR-specific markers were Amplified Genetic Consensus Markers as reported by Brunel et al. (1999).

A total of 253 RFLP probes, from four different origins, were used. Most of them (171 probes) corresponded to anonymous clones from a B. napus embryo cDNA library (Harada et al., 1988), which were provided by DNA Landmarks (St-Jean sur Richelieu, Canada). They were designated as BN103 to BN747 and have been already used to generate genetic maps from `Westar' x `Topas' (Landry et al., 1991) and Darmor-bzh x Yudal (Foisset et al., 1996) crosses. A total of 64 probes corresponded to Arabidopsis thaliana ESTs. They were supplied by the Arabidopsis Biological Resource Center DNA Stock Center (http://www.arabidopsis.org/madison97/at97abs/imar_12-27.html) and were named as AT1 to AT95. The other probes were amplified fragments from consensus primers of A. thaliana (L.) Heynd. resistance genes (RPM1 and RPS2). They were provided by M. Fourmann and D. Brunel (INRA Versailles, France) (Fourmann, 1998) and were coded Dkgr1 to Dkgr15, RPS2, 4-6St1, 8-10St2, Vers6-9 to Vers9-11. RFLP assays were realized following the procedure described by Sharpe et al. (1995). Six restriction enzymes, BamHI, DraI, EcoRI, HindIII, KpnI, and XbaI, were used. RFLP markers were coded by the name of their derived-probe, which could be followed by the letter `a' or `b' if two codominant markers were obtained from the same probe, or the letter `d' if a dominant marker was detected.

Linkage Maps
For each segregating marker, a chi-square analysis ({alpha} = 0.01) was used to test for deviations from the expected Mendelian ratios (1:1 in a the DH population, 1:2:1 or 3:1 in the F2:3 population with codominant or dominant markers, respectively). Linkage analyses from the DH and F2:3 Darmor x Samourai populations were performed applying Mapmaker/exp version 3.0 (Lander et al., 1987) and a chi-square test of independence (Mather, 1957) as described in Foisset et al. (1996). Linkage groups were established with a log10 of the odds ratio (LOD) threshold of 4.0 and 3.0 for the DH and the F2:3 populations, respectively, and a maximum recombination frequency of 0.4. Centimorgan distances were expressed with the Kosambi function (Kosambi, 1944).

Statistical Analyses and QTL Mapping
Experimental data for the mean disease index scored in 1998 and 1999 in the DH and F2:3 Darmor x Samourai populations as well as genotype x year interaction for the DH population were analyzed by a generalized linear model (SAS, 1989). Within each year, analysis of variance (ANOVA) partitioned total variation into effects of lines, replicates, blocks and errors ( where Pijk is the mean disease score of the ith DH line located in the kth block of the jth replicate, µ the mean of all the data, Li the DH Line i effect, Rj the Replicate j effect, Bk/j the block effect in the jth replicate and eijk the residual). Normality of each residual distribution was assessed by the PROC UNIVARIATE procedure. The Pearson coefficient was calculated with the PROC CORR procedure to determine correlations among adjusted means between the 2 yr. Heritabilities (h2) were also calculated from ANOVA with the formula: with {sigma}2G the genetic variance, {sigma}2E the environmental variance and n the number of replicates per line.

From adjusted means of the blackleg disease index, QTL mapping was performed by Interval mapping by Mapmaker/QTL 1.1 (Lincoln et al., 1992). In the DH progeny, QTL were first detected with a LOD threshold of 2.9, corresponding to an overall {alpha}-type I error risk of 5% (Lander and Botstein, 1989). Further QTL were revealed if the total LOD was more than 2.0 LOD of the total LOD for the first fixed detected QTL. In the F2:3 population, QTL were mapped by additive, dominant, and recessive models independently, with a LOD threshold of 2.3 ({alpha} = 0.05 according to Lander and Botstein (1989)) for each model. QTL detected by Interval mapping were checked (P < 0.05) by multiple regression of the resistance on the closest markers from the LOD score peaks of QTL (considered as the most-likely QTL position). Normality of residual distributions after multiple regression was also checked. Confidence interval of a QTL was defined by the region within 1 LOD from the QTL peak. The proportion of variance explained by the QTL detected (R2), their additive and dominant effects were estimated by Mapmaker/QTL 1.1 at the QTL peaks.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSION-PERSPECTIVES
 REFERENCES
 
Linkage Maps
Polymorphism between Darmor and Samourai was tested by means of 174 RAPD primers, 253 RFLP probes, 46 PCR-specific primers and 10 isoenzymatic systems. The proportion of RAPD primers and RFLP probes which generated polymorphic bands was 73 and 35%, respectively.

The genetic map constructed from 134 DH Darmor x Samourai lines included a total of 257 markers, including 161 RAPD, 81 RFLP, five isozymes, and 10 PCR-specific markers. Among them, six were unlinked and 251 were distributed on 24 linkage groups with a mean spacing of 6.1 ± 3.0 cM. This map covered 1475 cM, corresponding to about 67% of the Darmor-bzh x Yudal genetic map. Out of the set of markers, 19.5% (mainly RAPD) segregated following non Mendelian ratios, with an equal number of loci favoring the alleles of each parental line. Markers showing segregation distortion were essentially clustered in four linkage groups. Seventy five marker loci (68 RAPD, four RFLP, three isozymes) were common to the Darmor x Samourai DH map and the Darmor-bzh x Yudal genetic map described in Foisset et al. (1996) and Pilet et al. (1998a). These common loci allowed to assign the Darmor x Samourai linkage groups according to the Darmor-bzh x Yudal ones. DS2b linkage group was named according to common markers to another genetic map developed from the cross `Stellar' x `Drakkar' (Delourme et al., 2000, unpublished data). AFLP markers recently added to the maps from the three crosses (Darmor-bzh x Yudal, Darmor x Samourai and Stellar x Drakkar) confirmed the assignment of Darmor x Samourai linkage groups (data not shown).

The genetic map constructed from the 185 F2:3 Darmor x Samourai families included a total of 81 loci, among which 65 RFLP, 13 RAPD, and three PCR-specific markers. Sixty-three markers were codominant and 18 were dominant. Twelve loci were unlinked and 69 were distributed on 20 linkage groups, with a mean spacing of 8.9 ± 4.6 cM. The map covered 609 cM, i.e., about 28% of the Darmor-bzh x Yudal genetic map. Twelve percent of the loci presented segregation distortion at the 0.01 threshold. All of the 81 loci were common to the Darmor x Samourai DH genetic map. Their order and the distances were quite well conserved between the two Darmor x Samourai maps.

Blackleg Resistance Data
Phenotypic means of controls and Darmor x Samourai populations (DH and F2:3) for the disease index of blackleg resistance and variance components are indicated in Table 1. They were higher in 1999 than in 1998 (P < 0.001) expressing a lower infestation level in 1998. In 1998, the F2:3 progeny was globally more resistant than the DH one (P < 0.001).


View this table:
[in this window]
[in a new window]
  		Table 1. Statistical parameters of DH and F2:3 Darmor x Samourai populations assessed for field blackleg resistance in 1998 and 1999 at Le Rheu, France

 
Analysis of the experimental data showed no significant genotype x year interaction effect (P = 0.12) for the disease index in the DH progeny. In the DH and F2:3 progenies, genotypic effect was highly significant in 1998 and 1999 (P < 0.001 for both years and populations). Distributions of residues after variance analysis were normal in both years and populations according to the Shapiro-Wilk W statistic (Shapiro and Wilk, 1965) (in the DH population: P = 0.64 and 0.14 in 1998 and 1999, respectively; in the F2:3 progeny: P = 0.54). Within-year heritabilities were estimated at 0.65 and 0.64 in 1998 and 1999, respectively for the DH population. In the F2:3 progeny, heritability was lower .

Frequency distributions of the DH and F2:3 adjusted means for blackleg resistance in 1998 and 1999 are represented in Fig. 1 . Genotypes were continuously distributed for the resistance index across both years and populations. This confirmed the quantitative inheritance of Darmor field resistance to L. maculans. In the DH progeny, Pearson correlation between 1998 and 1999 for the index adjusted means was highly significant .



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 1. Frequency distributions of the Darmor x Samourai DH lines (in 1998 and 1999) and F2:3 families (in 1998) for adjusted means of the index of field resistance to Leptosphaeria maculans. Note: Adjusted means of parental control lines (D: Darmor, S: Samourai) are given in legend for each field trial

 
Blackleg Resistance QTL in Darmor x Samourai Cross
Characteristics of QTL detected from adjusted means of blackleg resistance index are described for each year and progeny type in Table 2. All the QTL identified were distributed over seven linkage groups.


View this table:
[in this window]
[in a new window]
  		Table 2. Characteristics of QTL associated with field blackleg resistance (for the mean disease index), detected in DH and F2:3 populations of the cross Darmor x Samourai over two years (1998 and 1999)

 
In the Darmor x Samourai DH population, a total of five QTL were revealed in 1998, collectively accounting for 42% of the variation. Two of them (on DS3 and DS11 groups) were detected with a higher LOD than 2.9 and explained individually 12.2 and 11.5% of the variation. Three further QTL, located on DS2b, DS6, and DS8 groups, were identified with a multiple-QTL model. Two of them (on DS6 and DS8 groups) were retained with an additional LOD below 2.0. These two last QTL were kept because their effect on the resistance was significant (P < 0.05) in a multiple regression model. At these two QTL, resistance alleles were derived from the susceptible parent Samourai.

In 1999, a total of four QTL were associated with resistance to blackleg in the Darmor x Samourai DH population, explaining 36.5% of the variation. The first one, situated on DS16 group and explaining 11% of the variation, was detected at a LOD threshold of 2.7. When the position of this QTL was fixed, three additional QTL already identified in 1998 were successively revealed (on DS2b, DS6, and DS11 groups). These four QTL displayed a significant effect on the disease index by multiple regression. At the QTL on DS6 and DS16 groups, resistance allele originated from Samourai. Out of the five and four QTL detected in the DH progeny in 1998 and 1999, respectively, three were common to the 2 yr.

When using the adjusted means over the 2 yr, we detected the QTL that are consistent over the 2 yr (on DS2b, DS6, and DS11) and the QTL on DS3 which was only detected in 1998.

In the F2:3 Darmor x Samourai population, we identified a total of four QTL associated with the index resistance in 1998. They collectively accounted for 41% of the variation and individually explained from 6.6 to 14.6% of the variation. The two major QTL, localized on DS6* and DS11* groups, were also associated with resistance in the DH population across the 2 yr, with the same parental contributions. They displayed overdominant effects favoring an increased level of resistance (on DS11*) or susceptibility (on DS6*) in heterozygous plants. The third QTL detected in the F2:3 progeny, located on DS3* group, was also found in the DH progeny in 1998. It showed a dominant effect in favor of the resistance. The fourth QTL, on DS1a* group, was specifically identified in the F2:3 population and exhibited an additive effect. Out of the four QTL detected in the F2:3 Darmor x Samourai progeny in 1998, three were revealed in the DH population in 1998 and/or 1999.

Blackleg Resistance QTL across Darmor-bzh x Yudal and Darmor x Samourai Crosses
Localization of blackleg resistance QTL detected in the cross Darmor x Samourai is presented in Fig. 2 , in comparison with those previously identified in the cross Darmor-bzh x Yudal using two resistance criteria (Pilet et al., 1998a). A total of 16 genomic regions were associated with resistance to L. maculans across the two crosses and the 4 yr (1995, 1996, 1998, and 1999).



View larger version (60K):
[in this window]
[in a new window]
  		Fig. 2. Genomic localization of QTL identified for field resistance to L. maculans in DH and F2:3 Darmor x Samourai populations compared with those revealed in the Darmor-bzh x Yudal DH population (Pilet et al., 1998a) across a total of 4 yr. Resistance was assessed by the mean disease index (I) in the two crosses and the percentage of lost plants (P) in the cross Darmor-bzh x Yudal. Note: Linkage groups (LG) of genetic maps constructed from the DH Darmor-bzh x Yudal population and the DH and F2:3 Darmor x Samourai progenies were named DY, DS and DS*, respectively. Only the LG where QTL were identified are represented. Distances between markers are indicated in Kosambi cM. QTL length is the confidence interval where the likelihood of the presence of a QTL is within tenfold (1 LOD) of its maximal value

 
Among them, four regions appeared consistent across the two crosses Darmor-bzh x Yudal and Darmor x Samourai. Three of them, localized on the linkage groups (LG) 2, 3, and 11, can be considered as major regions because they were revealed across 3 or 4 yr and explained individually up to 14.6% of the variation. At QTL located on LG3 and LG11, resistance alleles derived from the resistant parent Darmor. At QTL localized on DY2 and DS2b groups, resistance derived from the parents Yudal and Darmor, respectively. The fourth region, located on LG8, was identified for one year in each cross with minor opposite effects.

Twelve genomic regions were specifically detected in one cross, nine in Darmor-bzh x Yudal and three in Darmor x Samourai. Most of them corresponded to regions specifically expressed for 1 yr or one criterion, often with minor effects (on DS1a*, DY1b, DY6, DY9, DY10, DY13 groups), but sometimes with major effects (on DY6 and DS16 groups). Particularly, a year-specific resistance QTL with a major effect was detected on DY6 group near the `bzh' gene in Darmor-bzh x Yudal cross. Some other cross-specific regions were identified across 2 yr, criteria, or progenies (on DY5 and DS6 groups) and appeared to be strongly involved in blackleg resistance.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSION-PERSPECTIVES
 REFERENCES
 
QTL Detected in DH and F2:3 Darmor x Samourai Progenies
In this study, we identified QTL involved in blackleg resistance in DH and F2:3 populations of the cross Darmor x Samourai. Many resistance QTL mapping studies were carried out from F2:3 progenies (Bohn et al., 1996; Bubeck et al., 1993; Freymark et al., 1994, 1993; Mestries et al., 1998; Qiu et al., 1999; Rector et al., 1998; Saghai Maroof et al., 1996; Schechert et al., 1999; Schön et al., 1993; Welz et al., 1999; Xia et al., 1999) but none of them compared QTL detected in F2:3 and DH progenies from the same cross experimented in the same conditions. A total of seven genomic regions were associated with field resistance to L. maculans across the two Darmor x Samourai populations. QTL stability across years or populations was moderate since four regions (on DS2b, 3, 6, 11 groups) were common to the 2 yr and/or the two populations. These four QTL were revealed when we used adjusted means over the 2 yr in the DH population. This analysis was then useful to identify the most consistent QTL. We chose also to analyze 1998 and 1999 data separately since we wanted to assess QTL consistency over years. Three regions were population- and/or year-specific. Differences in QTL mapping observed between the two progenies derived from the same F1 hybrid population might be explained by QTL x environment interactions (the QTL on DS8 and DS16 were only detected for one year in the DH progeny) and by the genome coverage (no linkage group was constructed for the region corresponding to DS2b in the F2:3 progeny).

QTL Stability across Darmor-bzh x Yudal and Darmor x Samourai Crosses
Considering results obtained in the two crosses Darmor-bzh x Yudal (Pilet et al., 1998a) and Darmor x Samourai, a total of 16 genomic regions were identified for field blackleg resistance across 4 yr. Consistency of QTL results was moderate-to-low since only four regions were common to the two crosses. These common regions could correspond to the same or linked loci, polymorphic in the two crosses. For two of them (on LG3 and LG11), the Darmor allele always contributed to the resistance across years and populations, suggesting that these common regions could correspond to the same loci.

Numerous cross-specific regions were identified for field resistance to L. maculans. Most of them were also year-specific. Only two of them, located on DY5 and DS6, were detected in 2 yr. Nevertheless, populations were grown in same designs at the same location (Le Rheu, France). Resistance of the genotypes was scored using identical notation systems, DH population sizes did not differ greatly (n = 152 in Darmor-bzh x Yudal and n = 134 in Darmor x Samourai), marker density of the two DH genetic maps were similar. The homogeneity of these experimental conditions brings out three main reasons for the inconsistency of QTL number and positions observed among the two crosses.

  1. Genetic background is a major factor of the QTL instability observed. A lack of consistency between QTL detected in the two crosses could occur because different sets of QTL for blackleg resistance segregated in the two crosses and only polymorphic QTL were detected. Epistatic interactions could have also led to QTL instability among the two crosses. In the two crosses studied, the resistant parents Darmor-bzh and Darmor are near isogenic lines but the other parents shows different level of field resistance, Yudal being more susceptible to blackleg than Samourai. Consequently, we observed in this study that (i) the P criterion (percentage of dead plants) used in the Darmor-bzh x Yudal study (Pilet et al., 1998a) was not discriminant in the Darmor x Samourai cross and thus was not used (data not shown); (ii) for the disease index, a fewer number of QTL and/or QTL with lower effects were detected in the Darmor x Samourai cross than in the Darmor-bzh x Yudal one; and (iii) the number of QTL at which resistance allele derived from susceptible parent was higher in the Darmor x Samourai cross than in the Darmor-bzh x Yudal one (two DY regions with negative effect versus four DS regions). These QTL were specific to the cross including the susceptible donor parent of resistance alleles, except those located on LG2 and LG8 which were revealed in the two crosses. On these groups, QTL with opposite effects in the two crosses were revealed. This suggests that these common QTL located on LG2 and LG8 might be linked loci or identical loci with three different alleles. Further QTL analyses using composite interval mapping (Jansen, 1993; Zeng, 1993, 1994) or, more recently, multiple interval mapping methods (Kao et al., 1999) would allow us to precisely position QTL in these regions. (iv) Resistance QTL detected around the `bzh' gene (DY6 group) on the Darmor-bzh x Yudal map was not revealed in the Darmor x Samourai cross. This observation could support the hypothesis of a pleiotropic effect of the dwarf gene on blackleg resistance and other agronomic traits in oilseed rape, as previously suggested (Pilet et al., 1998a, b).
  2. The second reason for QTL instability concerns the infestation level in the field. Populations from the two crosses were not grown in the same years (Darmor-bzh x Yudal DH lines in 1995 and 1996; Darmor x Samourai DH and F2:3 progenies in 1998 and 1999). Inoculum pressure was high in 1995 and low in 1998 compared with 1996 and 1999, as demonstrated by control values (for the disease index in 1995, 1996, 1999 and 1998, respectively, Falcon: 4.4, 3.9, 3.6, 2.2; Eurol: 5.9, 5.4, 4.0, 2.4; Shogun: 7.9, 6.8, 6.5, 4.3). Hence, magnitude of variation for the disease index fluctuated according to years, influencing the number of QTL detected. This year effect was especially obvious in the cross Darmor-bzh x Yudal (Pilet et al., 1998a). Most of year-specific QTL presented an effect too low to be detected each year and a lot of them were detected in 1995. This indicates that consistent heavy disease pressure is required to accurately assess disease resistance for QTL mapping.
  3. Another simple explanation of QTL inconsistencies observed among the two crosses is related to the genome coverage level by the genetic maps generated. Sizes of the DH and F2:3 Darmor x Samourai genetic maps correspond to about 67 and 28% of the Darmor-bzh x Yudal map. In genomic regions uncovered by markers in the cross Darmor x Samourai, it is possible that resistance QTL exist but were not detected. This hypothesis is valid in the F2:3 Darmor x Samourai population, but not in the DH one where the genetic map covered all the regions identified for blackleg resistance in the cross Darmor-bzh x Yudal.


   CONCLUSION–PERSPECTIVES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION-PERSPECTIVES
REFERENCES
 
This study is the first to report QTL mapping of oilseed rape disease resistance across more than one cross. In other species, a few quantitative resistance mapping studies investigated QTL stability across different crosses; most of them were conducted with maize, Zea mays L. (Bohn et al., 1997; Bubeck et al., 1993; Groh et al., 1998; Lübberstedt et al., 1998a,b). These studies often displayed a low consistency of resistance QTL positions among crosses.

In this paper, we showed that the consistency of blackleg resistance QTL among two crosses was moderate-to-low since we identified four consistent and 12 specific regions among the two crosses Darmor-bzh x Yudal and Darmor x Samourai. We also revealed six QTL for which resistance alleles derived from the susceptible parents. These results imply that separate QTL mapping experiments must be conducted in the germplasm in which the final selection will be carried out in order to identify QTL usable for MAS. At least three genomic regions associated with field resistance to blackleg (on LG2, LG3 and LG11) could be interesting in MAS in order to increase resistance of Samourai. Regions on LG3 and LG11 could be all the more interesting in hybrid selection since they displayed a dominant or overdominant effect in favor of heterozygous for the resistance. A major cross-specific region detected on DS6 group expressed an overdominant effect for susceptibility. Thus, for oilseed rape hybrid breeding, a choice of QTL based on genetic analysis performed in DH (or BC) and F2:3 populations appears important. The construction of parental hybrid lines must take into account all the genetic effects at each chosen QTL.

In our study, blackleg resistance QTL were detected from experiments at one location (Le Rheu), where PG3 isolates of L. maculans are predominant (data not shown). Before initiating a MAS experiment, additional field trials would be necessary to determine QTL stability across different locations, where other pathotypes of L. maculans populations could occur.

At I.N.R.A. of Rennes, mapping of race-specific resistance genes at the cotyledon stage is in progress. Comparison of genomic localization of these genes with QTL mapped in this study will provide valuable information for a better understanding of the oilseed rape–Leptosphaeria maculans interaction and for breeding for durable resistance to blackleg in oilseed rape.


   ACKNOWLEDGMENTS
 
This work was supported by a grant from the French Ministry of Superior Education and Research (MESR), by CETIOM and PROMOSOL (France). We gratefully acknowledge B. Landry and N. Foisset (DNA Landmarks Society, St-Jean sur Richelieu, Canada) and M. Delseny (CNRS, Perpignan, France) for supplying RFLP probes. We are also grateful to D. Brunel and M. Fourmann (INRA, Versailles, France) for having provided us PCR-specific primers. We greatly thank H. Brun (INRA, Le Rheu, France) for her helpful advice in resistance tests and the team of the breeding experimental farm (INRA, Le Rheu, France) for their help in performing disease trials.

Received for publication February 14, 2000.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION-PERSPECTIVES
 REFERENCES
 




This article has been cited by other articles:


Home page
 Appl. Environ. Microbiol.Home page
L. Gout, M. Eckert, T. Rouxel, and M.-H. Balesdent
Genetic Variability and Distribution of Mating Type Alleles in Field Populations of Leptosphaeria maculans from France
Appl. Envir. Microbiol., January 1, 2006; 72(1): 185 - 191.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
T. R. Stefaniak, D. L. Hyten, V. R. Pantalone, A. Klarer, and T. W. Pfeiffer
Soybean Cultivars Resulted from More Recombination Events Than Unselected Lines in the Same Population
Crop Sci., December 2, 2005; 46(1): 43 - 51.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
R. Mihaljevic, H. F. Utz, and A. E. Melchinger
Congruency of Quantitative Trait Loci Detected for Agronomic Traits in Testcrosses of Five Populations of European Maize
Crop Sci., January 1, 2004; 44(1): 114 - 124.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pilet, M.L.
Right arrow Articles by Delourme, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Pilet, M.L.
Right arrow Articles by Delourme, R.
Agricola
Right arrow Articles by Pilet, M.L.
Right arrow Articles by Delourme, R.
Related Collections
Right arrow Other Oil Crops
Right arrow Cell Biology & Molecular Genetics
Right arrow Plant Disease
Right arrow Crop Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome