The aim of this study was to use multiple DNA markers for detection of QTLs related to resistance to white mold in an F2 population of common bean evaluated by the straw test method. The DNA from 186 F2 plants and from the parents was extracted for genotypic evaluation using SSR, AFLP, and SRAP markers. For phenotypic analysis, 186 F2:4 progenies and ten lines were evaluated, in a 14 × 14 triple lattice experimental design. The adjusted mean values of the F2:4 progenies were used for identification of QTLs by Bayesian shrinkage analysis. Significant differences were observed among the progenies for reaction to white mold. In identification of QTLs, 17 markers identified QTLs for resistance—13 SSRs and 4 AFLPs. The
Phytopathogenic organisms are the main agents responsible for significant losses in the common bean (
It is known that the best manner of controlling most common bean diseases is the use of cultivars with some level of genetic resistance. This measure is most recommended because it avoids or reduces the use of agricultural chemicals and is economically viable for the producer. However, for control of white mold, this measure is not efficient since cultivars with a satisfactory level of resistance that are adapted to Brazilian conditions are not available. Some lines and cultivars adapted to the conditions of west-central and southeast Brazil have partial resistance [
To evaluate common bean resistance to white mold, there are diverse methodologies that use artificial inoculation of the pathogen [
The complexity of resistance to white mold has led many researchers to analyses of QTLs for the purpose of locating efficient molecular markers to be used in marker-assisted selection. The distribution of molecular markers throughout the genome allows for detection and localization of QTLs. Some mapping techniques have been developed, and the interval mapping method has proven to be promising. In interval mapping, the QTL genotype is not observable, but it may be predicted based on markers around it; thus, the markers define an interval that may contain a supposed QTL [
However, certain traits that present inheritance of the oligo- or monogenic type may show low genetic variability, and even testing a large number of markers, few polymorphic tags may be found, limiting interval analysis to preestablished linkage groups or making their construction unviable. An alternative would be the association of the marker to the phenotype, that is, where it is assumed that a marker of significant effect is in linkage disequilibrium with the QTL. Nevertheless, if this disequilibrium is unknown, the effect of the marker becomes biased and its significance confuses the effect of the marker with its frequency in recombination with the QTL [
An alternative is simultaneous analysis of markers and the search for QTLs in a model where the establishment of linkage groups is not necessary.
“Moving away” analysis suggested by Doerge et al. [
The aim of this study was to apply the Bayesian method of analysis by multiple markers for detection of QTLs related to white mold resistance in an F2 population evaluated by the straw test method.
The lines CNFC 9506 and RP-2 were crossed. These parents were classified, based on reaction to oxalic acid, as susceptible and partially resistant to white mold, respectively, with CNFC 9506 receiving a score of 4.83 and RP-2 receiving 1.97 in the study developed by Gonçalves and Santos [
The CNFC 9506 line was developed by Embrapa Arroz e Feijão and the RP-2 line by UFLA, and they exhibit upright plant type and carioca (beige with brown stripes) type grains. Both are adapted and average yield (kg/ha) is greater for the RP-2 line [
As of the crossing of the parents, the F1 and F2 generations and the F2:3 and F2:4 progenies were obtained in field conditions. The F2:4 generation and ten lines (Corujinha, G122, CNFC 10720, CNFC 10722, M20, Ex-Rico 23, Small White, and Talismã) were used in the evaluation, with Corujinha being the susceptible control and Ex-Rico 23 the resistant control.
Evaluation of the F2:4 progenies was performed in the field. The experiment was conducted through a
Initially sterilized sclerotia were used for obtaining the mycelium. The fungus
Eight days after inoculation, evaluation of resistance of the common bean to white mold was performed by means of a diagrammatic scale described by Petzoldt and Dickson [
The DNA of the parents, CNFC 9506 and RP-2, and of the 186 F2 progenies, was extracted following the procedures used by Rodrigues and dos Santos [
Initially, random primers of SSRs (Simple Sequence Repeats—Microsatellite), AFLPs (Amplified Fragment Length Polymorphism), and SRAPs (Sequence Related Amplified Polymorphism) were tested, and the polymorphic ones were selected, namely, 17 SSRs, 31 AFLPs, and 11 SRAPs [
The genotypes of the SSR markers were identified with scores of −1, 0, and 1 for the genotypes of smallest number of base pairs, heterozygous, and genotype of greatest number of base pairs, respectively. The AFLP and SRAP markers were identified with scores 0 and 1, representing absence and presence of the band, respectively.
The “moving away from marker” analysis uses individual markers as a parameter in the search for QTLs. Thus, analysis is made using the conditional probabilities of the QTLs given to the reference marker. Thus, the linear model adopted is the following:
In this model, it is assumed that
A priori distributions of the effects of the QTLs (
In this model, only the phenotypic data are observed, whereas the genotypes of the QTLs
Assuming independence among the effects and variance and the genotypes of the QTLs, and also the independence of the observations in relation to the markers and their genetic distances, we have a new likelihood given by
In F2:4 populations, the probability of heterozygous plants within each family is given by 0.125.
Thus, each genotype is sampled directly from a Bernoulli distribution, with probability given by
Use of the Gibbs sampler is prohibitive because the
In the method presented, a uniform distribution may be used as an auxiliary function, where
Therefore, if
The other a posteriori conditional distributions for the
In Bayesian inference, the significance test is not as important as in likelihood analysis. More importantly, the aim of Bayesian analysis via MCMC is to obtain an empirical a posteriori distribution from which all information in respect to the QTL may be obtained. In simple Bayesian analysis, the position of the QTL is inferred based on the number of times the effect of the QTL passes through a small region (bin) in a determined position of the genome. This curve describes the intensity profile of the QTL. In the approach of Wang et al. [
The detection of QTLs associated with resistance to white mold through evaluation in the straw test is shown in Figure
Summary of the distance between the marker and QTL, position of the markers in Figures
Marker | Position | FRa (%) | Distanceb | Effect | Waldc | Heritd (%) |
---|---|---|---|---|---|---|
BM184 | 1 | 20.29 | 21.53 | 0.2 | 1038.04 | 84.48 |
BM187 | 2 | 4.47 | 4.49 | −0.077 | 164.57 | 49.7 |
BM211 | 3 | 1.45 | 1.45 | 0.068 | 120.37 | 76.93 |
BMd42a | 4 | 3.32 | 3.32 | −0.003 | 84.46 | 3.01 |
PVM02TC116 | 5 | 1.7 | 1.7 | 0.02 | 25.24 | 2.57 |
PV188 | 8 | 1.59 | 1.59 | −0.184 | 819.59 | 3.74 |
PV74 | 9 | 1.21 | 1.21 | −0.081 | 171.98 | 0.87 |
PVESTBR_185 | 10 | 45.32 | 75.33 | 0.197 | 952.96 | 75.71 |
PVESTBR_204 | 11 | 1.08 | 1.08 | 0.108 | 313.72 | 3.39 |
PV-gaat001 | 13 | 2.41 | 2.41 | −0.04 | 93.98 | 0.12 |
ME1 | 15 | 0.82 | 0.82 | 0.129 | 416.94 | 71.37 |
BMc94 | 16 | 38.39 | 50.76 | 0.189 | 873.32 | 84.94 |
BMc83 | 17 | 2.24 | 2.24 | −0.096 | 230.64 | 57.92 |
EAAG/MCAG224 | 26 | 9.42 | 9.54 | −0.019 | 37.03 | 77.51 |
EACC/MCAT141 | 36 | 1.8 | 1.8 | −0.019 | 35.74 | 0.39 |
EACC/MCAT126 | 37 | 3.68 | 3.69 | 0.022 | 52.11 | 2.01 |
EACA/MCAT148 | 45 | 1.8 | 1.8 | 0.021 | 41.08 | 4.92 |
Identification of the QTLs by the Wald test.
Among the 59 markers used, 17 identified QTLs for resistance to white mold, with 13 SSRs (BM184, BM187, BM211, BMd42a, PVM02TC116, PV188, PV74, PVESTBR_185, PVESTBR_204, PV-gaat001, ME1, BMc94, and BMc83) and four AFLPs (EAAG/MCAG224, EACC/MCAT141, EACC/MCAT126, and EACA/MCAT148). Of these, only BM184, PV188, PVESTBR_185, and BMc94 are associated with highly significant QTLs.
The effect of the QTLs on expression of resistance to white mold is represented in Figure
Effect of the QTL associated with the marker.
Among the 17 markers, nine are linked to QTLs with effects of increasing resistance to white mold. These QTLs are BM184, BM211, PVM02TC116, PVESTBR_185 PVESTBR_204, ME1, BMc94, EACC/MCAT126, and EACA/MCAT148. Of these, only three are related to highly significant QTLs (BM184, PVESTBR-185, and BMc94).
The frequency of recombination of the QTL with the marker is represented in Figure
Frequency of recombination between markers and QTLs.
It may be seen that the markers BM184, PVESTBR_185, and BMc94 segregate nearly independently of the QTLs associated with them, for they have a high frequency of recombination (from 20.29% to 45.31%) (Table
A summary of the distance data, in cM, between the marker and the QTL, the position of the marker in the figure, and of the effect and heritability of the QTL associated with resistance to white mold by evaluation in the straw test is shown in Table
The SRAP markers were not efficient in identifying QTLs for resistance to white mold by the straw test method.
Various common bean QTLs of resistance to white mold have already been identified; however, most of them are in other countries and under environmental conditions different from the crop conditions of the southeast of Brazil [
The BM211 marker was located separating two QTLs in the GL 8 [
The ME1 marker was initially marked in the GL 1 by Blair et al. [
The BM184 marker was initially mapped in the GL 11 [
The BM187 marker was mapped in the GL 6, in which up to now only one QTL was identified for resistance to white mold,
The BMd42a marker is described in the GL 10 [
The SSR markers PVM02TC116, PV188, PV74, PVESTBR_185, PVESTBR_204, BMc94, and BMc83, together with the AFLP markers EAAG/MCAG224, EACC/MCAT141, EACC/MCAT126, and EACA/MCAT148, were significant in identification of QTLs of resistance to white mold; however, they have not been reported in the literature.
In this study, the ME1 and PVESTBR_204 markers are those that are closest to the QTLs—at 0.82 cM and 1.08 cM, respectively. These QTLs have the effect of increasing resistance to white mold. However, the QTL identified as PVESTBR_204 has low heritability (3.39%) and is not promising for MAS.
The PVESTBR_185 and BMc94 markers identified highly significant QTLs with high heritability (75.71% and 84.94%, resp.); however, they segregate apart from the QTL, at 75.33 cM and 50.76 cM, respectively, and they are thus not efficient for MAS.
As for the markers BM187, BMc83, and EAAG/MCAG224, in spite of identifying QTLs with moderate to high magnitude heritability, these QTLs have the effect of decreasing resistance to white mold. The other markers identified QTLs of low heritability and were thus not efficient.
In this study, the SRAP markers did not identify QTLs; however, their efficiency has been reported in the literature. Soule et al. [
The “moving away” method under the Bayesian approach proved to be efficient in identification of QTLs when a genetic map is not observed due to the low density of tags. In this respect, new studies may be conducted for the purpose of estimating the position and order of the QTLs in the genome using a consensus map.
The ME1 and BM211 markers are near the QTLs, with the effect of increasing resistance to white mold and they are of high heritability; they are therefore promising for marker-assisted selection in progenies derived from cultivars adapted to the conditions of the southeast of Brazil.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG), and the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for financial support for the study and granting scholarships.