Towards effective gene pyramiding for disease resistance and trait selection in common bean (Phaseolus vulgaris L.)
Major common bean fungal and viral diseases concurrently occur in farmer’s fields leading to complete crop failure when susceptible varieties are grown. Inherent genetic resistance is thus the cheapest source of resistance against diverse pathogenic strains of key diseases affecting common bean as it reduces the need for frequent chemical control application. Deployment of multiple disease resistance bean varieties in farmer’s fields would therefore go a long way in reducing yield losses at farm level. Gene pyramiding is one of the breeding methodologies used to develop multiple disease resistant varieties. As a result, pyramided climbing bean populations targeting six major resistance genes (Co-42, Co-5 for Anthracnose; Phg-2 for Angular leaf spot (ALS); I and bc3 for Bean common mosaic virus (BCMV) and “P.ult” for Pythium ultimum root rots) was developed at CIAT, Uganda by hybridizing four parents (G2333 with Co-42 and Co-5, Mexico54 with Phg-2, MCM5001 with I and bc3 and MLB-4989A with “P.ult”) using marker-assisted backcrossing method. The objectives of this research was to determine recombination fractions and genetic linkage among key disease resistance genes pyramided, evaluate agronomic qualities of pyramided bean lines and to determine response to selection diseases and yield traits of pyramided bean lines based on field evaluations. For linkage analysis, the number of recombinants among linked gene pairs (Co-42/Phg-2 and Co-5/“P.ult”) was determined by counting Sequence Characterized Amplified Region (SCAR) marker’s DNA amplicons on electrophoresis gel pictures of bean lines at both BC3F6 and F2 generations. Recombination fraction r, among gene pairs, the probability of linkage and LOD scores were computed using the statistical relationship of likelihood which assumes a binomial distribution. The agronomic traits of pyramided bean populations were evaluated in four randomized incomplete blocks, with three replicates across two seasons in 2015 at CIAT, Uganda. Thirty seeds were planted per entry in plots of two rows, measuring three meters long, with spacing of 20 cm x 60 cm within and between rows, respectively. Traits evaluated were; days to 50% flowering (DF), number of internodes (Nodeno), number of flower buds per inflorescence (FLB), pods per plant (PDPL), seeds per plant (SPLA), weight of 100 seeds (100SW), growth habit (GH) and flower colour (FLCOL), while seed shape (SS) and primary seed color (PSC) were recorded after harvest. Data was summarized and analyzed for mean comparisons, correlations and trait frequencies recorded. Finally, to determine responses to selection, populations evaluated included parents, progenies of pyramids at BC3F9 generation and local checks. The trial was planted in 2016 in Uganda (Kabale and Kawanda) and Northern-Tanzania (Maruku and Kitengule) in randomized incomplete blocks with three replications. Sixty seeds were planted per xiv population in plots of three rows of three meters long, with spacing of 20 cm x 60 cm within and between rows respectively and climbing beans supported on wooden stakes. Yield traits recorded on all plant stand, were, days to 50% flowering, number of pods per plant and weight of 100 seeds. Diseases recorded were, ALS and BCMV on leaves and Anthracnose on pods using disease rating scale of 1-9. Traits means were compared using the combined un-balanced ANOVA, while Principal component analysis (PCA) was used to account for variations among bean populations. Narrow sense heritability of traits evaluated was computed using parent-offspring regression. Data and appropriate parameters were summarized and incorporated into the breeder’s equation (R= ih2S) for computing response to selection. The results were as follows; gene pair Co-5/“P.ult” exhibited weak linkage (r = 32 cM with a high LOD score of 9.20) in BC3F6 population, but relatively stronger linkage in F2 population (r = 21 cM with a high LOD score of 18.67). However, gene pair Co-42/Phg-2 had very weak linkage in BC3F6 population (r = 47 cM with a low LOD score of 0.16) and in F2 (r = 44 cM with a low LOD score of 0.70). The weak and incomplete linkages found, suggested that each of the four genes above had to be tagged with a specific marker during selection. All agronomic traits evaluated among pyramided bean populations across the two seasons (2015A and 2015B) had significant differences (p<0.05). Parental hybridisation improved number of pods per plant, number of seeds per plant and weight of 100 seeds. However, days to 50% flowering, number of internodes and numbers of flower buds formed per inflorescence were significantly different (P<0.05) among pyramids and blank populations (sister lines to pyramids). Transgressive segregation was observed among pyramided progeny lines. Positive and weak correlations (r < 0.5, P<0.01) were observed between number of genes pyramided and agronomic traits evaluated. The mean yield traits of the recurrent parent (G2333) were not yet recovered in segregates by the third backcross, while phenotypic similarity between pyramided lines and recurrent parent was still low for the number of pods per plant (12.5%) and weight of 100 seeds (10.7%). Response to selection was positive for mean days to flowering (five days) and negative for mean number of pods per plant and weight of 100 seeds (-9.5 pods and -1.9 grams respectively). Diseases had positive undesirable positive responses to selection. Narrow sense heritability was high for 100 seed weight (0.62) and BCMV (0.52), showing that these traits are easy to improve. Pyramided bean lines however still segregated at BC3F9 for diseases and yield traits. The following are conclusions and recommendations; selection with markers tightly linked to disease resistance genes helped to reduce the time for developing pyramids. Agronomic traits of progenies of segregating pyramided bean lines could be evaluated further and fixed in late xv generations such as BC3F11. The yield penalties associated with gene pyramiding reported in literature should be tested using fixed bean populations or genotypes. The superior bean lines identified among different populations, such as G63, G65 and G66 need further testing using multi-location trials to assess their adaptability and stability in mid and high altitude areas in Uganda and neigbouring countries.