Juan Carlos Rosas

Root architectural tradeoffs for water and phosphorus acquisition

Root architectural traits that increase topsoil foraging are advantageous for phosphorus acquisition but may incur tradeoffs for the acquisition of deep soil resources such as water. To examine this relationship, common bean genotypes contrasting for rooting depth were grown in the field and in the greenhouse with phosphorus stress, water stress and combined phosphorus and water stress. In the greenhouse, water and phosphorus availability were vertically stratified to approximate field conditions, with higher phosphorus in the upper layer and more moisture in the bottom layer. Under phosphorus stress, shallow-rooted genotypes grew best, whereas under drought stress, deep-rooted genotypes grew best. In the combined stress treatment, the best genotype in the greenhouse had a dimorphic root system that permitted vigorous rooting throughout the soil profile. In the field, shallow-rooted genotypes surpassed deep-rooted genotypes under combined stress. This may reflect the importance of early vegetative growth in terminal drought environments. Our results support the hypothesis that root architectural tradeoffs exist for multiple resource acquisition, particularly when resources are differentially localised in the soil profile. Architectural plasticity and root dimorphism achieved through complementary growth of distinct root classes may be important means to optimise acquisition of multiple soil resources.

Breeding beans for resistance to terminal drought in the Lowland tropics

In the lowland regions of Latin America, a large proportion of beans are sown at the beginning of a dry season where a guaranteed terminal (end-of-season) drought will reduce yields. This study was undertaken to identify lines within two black bean recombinant inbred line (RIL) populations with resistance to terminal drought. The two RIL populations were developed from crosses between a drought resistant line, B98311 from Michigan, with TLP 19 and VAX 5, two lines from CIAT with improved disease resistance and adaptation to growing conditions in Latin America. The RIL populations were evaluated in experiments conducted in Zamorano, Honduras and Veracruz, Mexico under drought stress and well-watered (non-stress) treatments. Yields were reduced in each experiment by drought and the fungal pathogen, Macrophomina phaseolina. Drought stress, disease pressure and low yields contributed to high coefficients of variation (CV), which made it difficult to select superior lines. Selection was based on rank of geometric mean (GM) yield calculated from the yield in the stress and non-stress treatments. One RIL, L88-63 ranked first in GM yield at both locations. Subsequent testing in Honduras and Michigan confirmed the high yield potential and broad adaptation of L88-63. Breeding beans for drought resistance in lowland tropical environments should also include breeding for resistance to M. phaseolina.

A method for screening Phaseolus vulgaris L. germplasm for preferential nodulation with a selected Rhizobium etli strain

As part of a breeding program to improve the nitrogen-fixing symbiosis between common bean (Phaseolus vulgaris) and Rhizobium etli, we developed a rapid screen for common bean accessions that preferentially nodulate with KIM5s, a high nitrogen fixing strain of R. etli. We constructed a mutant of KIM5s that did not fix nitrogen (Fix-) but was otherwise indistinguishable from KIM5s. We screened plants for symptoms of nitrogen deficiency when grown in a Honduran soil containing indigenous common bean-nodulating rhizobia (104 per gram) and KM6001, the Fix- mutant of KIM5s (104/seedling added 7 days after planting). Leaf color was scored on a scale of 1 to 5, in which 1 was dark green and 5 was bright yellow. Of 820 genetically diverse accessions of P. vulgaris screened, 51 were scored 1, 626 were scored 2 or 3, and 143 were scored 4 or 5. Selfed seed was produced from common bean plants of the accessions scored 1, 4 or 5. Twenty-four accessions that scored 1, and 58 that scored 4 or 5 were screened in soil containing indigenous rhizobia and the wild type KIM5s (Fix+), and nodule occupancy was determined by antibiotic resistance. On the 24 common bean accessions that were scored 1, KIM5s occupied 0-6% of the nodules, on 26 of the accessions that were scored 4 or 5, KIM5s occupied 90%-100% of the nodules, and on the remaining 34 that scored 4 or 5, there was a distribution of nodule occupancy. Foliar color was highly correlated with nodule occupancy (r = 0.786,p = 0.01). The results indicate that the rapid visual screen using the Fix- mutant accurately identified common bean accessions that preferentially nodulate with the wild-type KIM5s (Fix+) strain in soil containing indigenous rhizobia. This screen will facilitate introduction of the preferential nodulation trait into superior cultivars and provides the foundation for studies of the genetic basis of preferential nodulation.

Distribution of nitrogen in common bean (Phaseolus vulgaris L.) genotypes selected for differences in nitrogen fixation ability

The improvement of N2 fixation in legumes may lead to increased yields and reduced fertilizer requirement. Levels of N2 fixation were determined for three cultivars and nine progeny lines from two inbred backcross common bean (Phaseolus vulgaris L.) populations that were grown at Hancock, Wissconsin in 1984 and 1985 using 15N-depleted (NH4)2SO4. The high N2-fixing line ‘Puebla 152’ was the donor parent for both inbred backcross populations and the cultivars ‘Porrillo Sintetico’ and ‘Sanilac’ were the recurrent parents for populations 21 and 24, respectively. Total N yield, fixed N2 and % N derived from the atmosphere were determined for whole plants and plant parts at the R3 (50% bloom) and R9 (maturity) growth stages. Significant year-by-line interactions were found for N2 fixation traits among the population 21 lines and parents, but not for population 24 lines and their parents. Measures of N2 fixation at R3 were inadequate to predict N2 fixation at R9. Population 24 lines and parents differed for N2 fixation ability at R9, and fixed N2 was correlated with maturity. The recovery of an inbred backcross progeny line, 24-21, which matured earlier and fixed more N2 than the recurrent parent ‘Sanilac’ indicated that N2 fixation was heritable and that favorable alleles, independent of maturity, were recovered from a late-maturing, high N2-fixing donor parent by utilizing the inbred backcross breeding method. Since most fixed N2 and non-fixed N (>80%) was found in the seeds at maturity, and most lines did not vary for the distribution of nitrogen throughout the plant, selection for improved remobilization of nitrogen to the seed to increase yield is impractical in this genetic material. The highest N2-fixing lines tended to have high and similar % Ndfa in all plant parts.

The economic impact of bean disease resistance research in Honduras

This paper presents evidence of recent adoption rates of disease resistant bean varieties (RVs), the farm-level benefits of RV adoption, and the ex post rate of return to disease resistant bean research in Honduras. Results from a farm-level survey in 2001 in the two principal bean-producing regions of the country show that 41-46% of bean farmers (depending upon the season) have adopted an RV, and that adoption is neutral with respect to farm-size and market orientation. An expected utility framework was used to estimate the farm-level benefits of RV adoption, using a combination of experimental trial and farm-level survey data. Adopters gain the equivalent of 7-16% (depending on the variety) in bean income from the yield loss averted through RV use. The ex post rate of return to disease resistant bean research in Honduras from 1984 to 2010 is 41.2%.

Uromyces appendiculatus in Honduras:Pathogen Diversity and Host Resistance Screening

Bean rust, caused by the fungus Uromyces appendiculatus, is a major constraint for common bean production worldwide. Virulence of U. appendiculatus collected from wild and cultivated Phaseolus spp. was examined in 28 locations across Honduras. Host accessions representing wild and domesticated Phaseolus spp. collected at the same sampling locations were evaluated for resistance against U. appendiculatus. In total, 91 pathotypes were identified from 385 U. appendiculatus isolates according to their virulence on each of the 12 host differentials. No significant difference in pathogen total virulence, measured as the mean disease score, was found between locations. However, significant differences were found in pathotype virulence among isolates collected from different Phaseolus spp. within a location. Moreover, when locations were compared on the basis of pathotype occurrence and frequency, differences among locations were evident. No two locations had the same pathotype composition. The most common pathotype was virulent on 9 of the 12 differential lines. A high number of resistant accessions were identified in Phaseolus coccineus and P. lunatus. Although most wild P. vulgaris accessions were highly susceptible, rust resistance was observed in P. vulgaris landraces collected from farmers fields. Thirty-two (52%) of the accessions screened showed intermediate to high levels of resistance and, of those, 16% were P. coccineus accessions. Our findings support the hypothesis that interaction of U. appendiculatus in host populations composed of diverse Phaseolus spp. and genotypes has favored highly diverse and virulent pathotypes, creating a center for virulence diversity of the pathogen in Honduras. The high percentage of intermediate and highly resistant accessions identified in the present study supports the strategy of collecting plants from the center of diversity of a pathogen or in locations with high incidence of disease and pathogen diversity to maximize the probability of identifying new sources of resistance.

Candidate gene markers associated with cold tolerance in vegetative stage of rice (Oryzasativa L.)

Low temperature inhibits seedling emergence and early vegetative growth of rice (Oryza sativa L.), especially in cultivars of the subspecies indica, which are the most productive in the temperate region of South America. To improve breeding efficiency, we identified molecular markers associated with cold response related phenotypes, such as membrane integrity, photoinhibition and visual assessment of damage. A candidate gene approach was used to select genomic simple sequence repeat (gSSR) markers. For marker-trait association two strategies were used, candidate gene association mapping (CGAM) and single marker analysis (SMA) in a segregating F2:3 population. Thirty-six gSSR were identified but only 14 were polymorphic in the CGAM population. Using this approach, we found marker RM22034 associated with membrane integrity, RM6547 and RM14978 with photoinhibition, and RM144 with visual assessment of damage. The SMA strategy identified markers RM6651, RM14978, RM22034 and RM4 associated with membrane integrity and RM22034 with photoinhibition. All markers are currently available for marker assisted selection in rice breeding programs located in temperate regions.

Integrating molecular markers and environmental covariates to interpret genotype by environment interaction in rice (Oryza sativa L.) grown in temperate areas

Understanding the genetic and environmental basis of genotype × environment interaction (G×E) is of fundamental importance in plant breeding. If we consider G×E in the context of genotype × year interactions (G×Y), predicting which lines will have stable and superior performance across years is an important challenge for breeders. A better understanding of the factors that contribute to the overall grain yield and quality of rice (Oryza sativa L.) will lay the foundation for developing new breeding and selection strategies for combining high quality, with high yield. In this study, we used molecular marker data and environmental covariates (EC) simultaneously to predict rice yield, milling quality traits and plant height in untested environments (years), using both reaction norm models and partial least squares (PLS), in two rice breeding populations (indica and tropical japonica). We also sought to explain G×E by differential quantitative trait loci (QTL) expression in relation to EC. Our results showed that PLS models trained with both molecular markers and EC gave better prediction accuracies than reaction norm models when predicting future years. We also detected several milling quality QTL that showed a differential expression conditional on humidity and solar radiation, providing insight for the main environmental factors affecting milling quality in temperate rice growing areas.