Biofortification breeding of common bean (Phaseolus vulgaris L.)

Publiceret Januar 2009

Common bean (Phaseolus vulgaris L.) is one of five cultivated species from the genus Phaseolus and is a major grain legume crop, third in importance after soybean and peanut, but first in direct human consumption (Broughton et al. 2003).  Common beans originated in Latin America and have two primary centers of origin in the Mesoamerican and Andean regions that are easily distinguished by molecular means (Blair et al., 2006).  Major producing countries for national consumption are Brazil and Mexico; while the United States, Canada, Argentina and China are all exporting countries.   The crop is also important in a range of developing countries of Central America, of the Andean region of South America and of Eastern and Southern Africa (Singh, 1999).  In these regions, beans are grown both for subsistence agriculture and for regional markets where they play an important role in food security and income generation.   Common beans are important for nutritional well-being as well as poverty alleviation among consumers and farmers with few other food or crop options.  Much of the world’s bean production is on small farms ranging from 1-10 ha in size.   Multiple commercial seed types or horticultural classes exist based on seed color with white, yellow, cream, brown, pink, red, purple, black and mottled, pinto or striped seed types popular in different regions of the world and with different cultures (Voysest, 1994: Schoonhoven and Voysest, 1991). 

Per capita consumption varies with each producing and consuming country and also among regions within a country depending on consumer preferences but can be as high as 66 kg/capita/year in Rwanda and parts of western Kenya (Broughton et al. 2003).   Averages in the Americas are from 4-5 kg/capita/year in the United States, to more than 10 kg/capita/year in Brazil to as much as 35 kg/capita/year in Nicaragua.   These quantities of bean can provide substantial amounts of both protein and calories in the diet.  In nutritional terms, beans are often called the “poor man’s meat” for their inexpensive price as a protein source and their rich content of minerals (especially iron and zinc) and vitamins (Beebe et al., 2000).  In humans, iron is essential for preventing anemia and for the proper functioning of many metabolic processes while zinc is essential for adequate growth and sexual maturation and for resistance to gastro-enteric and respiratory infections, especially in children (Bouis, 2003).

In production terms, two general types of common beans are grown:  bush beans as a short season crop and climbing beans as a long-season crop (Shoonhoven and Voysest, 1991).  Bush beans produce a crop in as little as 65 days and may yield up to 2.5 t/ha per season (although average yields in Latin America are between 600-800 kg/ha and yields in Eastern and Southern Africa are lower still).  Climbing beans, on the other hand, have a slightly longer growing season (100-120 days; some even up to 240 days) and have a yield potential of 4.5 t/ha.  One advantage of bush beans over climbing beans is that in tropical regions two seasons might be grown, however early maturing climbing beans with adaptation to lower elevations have the potential to be grown in two seasons as well. Bush and climbing beans in small farmer fields are often intercropped, or used as a relay crop and planted at the end of the season to take advantage of residual moisture in the soil and are often not captured by official statistics.  One advantage of climbing beans over bush beans is that they fix larger amounts of nitrogen.

Figure 1. Steps in the nutritional breeding pathway: A) Germplasm screening (common bean diversity from a germplasm bank); B) Advanced line evaluation (inter-specific crosses with 97 ppm iron and 56 ppm zinc); C) Participatory Plant Breeding (Andean high iron advanced line).

Biofortification of common beans

As mentioned earlier, common bean is a valuable source of protein, minerals and vitamins.  In terms of biofortification, improvement of mineral content is advantageous precisely because the baseline grain iron content is high at 55 ppm (mg/kg) and variability for the trait is great, ranging up to 110 ppm, allowing initial breeding attempts to be much more successful than in the cereals in overall iron and zinc content increases (Beebe et al., 2000).  In addition, unlike many cereals that are polished before eating, resulting in significant loss of nutrients, common beans are consumed whole, thus conserving all their nutritional content.   Estimates for the Harvest Plus challenge program on biofortification are that an addition of approximately 40 ppm to baseline iron levels in common bean can meet a large proportion of the recommended daily intake of iron (Graham et al, 1999; Welch et al., 2000).  This target level takes into account the amount of bean consumed by the undernourished, any loses during storage, cooking and processing, and the extent humans will take up and absorb the extra iron.  Given a diet based approach of combining several biofortified foods, for example high iron beans and high iron rice or maize, this level could be even more rapidly enhanced.  The target areas for biofortified beans are in iron deficiency anemia prone areas of Latin America and Eastern and Southern Africa where the crop is important and consumption is high, such as the Central America, Northeast Brazil and the Great Lakes region of Africa.  The steps in common bean biofortification are outlined in Figure 1 and as separate sections below.

Germplasm Screening

In any breeding program, germplasm screening for a trait of interest is an important first step to genetic improvement.  In the case of biofortification, nutritional breeding also starts with assembly of parental germplasm for crosses based on the evaluation of a large amount of genetic material.  For beans, the CIAT core collection of 1400 genotypes with half from each genepool, was the starting point for screening of mineral traits.   A range of 30 to 110 ppm iron and 25 to 60 ppm zinc was found in the germplasm analyzed and the high iron genotypes, G14519 and G21242 were selected to initiate crosses (Beebe et al, 2000, Islam et al., 2002).  This variability for iron or zinc content is slightly larger than what was found in analysis of more limited range of genotypes (e.g. House et al., 2002; Guzman-Maldonado et al., 2003; Moraghan and Grafton 1999; Moraghan et al., 2002).  In addition, screening of advanced lines within each of the genepools has been important for identifying potential commercial type parents, as many of the core collection high iron or high zinc lines were of non-commercial seed types.  The range of mineral accumulation in the two gene pools of common bean (Andean and Mesoamerican) is similar, although many Andean beans or inter gene-pool hybrids have higher iron concentration than Mesoamerican beans (Islam et al., 2002).

For breeding, it has also been important to evaluate locally available germplasm to have a baseline of information on the nutritional traits.  Screening of local germplasm of the countries involved in biofortification has included a range of Andean varieties (Blair et al., 2005; Astudillo and Blair, 2009), a regional collection of Eastern and Southern African released varieties and a large collection of Rwandan genotypes (this laboratory, unpublished).   Additional diversity for mineral concentration has been found in wild or weedy germplasm (Guzman-Maldonado et al., 2004; unpublished data, this laboratory).  Finally, screening of related species such as P. coccineus or P. dumosus and P. acutifolius, has been used to identify high iron content genotypes in the secondary and tertiary genepools, respectively. 

One characteristic of this stage of nutritional breeding is that it has allowed the development of methodologies for nutritional analysis which was very important to avoid mineral contamination in iron analysis which is a common problem.   While initial sample preparation was done with regularly harvested seed and aluminum grinding equipment in a modified Retsch mill, currently we are using a more careful method for sampling and analysis that involves hand harvesting and threshing of grain followed by seed milling in Teflon chambers with zirconium grinding balls (Blair et al., 2005; Astudillo and Blair, 2009).  We have also 1) determined the validity and precision of various mineral analyses methods such as atomic absorption spectrophotometry versus inductively coupled plasma – optical emission spectrometry, 2) calibrated near infrared reflectance spectroscopy for iron and 3) measured the effect of surface cleaning with ethanol or a damp cloth to remove dust from the seed before grinding (this laboratory, unpublished). 

As part of germplasm screening and evaluation, the stability of the genotypes for a given trait is usually evaluated.  In the case of iron and zinc accumulation, GxE was best evaluated with the most promising local germplasm and potential parents to determine if mineral accumulation was stable across sites.  As part of the Harvest Plus challenge program, GxE has been analyzed for fast-track East African and Andean Region landraces, as well as for a high-mineral nursery of advanced lines tested across sites in Central and South America.

Breeding of high mineral bean varieties

An intitial goal in the biofortification of common beans has been to produce varieties with 80% more iron content and 40% more zinc while maintaining or improving the properties that farmers and consumers require in a variety, such as adaptation to abiotic or biotic stresses and seed shape or color.  Breeding at the International Center for Tropical Agriculture (CIAT) is concentrating on a range of commercial classes and on both genepools of common bean, the large-seeded Andeans and the small-seeded Mesoamericans using various strategies have been used for the current biofortification breeding effort, including backcrossing, recurrent selection and various permutations of gamete and pedigree selection.  

The first strategy applied for biofortification breeding of Andean beans at CIAT was backcrossing with gamete selection, where selection was applied in the BC1F1 plant stage and again in the BC1F3 generation at which point mineral analysis was used to select the most promising advanced lines in the following generation.  Two pedigrees were used in this stage: CAL 96 x (CAL 96 x G14519) and CAL 143 x (CAL 143 x G14519) based on the recurrent parents CAL96 and CAL143 which have been released as varieties in Uganda and Malawi, respectively.  Both of these red mottled beans were considered valuable recurrent parents because of their wide adaptation in Eastern and Southern Africa and in the Andes of South America. 

From these crosses, a number of BC1F4 Andean nutrition (NUA) lines were selected and used in multiple site, on-station and on-farm testing in Colombia, Kenya and Malawi.   Initial releases of the best lines with an increase of up to 25 ppm seed iron are planned for Bolivia, Colombia, Malawi and Zimbabwe with seed multiplication as part of the Harvest Plus, Fontagro and Agrosalud projects in all these countries as well as Kenya.  Participatory plant breeding (PPB) has been used in the process of nutritional improvement to define the best advanced lines and to simultaneously increase seed supply for variety promotion and nutritional testing.   In one case, PPB identified a moderately high iron genotype that was preferred by farmers due to its earliness and high production potential. 

Currently, backcrossing, multi-site testing and PPB is underway at CIAT or being planned for the improvement of climbing beans for nutritional quality based on the previous success of the bush bean crosses and the value of the high iron source, G14519.    Climbing beans were selected as a good delivery system for biofortified grain, because of their high potential impact and intrinsic advantages of high yield in small space, large grain, good nitrogen fixation and weed suppression, flexibility for various cropping systems (e.g. Maize x Beans), which makes them a good alternative for small farms.  The recurrent parents being used for climbing bean nutritional improvement have been BCMNV resistant genotypes with mid-altitude adaptation which are expected to also improve on the major productivity constraints of virus and heat stress susceptibility.  Markers assisted selection (MAS) is being used to select for both BCMV and angular leaf spot (ALS) disease resistance.

Within the Mesoamerican bean gene pool, breeding for nutritional value likewise focuses on combining high grain mineral content with preferred agronomic traits that will make new varieties attractive to farmers, so as to speed adoption and ultimate impact. Since the small and medium seeded Mesoamerican bush beans are often grown in more stressful environments, a top priority has been to combine high iron and zinc with drought tolerance, as well as resistance to important diseases such as bean golden yellow mosaic virus and angular leaf spot. Priority grain classes for this effort are small black, small red and cream striped grain types.  In a program that seeks to combine multiple quantitative traits, a recurrent selection scheme is being implemented, having completed two cycles of selection and attained an increase in iron of about 60% over standard varieties.

Inter-specific crosses to introgress high iron from related species appears to hold promise especially for the Mesoamerican beans where it has been difficult to reach levels of iron of 90+ ppm.  CIAT has made inter-specific crosses with high iron accessions of P. dumosus (P. polyanthus) and P. coccineus which have expressed as high as 127 ppm iron in grain harvested under greenhouse conditions (although field harvested grain is often lower in iron). One backcross of the interspecific F1 has been performed and so far a small number of F3 families average nearly 20 ppm more than high iron checks, while the range among families has been as high as 98 ppm.  While the highest levels are normally found in materials of poor adaptation (late maturity, poor pod set), even lines of 80 ppm iron would represent an important gain in genetic potential. Although interspecific crosses with an Andean bean (CAL 96) were also created, these did not show evidence of significant introgression of the high iron trait.  It is expected that individual lines from the inter-specific crosses will have different patterns of introgression from the related species that will determine the degree of expression of the high iron trait. 

In conclusion, we have outlined the steps typical of any crop improvement program starting with germplasm screening and followed by breeding, but with a focus on the nutritional traits of seed mineral content.   Within the breeding step, we have emphasized the use of hybridization to create wide crosses, the strategies for selection that are part of traditional breeding methods and the potential of molecular markers that are part of a modern plant improvement.


Astudillo C, Blair MW  (2009)  Contenido de hierro y cinc en la semilla y su respuesta al nivel de  fertilización con fósforo en 40 variedades de fríjol colombianas.  Agronomía Colombiana 26: 471-476.

Beebe S, Gonzalez AV, Rengifo J  (2000)  Research on trace minerals in the common bean.  Food Nutr. Bulletin 21:387-91.

Blair MW, Giraldo MC, Buendia HF, Tovar E, Duque MC, Beebe SE (2006) Microsatellite marker diversity in common bean (Phaseolus vulgaris L.)  Theor Appl Genet 113: 100–109.

Blair MW, Astudillo C, Grusak M, Graham R, Beebe S (2009) Inheritance of seed iron and zinc content in common bean (Phaseolus vulgaris L.). Molecular Breeding 23: 197-207.

Blair MW, Astudillo C, Restrepo J, Bravo LC, Villada D, Beebe SE  (2005)  Análisis multi-locacional de líneas de fríjol arbustivo con alto contenido de hierro en el departamento de Nariño.  Fitotecnia Colombiana 5: 20-27.

Bouis, H.E.  2003.  Micronutrient fortification of plants through plant breeding:  can it improve nutrition in man at low cost?  Proc. Nutr. Soc. 62:403-411 

Broughton WJ, Hernandez G, Blair MW, Beebe SE, Gepts P, Vanderleyden J (2003) Beans (Phaseolus spp.) – Model Food Legumes.  Plant and Soil 252: 55-128.

Cichy KA, Forster S, Grafton KF, Hosfield GL  (2005)  Inheritance of seed zinc accumulation in navy bean.  Crop Sci. 45:864-870.

Gelin JR, Forster S, Grafton KF, McClean P, Rojas-Cifuentes GA  (2007)  Analysis of seed-zinc and other nutrients in a recombinant inbred population of navy bean (Phaseolus vulgaris L.).  Crop Sci  47:1361-1366

Graham R, Senadhira D, Beebe S, Iglesias C, Monasterio I   (1999)   Breeding for micronutrient density in edible portions of staple food crops: conventional approaches.  Field Crops Res 60:57-80

Guzman-Maldonado SH, Acosta-Gallegos J, Paredes-Lopez, O. (2004) Protein and mineral content of a novel collection of wild and weedy common bean (Phaseolus vulgaris L.). J Sci Food Agric, 80:1874–1881

Guzman-Maldonado SH, Martínez O, Acosta-Gallegos J, Guevara-Lara FJ, Paredes-Lopez O (2003) Putative quantitative trait loci for physical and chemical components of common bean.  Crop Sci 43:1029–1035

House, W.A., R.M. Welch, S. Beebe. Z. Cheng.  2002.  Potential for increasing the amounts of bioavailable zinc in dry beans through plant breeding.    J. Sci. Food Agric. 82:1452-1457.

Islam FMA, Basford KE, Jara C, Redden RJ, Beebe SE  (2002)  Seed compositional and disease resistance differences among gene pools in cultivated common bean. Genet Resour Crop Evol 49:285–293

Moraghan JT, Grafton K  (1999)  Seed zinc concentration and the zinc-efficiency trait in navy bean.  Soil Sci Soc Am J 63:918-922

Moraghan JT, J Padilla, JD Etchevers, K Grafton JA Acosta-Gallegos  (2002)  Iron accumulation in seed of common bean.  Plant and Soil 246:175-183

Singh SP (ed) (1999) Common Bean Improvement for the Twenty-First Century.  Kluwer Acad. Publ., Dordrecht, Germ.

Singh SP, Westermann DT  (2002)  A single dominant gene controlling resistance to soil zinc deficiency in common bean.  Crop Sci  42:1071-1074

Schoonhovern A, Vosyest O (eds.) (1991) Common Beans:  Research for Crop Improvement.  C.A.B.  Int., Wallingford, UK.

Voysest O, Valencia M, Amezquita M  (1994)  Genetic diversity among Latin American Andean and Mesoamerican common bean cultivars. Crop Sci. 34:1100-1110

Welch RM, House WA,  Beebe S, Cheng Z  (2000)  Genetic selection for enhanced bioavailable levels of iron in bean (Phaseolus vulgaris L.).  J Agr Food Chem 48: 3576-3580