| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Plant, Soil and Entomological Sciences, Univ. of Idaho, 3793 North 3600 East, Kimberly, ID 83341-5076
* Corresponding author (singh{at}kimberly.uidaho.edu)
ABSTRACT
Knowledge, access, and use of diversity available in cultivated and wild relatives are essential for broadening the genetic base of cultivars to sustain improvement. My objectives are to review briefly the origin, domestication, and organization of genetic diversity in Phaseolus beans, highlight production problems and traits deficient in the common bean (P. vulgaris L.) cultivars, cite sources of useful gemplasm, and review progress achieved in broadening the genetic base of cultivars. Phaseolus beans originated in the Americas. Only five species, P. acutifolius A. Gray, P. coccineus L., P. lunatus L., P. polyanthus Greenman, and P. vulgaris L. were domesticated. The interspecies diversity in relation to common bean is organized in primary, secondary, tertiary, and quaternary gene pools. Intraspecies diversity in common bean is separated into two major gene pools (Andean and Middle American). Cultivars are further divided into races, each with their distinguishing characteristics. Also, cultivars of dry seed and snap bean exist. Abiotic and biotic stresses limit common bean production. The genetic base of cultivars within market classes is narrow and inadequate level of resistance to common bacterial blight [caused by Xanthomonas campestris pv. phaseoli (Smith) Dye] and white mold [caused by Sclerotinia sclerotiorum (Lib) de Bary] exist in cultivars. High levels of resistance to these and other desirable traits exist in the relatives and gene pools of P. vulgaris. Early maturity, adaptation to higher latitude, upright plant type, high pod quality and seed yield, and resistance to Bean common mosaic virus (a potyvirus) and/or rust [caused by Uromyces appendiculatus (Pers.:Pers.) Unger] have been bred into cultivars. However, most of the genetic variability available in the common bean races, gene pools, and wild relatives remains to be utilized. To maximize and sustain bean production, high yielding, high quality cultivars that are less dependent on water, fertilizer, pesticides, and manual labor should be developed. This need warrants sustained, comprehensive, and integrated genetic improvement, in which favorable alleles from cultivated and wild relatives are accumulated in superior cultivars. A three-tiered breeding approach involving: (i) gene introgression from alien germplasm, (ii) pyramiding favorable alleles from different sources, and (iii) simultaneous improvement of multiple traits for common bean cultivars would be the most appropriate strategies to meet theses needs.
Abbreviations: BCMV, Bean common mosaic virus • BGMV, Bean golden mosaic virus • CBB, common bacterial blight • CIAT, International Center for Tropical Agriculture • QTL, quantitative trait loci
THE GENETIC BASE of most common bean cultivars within a market class is narrow (Adams, 1977; McClean et al., 1993; Voysest et al., 1994; Zaumeyer, 1972). This is because only a small fraction of wild common bean populations were domesticated (Gepts et al., 1986). The narrow genetic base of cultivars is attributed also to stringent quality requirements imposed by both processors and consumers (Ghaderi et al., 1984; Hosfield et al., 2000; Myers, 2000; Wang et al., 1988), limited use of exotic germplasm (Miklas, 2000; Silbernagel and Hannan, 1988, 1992), and conservative breeding strategies employed by breeders (Singh, 1992).
To broaden the genetic base and maximize gains from selection, it is essential to accumulate favorable alleles from the crop's cultivated and wild populations and alien species. My objectives are to: (i) provide an overview of the origin, domestication, and organization of genetic diversity in the Phaseolus beans; (ii) highlight production problems and traits deficient in common bean cultivars; (iii) cite sources of useful germplasm in the common bean cultivars and wild relatives; and (iv) review recent successes achieved in broadening the genetic base of common bean cultivars.
Origin, Domestication, and Organization of Diversity
The genus Phaseolus is of American origin and comprises over 30 species (Debouck, 1991, 1999; Delgado Salinas, 1985; Maréchal et al., 1978; Westphal, 1974). Only five of these species, namely P. acutifolius A. Gray (tepary bean), P. coccineus L. (scarlet runner bean), P. lunatus L. (Lima bean), P. polyanthus Greenman (year-long bean), and P. vulgaris L. (common bean), were domesticated (Debouck, 1999, 2000; Gepts and Debouck, 1991). Among these species, common bean is the most widely grown, occupying more than 85% of production area sown to all Phaseolus species in the world.
Diversity among Phaseolus species in relation to common bean is organized into primary, secondary, tertiary, and quaternary gene pools (Debouck, 1999, 2000; Debouck and Smartt, 1995). The primary gene pool of common bean comprises both cultivars and wild populations. The latter are the immediate ancestors of common bean cultivars (Berglund-Brücher and Brücher, 1976; Brücher, 1988; Gentry, 1969; Kami et al., 1995; Kaplan, 1981; Miranda C., 1967; Weiseth, 1954). Wild populations are distributed from northern Mexico (Chihuahua) to northwestern Argentina (San Luis) (Gepts et al., 1986; Koenig et al., 1990; Toro et al., 1990). Moreover, common bean is a noncentric crop, with multiple domestication sites throughout the distribution range in Middle and Andean South America (Gepts et al., 1986; Harlan and De Wet, 1971). Hybrids between the wild and cultivated beans are fully fertile and have no major barriers (Motto et al., 1978; Singh et al., 1995). The exception is the occurrence of the Dl-1 gene in the Middle American and Dl-2 in Andean wild (Koinange and Gepts, 1992) and cultivated populations (Shii et al., 1980; Singh and Gutiérrez, 1984; Vieira et al., 1989). These genes may restrict gene introgression and exchange (Mumba and Galwey, 1999).
Common bean has evolved during domestication from extreme indeterminate climbing types to determinate bush types, from sensitivity to long photoperiod to insensitivity, from small- to large-seeded forms, from seed dormancy and water impermeability of the seed coat to lack of dormancy and water permeable seed coat, and from highly fibrous pod wall and shattering forms to lack of fibers and nonshattering types (Evans, 1980; Gepts and Debouck, 1991; Smartt, 1988). Major genes or quantitative trait loci (QTL) that influence domestication syndrome of the common bean have been identified and mapped (Freyre et al., 1998; Gepts, 1999; Gu et al., 1998; Koinange et al., 1996). These traits are growth habit (fin), photoperiod insensitivity (ppd, hr), pod fiber (St), seed dormancy, and seed weight.
The secondary gene pool of common bean comprises P. coccineus, P. costaricensis Freytag & Debouck, and P. polyanthus. These three species cross among themselves, and each can be crossed to common bean without embryo rescue, particularly when common bean is used as the female parent (Camarena and Baudoin, 1987; Ibrahim and Coyne, 1975; Manshardt and Bassett, 1984; Singh et al., 1997). However, hybrid progenies between crosses of common bean and any of the three species forming the secondary gene pool may be partially sterile, preventing the recovery of desired stable common bean phenotypes (Ibrahim and Coyne, 1975; Manshardt and Bassett, 1984; Wall, 1970). Crosses with common bean as a male parent are more difficult. Recombinants are unstable and there is a tendency to revert to the female-parent genotype (Ockendon et al., 1982; see reviews by Debouck, 1999; Hucl and Scoles, 1985; Smartt, 1970).
The tertiary gene pool of common bean comprises P. acutifolius and P. parvifolius Freytag. While these two species can be crossed without embryo rescue to produce fully fertile progenies (Singh et al., 1998b), crosses of common bean with these species require embryo rescue (Alvarez et al., 1981; Andrade-Aguilar and Jackson, 1988; Haghighi and Ascher, 1988; Mejía-Jímenez et al., 1994; Mok et al., 1978; Pratt, 1983; Singh et al., 1998b; Thomas and Waines, 1984). One or more backcrosses to the recurrent common bean parent are often required to restore fertility of the hybrids. Using P. acutifolius as the female parent of the initial F1 cross, and/or the first backcrossing the P. vulgaris x P. acutifolius hybrid on to P. acutifolius, is often more difficult than using P. vulgaris as the female parent of the initial cross and backcrossing the interspecies hybrid on to P. vulgaris (Mejía-Jiménez et al., 1994). The choice of parents (Federici and Waines, 1988; Mejía-Jiménez et al., 1994; Parker and Michaels, 1986) and use of the congruity backcross (i.e., backcrossing alternately to each species) over recurrent backcrossing (Haghighi and Ascher, 1988; Mejía-Jiménez et al., 1994) facilitate interspecific crosses of common and tepary beans, in addition to recovery of fertility and higher number of hybrid progenies. For additional information and earlier review of interspecific hybridization between common and tepary beans readers should refer to Waines et al. (1988).
Crosses of common bean with P. filiformis (Federici and Waines, 1988; Maréchal and Baudoin, 1978; Petzold and Dickson, 1987), P. angustissimus (Belivanis and Doré, 1986; Petzold and Dickson, 1987), and P. lunatus (Al-Yasiri and Coyne, 1966; Cabral and Crocomo, 1989; Kuboyama et al., 1991; Leonard et al., 1987; Mok et al., 1978; Rabakoarihanta et al., 1979) have been attempted without producing fertile viable hybrid progenies. Thus, for gene introgression purposes, these species would be considered in the quaternary gene pool of the common bean. Identification and cloning useful genes from these species, and successful regeneration and transformation of common bean, may facilitate gene introgression in the future.
Diversity in Common Bean Cultivars
There are two major commercial classes of common bean, snap and dry beans. Snap bean cultivars possess a thick succulent mesocarp and reduced or no fiber in green pod walls and sutures (Myers, 2000; Myers and Baggett, 1999; Silbernagel, 1986). The green pods are harvested for fresh, frozen, and canning purposes. Different market classes of snap bean cultivars are largely determined on the basis of pod shape (flat, cylindrical, or oval), color (dark green, light green, yellow, or purple), and length (or sieve size). Among snap bean cultivars, there can be a large variation in growth habit and adaptation traits.
Similarly, large variation in growth habit, phenological traits, seed size, shape, color, and canning and cooking qualities are found among dry bean cultivars (Singh, 1992; Voysest and Dessert, 1991). In the highlands of Peru and Bolivia, unique popping beans called ñuñas are grown. However, the largest production (>14 million hectares) and consumption of P. vulgaris in the world is of dry beans, followed by a much lower production of snap bean.
Genetic diversity in common bean is organized in large-seeded (>40 g 100-seed weight-1) Andean and small- (<25 g 100-seed weight-1) and medium- (25–40 g 100-seed weight-1) seeded Middle American gene pools (Evans, 1973, 1980). Further evidence for the existence of the two gene pools was provided by the relationship of seed size (small versus large) with (i) the Dl genes (Dl-1 versus Dl-2) and the F1 hybrid incompatibility (Gepts and Bliss, 1985; Singh and Gutiérrez, 1984; Vieira et al., 1989); (ii) phaseolin seed proteins (Gepts et al., 1986); (iii) allozymes (Koenig and Gepts, 1989; Singh et al., 1991c); (iv) morphological traits (Singh et al., 1991b); and (v) DNA markers (Becerra Velasquez and Gepts, 1994; Haley et al., 1994c; Khairallah et al., 1990). Singh (1989) described in detail the patterns of variation in common bean cultivars. Singh et al. (1991a) further divided the Andean and Middle American cultivated gene pools into six races: Andean (all large-seeded) = Chile, Nueva Granada, and Peru; Middle American = Durango (medium-seeded semi-climber), Jalisco (medium-seeded climber), and Mesoamerica (all small-seeded), each with their distinguishing characteristics, ecological adaptation, and agronomic traits. Beebe et al. (2000) reported the existence of additional diversity within Middle American races, especially a group of Guatemalan climbing bean accessions that did not group with any of the previously defined races.
Production Problems
Common bean suffers from both abiotic and biotic production constraints (Graham, 1978; Schwartz and Pastor-Corrales, 1989; Singh, 1992; Wortmann et al., 1998; Zaumeyer and Thomas, 1957). Among abiotic constraints, the most widely distributed are low soil fertility, particularly deficiency of nitrogen, phosphorus, and zinc, as well as toxicities of aluminum and manganese. Similarly, drought is among the most widely distributed and endemic abiotic problems affecting bean production in many regions of the world, especially in northeastern Brazil, the central and northern highlands of Mexico, the Rift Valley of East Africa, and the intermountain regions of the USA. Complete crop failures under dryland conditions in these regions are not uncommon.
High temperatures (>30°C day and/or >20°C night) in tropical lowlands (below 650 m elevation) and at higher latitudes (e.g., California, Colorado, Idaho, Nebraska, Washington, and Wyoming in the USA) can severely limit bean production. Low temperatures (below 15°C), as well as frost at the beginning and end of the growing season in the highlands (above 2000 m elevation) of Latin America and at higher latitudes (e.g., >40°C in the USA and Canada) also can reduce bean yields.
Among biotic stresses, bacterial diseases such as common bacterial blight (CBB) is a widespread problem from tropical to temperate bean growing environments. In relatively cooler and wetter areas, halo blight [caused by Pseudomonas syringae pv. phaseolicola (Burkh.)] and bacterial brown spot (caused by Pseudomonas syringae pv. syringae van Hall) may cause severe yield losses.
Angular leaf spot [caused by Phaeoisariopsis griseola (Sacc.) Ferr.], anthracnose [caused by Colletotrichum lindemuthianum (Sacc. & Magnus) Lams.-Scrib.], and rust are considered among the most widely distributed foliar fungal diseases that cause severe yield losses of common bean in the Americas, Africa, and other parts of the world. Root rots caused by Fusarium solani f. sp. phaseoli (Burkh.) Snyd. & Hansen and other soil borne pathogens in most bean growing environments, web blight [caused by Thanatephorus cucumeris (Frank) Donk.] in the warm humid tropics, and white mold and ascochyta blight [caused by Phoma exigua var. diversispora (Bub.) Boerma] in cool wet regions, occasionally become severe on common bean.
Bean common mosaic virus (BCMV, a potyvirus) in most bean production regions of the world, and Bean golden mosaic virus (BGMV, a geminivirus) which occurs in tropical and subtropical Central America, coastal Mexico, the Caribbean, Brazil, and Argentina cause severe yield losses in common bean. Curly top (caused by Beet curly top virus, a geminivirus) in the western and Pacific Northwestern (PNW) USA, and Bean yellow mosaic virus (a potyvirus) in the PNW, Europe, Middle East, North Africa, and Asia also can cause severe yield losses in susceptible cultivars.
Among the insect pests, leafhoppers Empoasca kraemeri Ross & Moore (in the tropics and subtropics) and E. fabae (Harris) (in the temperate and cooler environments) are the most widely distributed problems, especially in relatively drier areas. Bean pod weevil (Apion godmani Wagner and A. aurichalceum Wagner) causes severe damage in the highlands of Mexico, and in Central America. In the highlands of Mexico and in the USA, Mexican bean beetle (Epilachna varivestis Mulsant) also causes severe leaf damage, especially in late maturing cultivars. Bean fly (Ophiomyia phaseoli Tryon) is by far the most damaging insect pest of common bean in Africa (Abate and Ampofo, 1996; Wortmann et al., 1998). The bean weevil Zabrotes subfasciatus Boheman (in warm tropical and subtropical environments) and Acanthoscelides obtectus (Say) (in cool and temperate environments) cause severe losses when dry beans are not properly stored.
Traits Deficient in Common Bean Cultivars
Moderate to high levels of resistance are found in common bean cultivars to angular leaf spot (Ferreira et al., 2000; Guzmán et al., 1995; Pastor-Corrales et al., 1998; Schwartz et al., 1982), anthracnose (Alzate-Marin et al., 1997a; Balardin and Kelly, 1998; Bannerot, 1965; Melotto and Kelly, 2000; Menezes and Dianese, 1988; Pastor-Corrales et al., 1995; Schwartz et al., 1982; Young and Kelly, 1996a,b), BCMV (Drijfhout, 1978; Kelly, 1997), bean pod weevil (Garza and Muruaga, 1993; Garza et al., 1996; Mancía, 1973), BGMV (Blair and Beaver, 1993; Molina Castañeda and Beaver, 1998; Morales and Niessen, 1988; Urrea et al., 1996; Velez et al., 1998), root rots (Abawi and Pastor-Corrales, 1990; Beebe et al., 1981), rust (CIAT, 1985; Stavely, 1984, 1988, 1989, 1999), and drought stress (Abebe et al., 1998; Acosta-Gallegos and Adams, 1991; Acosta-Gallegos and Kohashi-Sibata, 1989; Acosta et al., 1999; Acosta-Gallegos et al., 1997; Singh and Terán, 1995; Terán and Singh, 2002). The levels of resistance to CBB, halo blight, bacterial brown spot, ascochyta blight, web blight, white mold, bean fly, leafhoppers, and bruchids, among other factors, are not adequate. Later in this paper, I will discuss alien germplasm from primary, secondary, and tertiary gene pools of common bean as promising sources of resistance to abiotic and biotic stresses.
Broadening the Genetic Base of Common Bean Cultivars
Approximately 29000 cultivated and >1300 wild accessions of common bean, >1000 of the secondary, and >350 of tertiary gene pools are available at the International Center for Tropical Agriculture (CIAT), Cali, Colombia (Cardona and Kornegay, 1999; CIAT, 1996; Debouck, 1999). Similarly, at the USDA-ARS Western Regional Plant Introduction Station, Pullman, WA, there are 11 809 accessions of primary, 475 of secondary, and 121 of tertiary gene pools of common bean (Silbernagel and Hannan, 1992; M. Welsh, 2001, personal communication). However, much of the variability in these sources has yet to be utilized for common bean improvement (Miklas, 2000; Silbernagel and Hannan, 1992; Singh, 1992).
In common bean, large differences are found in seed characteristics, growth habit, and adaptation traits of cultivars of different market classes. There are large differences in combining ability (Nienhuis and Singh, 1986, 1988a; Singh et al., 1992b) and occasional problems occur affecting recombination and gene exchange between common bean races and gene pools (Koinange and Gepts, 1992; Kornegay et al., 1992; Mumba and Galwey, 1999; Singh and Gutiérrez, 1984; Singh and Molina, 1996; Welsh et al., 1995). These and other considerations have made the task of maximizing the use of available germplasm a challenging one. Nonetheless, introgression and pyramiding of useful alleles from within and across cultivated races and gene pools, wild populations of common bean, and its secondary and tertiary gene pools would broaden the genetic base of cultivars, maximize gains from selection, and increase the durability of resistance to diseases such as anthracnose, BCMV, halo blight, and rust.
For the use of Phaseolus germplasm for common bean improvement in the USA, readers should refer to Miklas (2000) and Silbernagel and Hannan (1988)( 1992). In addition, Debouck (1999), Hucl and Scoles (1985), Smartt (1970), and Waines et al. (1988) reviewed interspecific hybridization of P. vulgaris with its related species. It should be apparent from these reviews that only a small portion of genetic variability available in Phaseolus species has bean utilized thus far for common bean improvement since organized breeding was initiated in the late nineteenth and early twentieth century.
Common bean originated in tropical and subtropical regions (Gepts et al., 1986), it is a short-day species (White and Laing, 1989), and it was introduced in the USA (Kaplan and Lynch, 1999). Because of this, most introduced tropical and subtropical common bean and other Phaseolus species germplasm are unadapted in the USA, Canada, and other higher latitude environments. Thus, genes for early maturity and photoperiod insensitivity may need to be incorporated (i.e., through a conversion program) into unadapted germplasm, a priory for their adequate evaluation and use in improvement programs. To facilitate crosses with insensitive parents, photoperiod sensitive genotypes could be grown under shorter day lengths. Often, gene introgression from alien germplasm is accomplished first, followed by pyramiding favorable alleles from diverse sources for specific traits into a common genotype (i.e., character improvement, germplasm enhancement, parental development, or prebreeding). A separate, sustained, and integrated genetic improvement program is warranted for each major market class for subsequent utilization of the resulting germplasm to improve upon adaptation, yield, resistance to abiotic and biotic stresses, and processing and culinary qualities of snap and dry bean cultivars (Singh, 1992, 1999a). A three-tiered breeding approach (Kelly et al., 1998b, 1999e; White and Singh, 1991) would be the most appropriate for a long-term program encompassing germplasm enhancement and cultivar development, especially in temperate climates such as in the USA and Canada. This involves (i) gene introgression from alien germplasm, (ii) gene pyramiding from different sources for specific traits, and (iii) simultaneous selection for multiple agronomic traits for cultivar development.
The mass-pedigree (Beebe et al., 1993; Grafton et al., 1993; Singh et al., 1989, 1993), pedigree (Kelly et al., 1994a, 1994b), and recurrent backcross (Alberini et al., 1983; Bliss, 1993; Miranda et al., 1979; Pompeu, 1980, 1982) methods and their modifications often were used for common bean improvement. Researchers also have used congruity backcrossing (Haghighi and Ascher, 1988; Mejía-Jiménez et al., 1994; Urrea and Singh, 1995), single seed descent (SSD) (Kelly et al., 1989; Urrea and Singh, 1994), recurrent (Beaver and Kelly, 1994; Duarte, 1966; Kelly and Adams, 1987; Singh et al., 1999; Sullivan and Bliss, 1983), and gamete selection (Singh et al., 1998a) methods. Thus, most, if not all, commonly used crop breeding methods have been employed with common bean. However, objective data comparing the efficiency of different selection methods, with few exceptions (Beaver and Kelly, 1994; Gutiérrez and Singh, 1992; Singh and Terán, 1998; Urrea and Singh, 1994, 1995) are not available. Urrea and Singh (1994) found that the F2-derived family method of selection was superior to the SSD and bulk methods commonly used for advancing early generation of hybrid populations. Singh and Urrea (1995) and Singh et al. (1990) suggested selection for seed yield in early generation of interracial and intergene pool populations to identify promising populations with desirable recombinants. From early generation yield tests (F2–F4), Singh and Terán (1998) identified high- and low-yielding populations that eventually produced high- and low-yielding advanced generation (F7) lines.
Gene Introgression from Alien Germplasm
Introgressing useful genes across market classes within a cultivated common bean race, from distantly related races and gene pools, from wild populations, and/or from alien species forming the secondary and tertiary gene pools, is accomplished separately. This is because of differences in genetic distance between different Phaseolus species and P. vulgaris (Debouck, 1999, 2000; Debouck and Smartt, 1995) and between gene pools and races within the common bean cultivars (Singh, 1989; Singh et al., 1991a), and because of the different breeding methods and strategies that are required. Knowledge of parental performance for seed yield and other quantitative and qualitative characters, combining ability, occurrence of incompatibility genes, and undesirable linkages are essential for introgressing desirable alleles from alien germplasm. The problem of gene recombination and recovery of useful genotypes often increases with increase in genetic distance between parents to be hybridized (Kornegay et al., 1992; Mumba and Galwey, 1999; Singh and Molina, 1996; Welsh et al., 1995; see review by Debouck, 1999 for interspecific hybrids).
Introgression between Market Classes within a Common Bean Race
In the small-seeded Mesoamerica race, an indeterminate upright Type II growth habit was introgressed from Central American cultivars to determinate Type I growth habit navy bean to improve plant type and seed yield (Adams, 1982). The taller Type II cultivars also are required for large scale mechanized commercial farming to reduce disease incidence, dependence on hand labor, and production costs (Vandenberg and Nleya, 1999). The I gene for BCMV resistance, gene(s) for anthracnose resistance, and/or Type II growth habit were transferred from cultivars of different market classes within Mesoamerica race in the Central American small reds (Beebe and Pastor-Corrales, 1991; Beebe et al., 1993) and Brazilian carioca (cream striped), jalinho (beige), and mulatinho (cream) beans that were characterized by prostrate growth habit Type III (Alberini et al., 1983; Miranda et al., 1979; Pompeu, 1980, 1982; Singh et al., 1998a; Thung et al., 1993).
In Durango race, the bc-2 and/or bc-22 gene for BCMV resistance originally from common red Mexican landrace (Miklas, 2000) was transferred to pinto beans (Silbernagel, 1994; Stavely et al., 1995). Similarly, curly top resistance from common red Mexican landrace was introgressed into great northern and pinto beans (Schultz and Dean, 1947).
Introgression between Cultivated Races within a Primary Gene Pool
Despite considerable differences in seed and plant characteristics, introgression of useful genes among and across the three Middle American races, namely, Durango, Jalisco, and Mesoamerica has been common for tropical and subtropical Latin American environments. Differences in photoperiod response and phenological traits are not large at intermediate elevations (1000–1800 m). Often it is sufficient to use a three-way, modified-double cross (Singh, 1982), or backcross, ensuring >65% genetic contribution of the parents of the same race and market class under improvement.
Most landraces and improved cultivars belonging to Durango race possess the prostrate semiclimbing Type III growth habit (Singh, 1989; Singh et al., 1991a). In semiarid production regions of the central highlands of Mexico and the northwestern USA, this prostrate Type III growth habit often may not impose serious problems for disease management and their high yield potentials are often realized. But in comparatively wet or humid growing environments (e.g., Michigan, Minnesota, and North Dakota), such cultivars are prone to white mold and seed discoloration, because of their closed canopy and pods touching the ground. Thus, upright erect cultivars that permit better air movement through the canopy and keep pods away from the soil are preferred (Coyne et al., 1974). Type II growth habit was introgressed from the Mesoamerica race (small black and navy beans) to the Durango race cultivars by phenotypic recurrent selection (Kelly and Adams, 1987). Subsequently, this trait and the I gene for BCMV and Ur-3 gene for rust resistance from small black or navy beans were transferred into other Durango race cultivars including great northern, pink, pinto, and red market classes (Coyne et al., 2000; Grafton et al., 1999; Hang et al., 1997, 1998; Hosfield et al., 1995; Kelly et al., 1990, 1992a,b; Park et al., 1999; Silbernagel and Hang, 1997a,b,c,d; Stavely and Grafton, 1989; Stavely et al., 1995, 1989, 1998, 1999a,b). Kelly (1992) transferred bc-3 gene for BCMV resistance from small white (IVT 7214) to great northern and pinto beans. Coyne et al. (1991)(1994, 2000) introgressed moderate levels of tolerance or avoidance for white mold from small black and navy to great northern and pinto beans. Tolerance to Fusarium root rot from Mesoamerica race beans was introgressed into race Durango pink, pinto, and red market classes (Burke, 1982a,b,c; Burke et al., 1995a,b,c; Hang et al., 1997, 1998; Kelly et al., 1990, 1999a,d; Schneiter et al., 1982, 1983; Silbernagel and Hang, 1997b, 1998).
Extremely high levels of resistance to Mexican pod weevil (A. godmani) were found in the Jalisco race climbing bean germplasm from the highlands of Mexico (Beebe et al., 1993; Garza and Muruaga, 1993; Garza et al., 1996; Mancía, 1973). The insect also is a severe problem in the warmer regions of Guatemala, El Salvador, Honduras, and northern Nicaragua, where small-seeded popular cultivars are highly susceptible to this insect (Mancía, 1973). The Agr gene confers intermediate levels of resistance, but the resistance to the pod weevil is higher when combined with the Agm gene (Garza et al., 1996). However, the Agm gene alone has no effect. High levels of resistance from Jalisco race have been transferred into Middle American small red- and black-seeded cultivars (Beebe et al., 1993).
Resistance to BGMV can be expressed as percent infection, plant dwarfing, leaf chlorosis, and pod deformation (Morales and Nissen, 1988). Resistance to each of these component traits is under different genetic control (Blair and Beaver, 1993; Molina Castañeda and Beaver, 1998; Morales and Singh, 1991; Urrea et al., 1996; Velez et al., 1998). Moreover, at least two different sources of resistance to leaf chlorosis are known to exist. One of the sources found in race Durango (e.g., landrace ‘Garrapato’) is controlled by the recessive gene Bgm-1. Bgm-1 was fortuitously transferred into common bean line A 429 (Morales and Singh, 1993). Subsequently, it was used to develop highly resistant small-seeded cultivars of red (Don Silvio, Tio Canela 75), black (Turbo III), and carioca (Carioca 2000) market classes for Central America, Brazil, and Bolivia by pedigree, mass-pedigree, and gamete selection methods. Moreover, availability of RAPD (random amplified polymorphic DNA) and SCAR markers (Urrea et al., 1996) minimize the need for direct screening in presence of the virus.
Gene Introgression between Two Cultivated Primary Gene Pools
Recovery of essential agronomic characteristics such as adaptation, yield, and seed and pod quality characteristics of cultivars is challenging while introgressing desirable alleles by means of biparental crosses between Andean and Middle American gene pools of common bean. Problems exist irrespective of breeding systems such as pedigree (Kornegay et al., 1992), SSD (Welsh et al., 1995), or mass selection (Singh et al., 1989). Similarly, it is difficult to develop a dry or snap bean cultivar from the single crosses between the two commercial classes (Myers, 2000). Recurrent or congruity inbred-backcrossings (Bliss, 1993; Urrea and Singh, 1995) and recurrent selection (Beaver and Kelly, 1994; Singh et al., 1999) are required. Moreover, bridging-parents may be needed (Singh and Gutiérrez, 1984), if the Dl-1 allele in the Middle American interacts with the Dl-2 allele in the Andean bean.
Resistance to leafhopper was introgressed from the Mesoamerica race to Nueva Granada (CIAT, 1992). The Bgm-1 allele for resistance to BGMV from the Durango race was transferred into large-seeded red mottled Andean bean (Beaver and Steadman, 1999) and into snap bean (McMillan et al., 1998). Beaver et al. (1999) also transferred Bgm-2 allele for BGMV resistance from the Middle American bean to the light red kidney Andean bean. Stavely and McMillan (1992) transferred rust resistance from the Middle American dry bean to snap beans. In contrast, rust resistance controlled by the Ur-6 allele was transferred from Andean kidney bean to race Durango pinto bean (Wood and Keenan, 1982). Similarly, Ur-4 from ‘Early Galatin’ snap bean was transferred into great northern and pinto bean (Stavely and Grafton, 1989; Stavely et al., 1989, 1998, 1999a,b).
Gene Introgression from Wild Populations of Common Bean
Resistance to bruchid (Z. subfasciatus) was absent in thousands of accessions of cultivated common bean (Schoonhoven and Cardona, 1982). Only a few wild bean populations from the highlands of Mexico were highly resistant (Acosta-Gallegos et al., 1998; Cardona and Kornegay, 1989, 1999; CIAT, 1996; Schoonhoven et al., 1983).
By extracting purified arcelin seed protein from wild beans and using a range of concentrations in artificial seeds, Osborn et al. (1988) demonstrated that bruchid resistance was due to the presence of a unique protein in the seed cotyledons of wild bean. A single dominant gene controls the presence of arcelin (Osborn et al., 1986), and six alleles with varying effects have been characterized (Acosta-Gallegos et al., 1998; Cardona and Kornegay, 1989, 1999; Cardona et al., 1989, 1990; Kornegay et al., 1993). The presence of arcelin is easily detected by SDS-PAGE electrophoresis. Since antiserum to the protein is available, it is easier, faster, and cheaper to use the ELISA technique for large scale screening of segregating populations, families, and derived advance-generation lines. This development has facilitated and expedited the breeding for bruchid resistance, and minimized dependence on screening using insect infestation, which is time-consuming and dependent on the environment. A recurrent backcrossing was used to transfer extremely high level of resistance to bruchids from the Mexican wild bean populations to a range of cultivars (Cardona and Kornegay, 1989, 1999). Arcelin is a heat-labile seed storage protein. Its negative effects on humans and animals are lost after cooking beans.
Introgression from the Secondary Gene Pool
Resistance to angular leaf spot (Busogoro et al., 1999), anthracnose (Hubbeling, 1957), ascochyta blight (Schmit and Baudoin, 1992), BGMV (Beebe and Pastor-Corrales, 1991; CIAT, 1986; Singh et al., 1997), Bean yellow mosaic virus (Baggett, 1956), CBB (Mohan, 1982; Schuster et al., 1983; Singh and Muñoz, 1999), root rots (Azzam, 1957; Hassan et al., 1971; Yerkes and Freytag, 1956), white mold (Abawi et al., 1978; Hunter et al., 1982), and cold (Bannerot, 1979) are found in the secondary gene pool of common bean comprising P. coccineus, P. costaricensis, and P. polyanthus. Wilkinson (1983) suggested that P. coccineus could be a potential source of high yield for common bean.
Despite some difficulties in developing true breeding lines, researchers have successfully introgressed moderate levels of resistance from P. coccineus to common bean for CBB (Freytag et al., 1982; Miklas et al., 1994a,b; Park and Dhanvantary, 1987), for Fusarium root rot (Wallace and Wilkinson, 1965), and for white mold in dry bean (Miklas et al., 1998a) as well as in snap bean (Abawi et al., 1978; Dickson et al., 1982; Lyons et al., 1987). In contrast, resistance to halo blight from the common bean was incorporated into P. coccineus (Ockendon et al., 1982).
Introgression from the Tertiary Gene Pool
Resistance to ashy stem blight [caused by Macrophomina phaseolina (Tassi) Goid.] and Fusarium wilt (caused by Fusarium oxysporum f. sp. phaseoli Kendrick & Snyder) Miklas et al. (1998b), BGMV (Miklas and Santiago, 1996), bruchids (CIAT, 1995, 1996; Dobie et al., 1990; Shade et al., 1987), CBB (Coyne et al., 1963; Schuster et al., 1983; Singh and Muñoz, 1999), drought (Federici et al., 1990; Markhart, 1985; Parsons and Howe, 1984; Rosas et al., 1991; Thomas et al., 1983), leafhopper (CIAT, 1995, 1996), and rust (Miklas and Stavely, 1998) are found in accessions of P. acutifolius. A high level of resistance to CBB was transferred from tepary to common bean (Coyne et al., 1963; McElroy, 1985; Scott and Michaels, 1992; Singh and Muñoz, 1999). However, introgression of resistance to other factors has not been achieved. Poor choice of parents, use of comparatively small population sizes of interspecific hybrid progenies, and/or inappropriate screening techniques for segregating populations might have restricted effective gene introgression between species. To overcome some of the difficulties associated with introgressing quantitatively inherited traits from P. acutifolius to P. vulgaris, the following procedures may be helpful: (i) use of congruity backcrossing, (ii) production of large interspecies hybrid progenies from plant-to-plant paired pollinations at each step of the crossing, and (iii) development of a large number of inbred-backcross lines before subjecting them to appropriate screenings (Singh and Muñoz, 1999). Moreover, molecular marker-assisted introgression (Tanksley et al., 1996) may expedite and permit more focused use of tepary bean and other alien cultivated and wild germplasm.
Gene Pyramiding for Specific Traits
Pyramiding favorable alleles for specific traits has been used in the following situations: (i) for leafhopper resistance from within a given cultivated race of common bean, (ii) for seed yield and BCMV and drought resistance from among races within a primary gene pool, (iii) for anthracnose, rust, and BGMV resistance from across the two cultivated primary gene pools within P. vulgaris, and (iv) for common bacterial blight from across Phaseolus species.
Leafhopper Resistance
Tolerance to feeding damage by leafhoppers E. kraemeri (Calderón and Backus, 1992; Kornegay et al., 1986; Wolfenbarger and Sleesman, 1961) and differences for ovipositional nonpreference were found in common bean cultivars (Kornegay et al., 1986, 1989; Wilde and Schoonhoven, 1976). Both traits are inherited quantitatively (Galwey and Evans, 1982; Kornegay and Temple, 1986). Through recurring cycles of the bulk-pedigree method of selection, significantly higher levels of tolerance were accumulated in breeding lines (Cardona and Cortés, 1991; Cardona and Kornegay, 1999; CIAT, 1993; Kornegay and Cardona, 1990; Kornegay et al., 1989). These E. kraemeri tolerant lines also were tolerant to E. fabae (Harris) in Ontario, Canada (Schaafsma et al., 1998).
Bean common mosaic virus
Both strain-specific and nonspecific resistances to BCMV have been reported (Drijfhout, 1978; Drijfhout et al., 1978; Kelly, 1997). Molecular markers for the dominant I (Haley et al., 1994b; Melotto et al., 1996), and for recessive bc-12 (Miklas et al., 2000a) and bc-3 (Haley et al., 1994a; Johnson et al., 1997) genes also have been reported. Undesirable linkages of I with B gene for seed coat color occur (Cáceres et al., 1985; Kyle and Dickson, 1988; Temple and Morales, 1986) with some exceptions (Kelly, 1992). The I and bc-12 genes have been combined in pinto ‘Kodiak’ (Kelly et al., 1999a), and I and bc-3 were combined in recent rust resistant great northern BelMiNeb RMR-6 and 7, pinto BelDakMi RMR-14 to 18 bean germplasm lines (Stavely et al., 1998, 1999a,b), and ‘Raven’ black bean (Kelly et al., 1994b).
Seed Yield
Heritability of seed yield varies from low to moderately high (Nienhuis and Singh, 1988b; Singh et al., 1991e; Welsh et al., 1995). Good general combining ability exists among the three common bean races of Middle American gene pool (Nienhuis and Singh, 1986, 1988a). Thus, high yielding genotypes have been developed by mass-pedigree (Singh, 1995b; Singh et al., 1989, 1993) and recurrent selection (Singh et al., 1999) methods from interracial populations within the Middle American gene pool. While Kornegay et al. (1992) and Singh et al. (1989) were unable to develop high yielding lines, Beaver and Kelly (1994) and Singh et al. (1999) successfully selected for high yield in Andean x Middle American populations developed through recurrent selection.
Drought Resistance
Among physiological and agronomic traits used as selection criteria in breeding for drought resistance in common bean, seed yield was the most reliable (Ramirez-Vallejo and Kelly, 1998; White et al., 1994a). Heritability for seed yield under drought stress ranged from 0.09 to 0.80 (Schneider et al., 1997; Singh, 1995a). Cultivars from Durango race possess the highest levels of drought resistance (Abebe et al., 1998; Acosta-Gallegos et al., 1997; Terán and Singh, 2002) as would be expected on the basis of their origin in the semiarid Mexican highlands (Singh, 1989; Singh et al., 1991a). White et al. (1994b) observed large differences in combining ability for drought stress resistance in common bean races of Middle American origin. Lines possessing higher levels of drought resistance were developed from Durango x Mesoamerica and/or Mesoamerica x Nueva Granada populations (Rosales-Serna et al., 2000; Schneider et al., 1997; Singh, 1995a; Singh et al., 2001b).
Anthracnose
Despite the fact that C. lindemuthianum that causes anthracnose in common bean is highly variable pathogen (Balardin et al., 1997, 1999; Melotto et al., 2000; Sicard et al., 1997), dozens of highly resistant common bean cultivars have been identified (Pastor-Corrales et al., 1995; Schwartz et al., 1982). Some landrace cultivars (e.g., G 2333) possess multiple genes for resistance (Pastor-Corrales et al., 1994; Young et al., 1998). Furthermore, Balardin and Kelly (1998) proposed pyramiding different resistance genes of Andean and Middle American origins as the long-term approach to control bean anthracnose. Kelly et al. (1998a)( 1999b, 2000) combined the Andean Co-1 and Middle American Co-2 genes for anthracnose resistance in small-seeded black and large-seeded dark and light red kidney beans. Availability of molecular markers for five independent dominant resistance genes (Adam-Blandon et al., 1994; Alzate-Marin et al., 1997b; Young and Kelly, 1996b, 1997; Young et al., 1998) should facilitate their pyramiding and deployment for specific common bean production regions to control anthracnose.
Bean golden mosaic virus
Two independent recessive genes, bgm-1 (Blair and Beaver, 1993; Urrea et al., 1996) and bgm-2 (Velez et al., 1998), controlling resistance to leaf chlorosis caused by BGMV exist. Also, a recessive gene dwf (Velez et al., 1998) for plant dwarfing and a dominant gene Bgp for resistance to pod deformation (Molina Castañeda and Beaver, 1998), both symptoms also caused by BGMV, were identified. The latter requires bgm-1 to be expressed. Singh et al. (2000a)(b) combined the Bgm-1 gene imparting resistance to leaf chlorosis from Garrapato (Durango race) with resistance to plant dwarfing from Royal Red (Nueva Granada race).
Bean Rust
U. appendiculatus, the causal organism of bean rust, is among the most variable pathogens that cause disease in common bean (CIAT, 1985; Stavely, 1984, 1988, 1989, 1999). Being wind-borne, it is highly mobile and thus possesses great potential for changing its population composition within and between growing seasons and production regions. High levels of resistance have been identified in a range of cultivated bean types of both Andean and Middle American origins (CIAT, 1985; Stavely, 1984). While the genetics of resistance found in some important sources of germplasm still need to be fully understood, at least 11 known genes carrying race-specific resistance have been identified (Carvalho et al., 1978; Christ and Groth, 1982; Grafton et al., 1985; Kolmer and Groth, 1984; Stavely, 1990) and classified according to their origins in common bean gene pools (Kelly et al., 1996). RAPD markers linked to resistance genes are available (Boone et al., 1999; Haley et al., 1993; Johnson et al., 1995; Jung et al., 1996; Miklas et al., 1993). These markers are being used to incorporate and pyramid rust resistance into common bean cultivars, and/or to combine rust resistance with resistance to other diseases, such as BCMV, BGMV, anthracnose, and/or CBB (see reviews by Kelly and Miklas, 1998, 1999). For example, Ur-3, Ur-4, and Ur-11 have been combined with I and bc-1 gene for BCMV resistance into great northern BelMiNeb RMR-7 (Stavely et al., 1999b), and pinto BelDakMi RMR 14 to 17 have Ur-3, Ur-6, and Ur-11 rust resistance genes (Stavely et al., 1998, 1999a). In addition, the recent pinto bean germplasm release, BelDakMi-RMR-18, has four rust resistance genes (Ur-3, Ur-4, Ur-6, and Ur-11) of which Ur-3 and Ur-11 are of Middle American and Ur-4 and Ur-6 are of Andean origin. Thus, this is a singular example in the history of common bean disease resistance breeding, whereby four genes of the two distinct evolutionary origins have been combined to impart resistance to all known races of the rust pathogen in the USA (>70 races). Over 20 yr of informal but close collaboration between federal and state institutions have been crucial for this remarkable achievement.
Common Bacterial Blight
Only low to intermediate levels of CBB resistance are found in common and scarlet runner bean (P. coccineus), but tepary bean (P. acutifolius) possesses the highest levels of resistance (Coyne et al., 1963; Schuster et al., 1983; Singh and Muñoz, 1999; Zapata et al., 1985). Major dominant and/or recessive genes control the race-specific resistance to CBB in tepary bean (Drijfhout and Blok, 1987; Dursun et al., 1995; Freytag, 1989; Urrea et al., 1999). However, resistance found in common bean, including that introgressed from the scarlet runner (Freytag et al., 1982; Miklas et al., 1994a; Park and Dhanvantari, 1987) and tepary beans involve one or more genes with major effects and five to eight genes or QTLs with small effects (Jung et al., 1996; McElroy, 1985; Nodari et al., 1993; Silva et al., 1989).
CBB resistance found in common bean (Beebe and Pastor-Corrales, 1991), scarlet runner (Freytag et al., 1982; Miklas et al., 1994a; Park and Dhanvantari, 1987), and tepary (Coyne et al., 1963; McElroy, 1985; Scott and Michaels, 1992) beans have been introgressed independently. However, the highest level of CBB resistance from different species was recently introgressed and pyramided (Miklas et al., 2000b; Singh and Muñoz, 1999; Singh et al., 2001a). Since different levels of CBB resistance are found in common, scarlet runner, and tepary beans (Singh and Muñoz, 1999), efforts first be made to systematically pyramid different CBB resistance genes within each species. Next, CBB resistance be combined between common and scarlet runner beans, since they possess comparatively lower levels of CBB resistance than the tepary beans (Singh and Muñoz, 1999). Eventually, the pyramided CBB resistance thus obtained could be combined with that of the tepary bean. Use of tightly linked molecular markers for each of the CBB resistance genes found across these three Phaseolus species, access to the widest range of pathogenic populations of the bacterium, and molecular marker-assisted introgression and recurrent selection would expedite this breeding process.
Simultaneous Selection for Multiple Agronomic Traits for Cultivar Development
High yielding cultivars with the maximum expression of valuable trait and a combination of the maximum number of desirable traits are sought in each successive breeding cycle. To achieve this objective, Grafton and Singh (2000), Kelly et al. (1998b)(1999e), and Singh (1992)(1999a) discussed general strategies for common bean improvement.
For each market class, all commercial cultivars, elite lines, and donors of favorable alleles (including those obtained from gene introgression from alien germplasm and gene pyramiding) must be similar in growth habit, maturity, seed color and size, and be well adapted. Thus, each cross is made among only high yielding, well-adapted, elite recipient and donor parents. When the necessary genes for each of the major traits of interest are found in separate parents, bi-parental and backcrosses are not adequate. A few multiple-parent crosses should be preferred over a large number of single crosses and backcrosses. Although comparatively more time is spent during hybridization to generate multiple-parent crosses, the process allows production of recombinants with favorable alleles for multiple traits. This production of recombinants is not possible through single crosses and backcrosses without repeated cycles of selection for specific traits, one trait at a time. Thus, all desirable alleles are combined in the first step, and the time required for cultivar development is reduced. The single crosses between the two most promising or the highest yielding lines or cultivars of the diverse complementary origins within a market class should provide still higher yielding cultivars.
Gamete selection in the F1, when combined with early generation (F2–F4) selection, should help identify promising populations and families within populations (Singh, 1994). These are then used to develop superior lines for subsequent cultivar identification. Moreover, for biotic and abiotic stresses that cannot be screened simultaneously, different locations and nurseries may be required to select promising populations and families within populations (Singh et al., 1991d, 1992a, 1998a). The F1-derived families in F2 and subsequent generations were evaluated in different locations for development of high yielding Type II growth habit lines and cultivars with resistance to five diseases and leafhopper (Singh et al., 1998a, 2000c). The use of different locations and complementary nurseries was crucial to select for resistance to multiple pests in a relatively short time.
Mesoamerica Race Cultivars
The major achievements of breeding in this group of cultivars in Canada and the USA include earliness, adaptation to higher latitude, high yield, upright plant type, combination of I and bc-3 and I and bc-12 genes for resistance to BCMV, and rust and anthracnose resistance (Grafton et al., 1993; Kelly et al., 1989, 1994a,b, 1995, 2000; Myers et al., 1991). Resistance to BCMV, BGMV, CBB, leafhopper, bruchid, and bean pod weevil, and upright plant type in red and black beans in Central America were improved (Beebe and Pastor-Corrales, 1991; Beebe et al., 1993; Kornegay and Cardona, 1990, 1991). Most of these traits, with the exception of bean pod weevil resistance (which was not required), and resistance to angular leaf spot, anthracnose, and drought stress were bred into cream, cream striped, and beige types for Brazil (Alberini et al., 1983; Miranda et al., 1979; Pompeu, 1980, 1982; Singh et al., 1992c, 1998a; Thung et al., 1993). Additional information regarding genetic improvement of Mesoamerica race cultivars can be found in a recent article by Singh (1999b).
Development of high yielding, early maturing (
90-d duration) Type II growth habit cultivars suitable for direct combine harvesting has proven to be very difficult. In addition, these cultivars should have adequate resistance to abiotic (e.g., drought and soil mineral deficiency and toxicity) and biotic (e.g., white mold and CBB) stresses for low-input sustainable farming systems.
Durango Race Cultivars
Resistance to drought, soil mineral deficiency (zinc), and soil compaction seems to have been inadvertently reduced in modern cultivars compared with landraces grown in the first half of twentieth century in the USA (S. Singh, 2000, unpublished data). Improving resistance to these abiotic stresses simultaneously with resistance to biotic stresses, especially white mold, in otherwise high yielding cultivars for North America is the most formidable challenge.
Nueva Granada Race Cultivars
Improving seed yield of Type I growth habit cultivars, and combining high yield, high seed quality, and early maturity in Type II and Type III growth habit cultivars for North America is challenging. In addition, resistance to drought and low fertility soils along with resistance to major biotic constraints will be essential for sustained common bean production in the tropics and subtropics.
Snap Bean Cultivars
Maximizing pod yield and quality of Type I growth habit snap bean cultivars destinned for a single destructive harvest offers a daunting challenge to breeders. Moreover, cultivars adapted to low-input farming systems, possessing resistance to major abiotic and biotic stresses will be required.
Conclusions and Future Prospects
Collection, characterization, and understanding of genetic diversity in Phaseolus beans in the later half of the twentieth century has been remarkable. Similarly, principal production constraints and traits deficient in common bean cultivars were determined. The major achievements of several decades of organized breeding include earliness, adaptation to higher latitude, high yield, improved pod and seed quality, upright plant type, and resistance to diseases (angular leaf spot, anthracnose, BCMV, BGMV, CBB, curly top, and rust), insect pests (bean pod weevils, bruchids, and leafhoppers), and drought. Nonetheless, the genetic base of commercial cultivars within specific market classes is narrow, the average global yield of common bean remains low (<900 kg ha-1), and its production suffers from a wide range of abiotic and biotic constraints, some causing total yield losses. Resistance to drought, soil mineral deficiency and toxicity, and soil compaction seem to have been inadvertently reduced in modern cultivars. Improving resistance to these abiotic stresses simultaneously with resistance to biotic stresses, especially white mold, in otherwise high yielding cultivars for North America is the most formidable challenge.
Useful alleles and traits for many agronomic traits deficient in common bean cultivars, including resistance to storage insects, leafhoppers, ascochyta blight, CBB, white mold, drought, and soil fertility problems have been identified in cultivated and wild common bean and related species in secondary and tertiary gene pools. As much as 90% of the genetic variability available in the primary gene pool and related species (mostly of tropical and subtropical origins) remains underutilized or not utilized at all. Problems of adaptation associated with introduced germplasm and the lack of long-term sustained public funding might have hindered extensive use of this Phaseolus bean germplasm.
To maximize and sustain common bean production, it is essential to develop high yielding, high quality cultivars that are adapted to low-input sustainable farming systems. This need warrants sustained, comprehensive, and integrated genetic improvement programs in which favorable alleles from cultivated and wild populations of common bean's primary, secondary, and tertiary gene pools are accumulated in superior cultivars of each of the major market classes. Long-term publicly funded germplasm conversion programs with collaboration between federal and state institutions are crucial. In the meantime, breeders in North America may find improved common bean germplasm and cultivars from national, regional, and international programs in the tropics and subtropics very useful to broaden the genetic base of cultivars in different market classes. Recombination and selection methods will vary depending upon the genetic distance between parents, breeding objectives, and available resources. Availability of an efficient and repeatable transformation system for P. vulgaris, and marker-assisted introgression and selection of useful alleles from wild relatives and alien species should expedite and facilitate the integrated genetic improvement of common bean.
NOTES
Published as Idaho Agric. Exp. Stn. Journal Article No. 01711. Univ. of Idaho, College of Agriculture, Moscow, ID 83844.
Received for publication January 20, 2001.
REFERENCES
Acosta-Gallegos et al. (1995a)(b, 2001) and Singh et al. (1993) combined resistance to anthracnose, BCMV, and rust into high yielding ‘azufrado’, ‘bayo’, black ‘Flor de Mayo’, and pinto bean cultivars for Mexican highlands. Brick et al. (1991)(1995), Burke (1982a)(b,c), Burke et al. (1995a)( b,c), Coyne et al. (1974)(1991, 1994, 2000), Grafton et al. (1993)(
Major advances include incorporation of earliness, adaptation to higher latitude, and resistance to BCMV, curly top, and rust in determinate Type I growth habit cultivars of different market classes. Good canning quality and resistance to anthracnose, BCMV, curly top, and rust, either singly or in combinations, have been bred into light red kidney (Kelly et al., 1999b), dark red kidney (Kelly et al., 1998a), and white kidney (Kelly et al., 1999c) bean cultivars for North America. Beaver (1999), Beaver and Molina (1997), and Beaver and Steadman (1999) reviewed breeding for cultivars of the large-seeded Nueva Granada race for the tropics and sub-tropics. Resistance to angular leaf spot, anthracnose, BCMV, BGMV, CBB, and/or rust has been bred into Nueva Granada race cultivars for Africa, the Caribbean countries, and Latin America.
Skroch and Nienhuis (1995) reported that snap bean cultivars have a broad genetic base compared with dry beans. For example, ‘Oregon 91G’ has "S" Mesoamerican phaseolin, while its morphological and horticultural characters are of the Andean beans (Myers, 2000). For decades, breeders in the USA and Europe have crossed extensively to Mesoamerican germplasm to introgress disease resistance, small seed and pod size, and other traits. Improvements in snap bean include a change from climbing to bush growth habit, increased lodging resistance, concentration of pod set, stringless or low pod fiber, round pod cross-section, straight and smooth pods, darker green interior and exterior color, reduced interlocular cavitation, slow seed development, and incorporation of pod pubescence (Myers, 2000; Myers and Baggett, 1999; Silbernagel, 1986; Silbernagel et al., 1991). Similarly, resistances to bacterial blights, BCMV, BGMV, Cucumber mosaic virus, curly top, root rots, rust, and white mold, as well as heat, cold, and ozone have been incorporated singly or in various combinations (McMillan et al., 1998; Silbernagel, 1979, 1987; Stavely and Steinke, 1985).
