Breeding Dry Beans for the Northern Plains
April 15, 2002
By Ken Grafton, Plant Sciences Department, North Dakota State University
The objective of the dry bean breeding program at North Dakota State University is to develop high yielding, high quality bean genotypes adapted to the northern Great Plains. In 2000-2001, we made more than 600 unique hybridizations. To date, more than 350 unique hybridizations were made for 2001-2002 and we are now completing our navy and black bean crossing series. Parental germplasm consists of adapted cultivars grown in the Northern Plains, breeding lines developed at NDSU, and germplasm possessing desirable traits from other breeding programs. Unadapted germplasm lines from other sources are evaluated for desirable traits and introgressed into adapted material (e.g., pre-breeding). Each year, the breeding program evaluates material from around the world as possible sources of resistance to white mold, rust, root rot, virus, and bacterial blights.
A modified pedigree breeding method is used, which allows continual evaluation and selection of desirable plants and/or lines. Important events using this breeding scheme are as follows: Hybridizations are made in the greenhouse and the F1 plants are grown to ensure quality F2 seed. These events occur in the fall and winter so that we can plant F2 populations in the field and save one year in the breeding cycle. The F2 seed harvested from these F1 plants are planted in a spaced plant arrangement in North Dakota and Minnesota (in the case of kidney bean types) so that single F2 plants can be selected for maturity, plant architecture, number and location of pods, lack of diseases, and appropriate seed characteristics. In order to reduce the cost of the winter nursery (these costs continually increase with air freight costs now passed on to breeding projects and increased air fare to Puerto Rico), most F3 progeny from selected individual F2 plants will be grown in the field in ND and MN. Only highly superior pinto and navy lines will be planted in the winter nursery, along with advanced lines. Lines were tested in the field for rust and in the greenhouse for bean common mosaic virus (BCMV). Those found to be resistant are harvested in bulk and the resultant F3:4 lines are planted in plant rows at two locations in North Dakota. The best single plants are selected from lines that perform well at both locations and advanced one generation (F4:5 generation) in the winter nursery. The F4:6 seed is then used for a preliminary yield test (2-3 reps at each of 3 locations, depending on seed availability) for one year. If the line continues to perform well, it is single plant selected to begin pure line development and also entered into the advanced yield tests (3 reps/location; 4 locations). A selection pressure of 10-15% is placed on materials entered into plant rows and yield tests, ensuring that only the best lines will be selected. Elite lines that perform well in the advanced yield trials are then entered into variety trials for three years. The time required to develop and release a variety after the initial cross is made is usually 10-12 years, although it may be reduced considerably if off-season nurseries are used.
Off-season winter nurseries are used to reduce the number of years needed for experimental lines to reach near-homozygosity (the period during which evaluation and selection takes place on individual plants and rows). This allows the breeder to begin testing materials for yield and other agronomic traits in a shorter period of time than if limited to a breeding strategy of only one generation per year. Several years can be eliminated from cultivar development if off-season winter nurseries are used. A winter nursery was conducted near Isabela, Puerto Rico, in cooperation with Dr. J. Beaver, University of Puerto Rico. This is an excellent environment to select for stress resistance. The 2001-2002 nursery was much reduced from last years nursery, comprising only 3,200 rows.
Yield Tests, Breeding Nurseries, and Variety Trials
The breeding program had yield tests and/or breeding nurseries at nine locations in North Dakota and Minnesota [Erie, Hatton (2), Forest River, Johnstown, Fargo, Oakes, Perham, and Park Rapids] in 2001. In 2002, some modifications may occur in order to improve the efficiency of the breeding program. The white mold nursery will be moved from Hatton to Carrington to take advantage of micro-irrigation and to allow both the plant and the pathogen a better opportunity to interact. The Hatton site remained exceedingly wet during the three-year period of this nursery. Because of the extremely wet conditions, either plant growth was severely limited (usually the case) or the pathogen never developed to levels needed for uniform infection. Two sites near Hatton, one for the variety trial and the other for the breeding nursery, will remain. In 2001, pinto (P), navy (N), and miscellaneous (M) bean class variety trials were grown near Erie (P, N, & M), Hatton (P, N, & M), Forest River (P & N) and Oakes (P & N); preliminary and advanced yield tests were grown at Erie, Hatton, Johnstown, and Perham, and Park Rapids. The breeding program tested 42 pinto, 20, navy, 18 black, 8 cranberry, and 20 kidney bean lines in advanced yield trials. We also tested 134 pinto (84 harvested), 102 navy (75 harvested), 59 great northern (20 harvested), and 10 small reds (7 harvested) in preliminary yield tests. Preliminary tests of kidney and black were not conducted due to a shortage of seed. Approximately 1,500 plots were harvested for the variety trials alone; an additional 4,000 test plots of preliminary and advanced yield trials also were harvested. More than 8,000 single plant selections were made and harvested from breeding nurseries grown in 2001. Breeding nurseries were located at Erie, Fargo, Hatton, Johnstown, Perham, and Park Rapids. Germplasm evaluation nurseries (exotic material) were at Erie, Fargo, Perham, Hatton, and a rust nursery also was at Fargo. Breeding nurseries, yield trials, and other trials totaled more than 10,000 plots planted on about 40-45 acres.
The 2001 growing season was, in general, excellent. While some areas received surplus moisture, these areas were considerably reduced from rainfall extremes experienced in previous years. The weather pattern and reduced disease pressure brought about record yields in ND. Even though the region looked as if white mold pressure would be great early in the season, limited rainfall and warm temperatures in late July and early August resulted in minimal damage. Rainfall was critical at the Erie site, which developed into a very good nursery, in spite of receiving only about five inches of rain the entire growing season. This lack of rainfall hastened maturity (reduced the range between early and late-season material) and brought about reduced seed sizes. Yields also were affected, although the yield trials had very good yield averages. The breeding nursery site at Perham, however, had moderately severe root rot pressure, which allowed us to differentiate those that had some levels of root rot resistance compared to the susceptible check Montcalm. Also, the latter half of the season offered us an opportunity to evaluate lines for the ability to continue seed filling during relatively dry weather, an objective that has been difficult to select for the last few years.
The breeding program made approximately 8,000 single plant selections of pinto, navy, pink, black, great northern, kidney, and small red market classes made in early generation (F2) populations grown at Hatton and Erie, about the same number as in 2000. After harvest, these single plant selections were evaluated again for seed type in the seed lab, and the lines with the best pinto and navy seed traits were entered in the winter nursery in Puerto Rico for generation advance and additional evaluation. In Puerto Rico, individual rows are evaluated and the best rows are harvested. No more than 40% of the lines are selected for further evaluation. Lines from the winter nursery will entered into preliminary yield tests in 2002 or will be evaluated again in plant rows. Approximately 90% of the breeding effort is placed on pinto bean and navy bean improvement. Lines from single plant selection of pink, great northern, small red, black, and kidney will be evaluated in 2002 in ND and Minnesota.
Breeding lines were evaluated in the greenhouse during the winter for the presence of BCMV resistance genes I, bc22, and bc3. Lines are discarded if resistance genes are not present. Incorporating adequate levels of BCMV resistance into pinto bean breeding lines improves the overall desirability of experimental lines because of the threat of virus infection in western seed producing states. Lines also were evaluated for rust resistance, using a composite of isolates collected from North Dakota. High levels of resistance were identified in lines from a number of market classes.
In cooperation with the NDSU Plant Pathology Department, I obtained two anthracnose isolates from Michigan State University that are being used to screen for resistance to this disease. The two isolates, race 7 and race 73, represent the two most common anthracnose races in the U.S. Now that this disease has been found in North Dakota, it is imperative that we place greater resources on incorporating resistance to this pathogen in adapted lines.
Breeding Other Market Classes for the Northarvest Region
The objective of this program is to allow the dry bean growers of the Northern Plains to diversify into other market classes, thereby increasing the possibility of further exports. As indicated, breeding programs for dark and light red kidney, pink, small red, black, and great northern bean market classes also are underway. Pink, small red, and great northern market classes are closely related to pinto bean, while black is similar to navy bean; materials used to improve the two major market classes may also be used to improve these other, minor classes. We are currently re-evaluating the need for a cranberry program, as acreage has diminished in recent years. Plant rows were evaluated in 2001 and preliminary yield trials of black beans, pinks, and great northern also were grown. The 2002 season will see PYTs and AYTs of kidney, blacks, northerns, pinks, and reds in addition to pinto and navy.
White mold -- White mold (Sclerotinia sclerotiorum) continues to be a serious problem for dry bean growers and is ranked as the disease of most concern to producers. The potential for improved genetic resistance as a control measure has been demonstrated. Two major mechanisms of genetic resistance to white mold exist: 1) avoidance - usually associated with improved plant architecture; and 2) physiological - associated with biochemical functions at the cellular level. Generally, greenhouse and laboratory methods screen solely for physiological resistance, whereas field plantings screen for both physiological and avoidance mechanisms. Thus, greenhouse and laboratory resistance may not correlate well to field resistance. The breeding project has relied on evaluating material in the field at Page, where a 12 entry National White Mold Nursery also is grown. This national nursery is grown in Michigan, New York, Nebraska, Colorado, and Ontario, Canada.
An NDSU navy bean breeding line, 88-106-04, continues to provide good levels of resistance and is used in the NDSU program as a source of resistance. We are transferring this resistance into pinto bean, which is our greatest challenge. We are collaborating with USDA-ARS in this area in order to obtain an additional white mold site. This population was tested in two environments in 2001 (Prosser, WA at a site with center pivot irrigation) and near Hatton, ND (natural rainfall, natural infection, well protected). Lines differed dramatically for reaction to white mold, with several lines showing minimal white mold infection at both sites. One line, AN-37, had the best overall score for white mold tolerance. This line, which is early with a small pinto seed type, was incorporated into the 2002 pinto crossing block. Several NDSU navy bean breeding lines have shown excellent white mold resistance in both field and greenhouse trials, but this may be due, primarily to escape. We are continuing to develop populations using Othello and lines from the USDA Plant Introduction Collection that contain some level of white mold resistance, in cooperation with Dr. Jack Rasmussen, Plant Pathology, NDSU.
Rust -- Rust (caused by Uromyces appendiculatus) has, at times, been a severe disease problem, particularly for pinto bean producers. This fungal pathogen is composed of many races (at least 46 in the United States and over 200 worldwide) and has the capability to increase in variability, since it can complete its entire life cycle on the bean plant. Resistance usually is controlled in a single gene fashion, whereby one gene in the plant imparts resistance to one specific race of the pathogen. Often, this type of vertical resistance is short-lived, because the selection pressure placed on the rust race forces a change in the rust population. As a result, a new rust race is developed, and the former resistant variety becomes susceptible. This was observed in the pinto variety Olathe which, when first grown in North Dakota, was completely resistant, but was susceptible to at least one component of the rust population.
A rust nursery grown annually at Fargo is used to identify field response to the rust population that exists in North Dakota. The nursery is artificially inoculated with a collection of rust made the previous year. A collaborative nursery to determine potential adaptation of material into which exotic genes for resistance have been incorporated was planted at Erie. In 2001, we evaluated 128 lines (pinto, great northern and red) for reaction and overall adaptation. These lines were developed cooperatively between NDSU and USDA-ARS for high levels of resistance to rust and other bean diseases. From this effort, we have jointly released with USDA more than 28 lines possessing virtual immunity to rust. The breeding program assisted Dr. Jack Rasmussen, Plant Pathology Department, NDSU, in developing populations to understand the genetic control of rust resistance in Compuesto Negro Chimaltenango (CNC). This exotic black bean from Guatemala has excellent rust resistance that is unique from other resistance sources. While not effective against all races, the rust resistance from CNC, when combined with other major genes already present in NDSU bean breeding material, provides essential immunity to almost all races present in the U.S. Other sources of resistance also are being evaluated and molecular markers to assist in selection of resistant lines are being developed.
Root rot -- A root rot nursery was established at Perham to identify resistant genotypes and incorporate the resistance into adapted, high quality, high yielding kidney and cranberry beans. I have identified many lines with less disease damage, and some lines that essentially have no disease development on the roots. These lines, which include FR-266, Wisc. MDR 147, Wisc. RRR-36, and G-122, along with more exotic germplasm such as PR9434-17 as well as NDSU cranberry beans breeding lines have performed well the past three years in this evaluation nursery. We need to continue to evaluate material in the field because of the nature of the complex. Efficient evaluation of material can occur in the greenhouse, but evaluation for reaction to only one pathogen at a time can be made. In Minnesota, there may be as many as five species that may be causing damage: Fusarium, Pythium, Rhizoctonia, or Aphanomyces, so field screening is critical.
In 2000, we planted VAX 3 and VAX 5, two common bacterial blight lines developed at the International Center for Tropical Agriculture, Cali, Colombia, at Perham. I wanted to see how their CBB resistance held up in the field because I wanted to transfer their CBB resistance into kidney bean and cranberry bean. We found that these lines had excellent levels of resistance to root rot pathogens in the field and are now using these lines as sources of resistance to both CBB and root rot. This was again verified in the 2001 trials at Perham. We have developed populations with these lines as sources of resistance, which are being advanced rapidly in the greenhouse for further evaluation. VAX 3, which possesses a red seed coat, also has high levels of resistance to anthracnose, bacterial blights, and rust.
Bean common mosaic virus -- We continue to screen for bcmv resistance in order to assure production of quality seed. This virus, which is occasionally observed in ND, is prevalent in the PNW region, which affects seed production and, ultimately, supply. Selection for two resistance genes allows us to assure that the virus will not be problematic in NDSU material. Since the virus is seldom seen in ND, we evaluate material for reaction to this disease only during the winter months in the greenhouse. We have transferred resistance genes into Maverick and will begin yield testing these lines in 2002.
Anthracnose -- In 1998-99, we began identifying parental material that was used in the breeding program which possessed resistance to this disease. Several races of this disease are found worldwide, but the two most prevalent in the U.S. are races 7 and 73. Race 7 primarily attacks kidney bean and cranberry beans, but usually does little damage to cultivars of pinto, navy, black, and other market classes. Race 73, on the other hand, primarily incites damage to pinto, navy, black, and others, but usually has no effect on kidney and cranberry. Now that the disease has been found in ND, we have given breeding for resistance to this pathogen our greatest disease priority. Focusing our efforts on this problem will allow us to make good progress in incorporating resistance genes into adapted material. We are primarily interested in using the Co-42 gene for resistance, simply because we can screen for the presence of this gene in the absence of the pathogen by using molecular markers. This will allow us to rapidly backcross this resistance gene into desirable genotypes. NDSU is collaborating with USDA-ARS personnel in this area. Other sources of resistance also will be exploited, because our greatest chance of controlling this pathogen is to combine several resistance genes into our cultivars.
Germplasm Enhancement, Use of Molecular Markers
As described elsewhere, the breeding project is using laboratory procedures in an attempt to identify molecular markers associated with white mold resistance. Breeders could then select for the presence of the associated marker for the desired trait. The dry bean breeding project has shared laboratory space with the barley and oat breeding projects. Work at other laboratories has identified markers associated with three rust resistance genes, anthracnose resistance gene, and BCMV resistance genes. We are focusing on white mold resistance introgressed into pinto bean, rust resistance, root rot resistance, and quality traits. As additional markers are identified, it may be possible to select for a number of traits simultaneously using only leaf tissue. One potential aspect of this research would be to identify desirable genes in related species (e.g., drought tolerance, heat stress tolerance, and common blight resistance, all from tepary bean, Phaseolus acutifolius, and white mold resistance from scarlet runner bean, P. coccineus). We are working with USDA-ARS and the University of Puerto Rico on white mold resistance and use of other species as sources of resistance, respectively.
We have developed populations to identify markers associated with genetic resistance for white mold and root rot from the core collection evaluation. We also routinely use molecular markers to evaluate for the presence of genes resistant to bean common mosaic virus and rust. We will begin routine evaluation of material for markers associated with common bacterial blight resistance.
Our next challenge is to combine use of molecular markers and greenhouse screening for anthracnose. Plant Pathology personnel will evaluate lines for reaction to anthracnose using greenhouse-grown seedlings, while we will concentrate on molecular markers. Information obtained from both sources will help us make important decisions on disease reactions and host genetics. We can utilize a molecular marker for the Co-42 resistance gene for anthracnose, which will provide effective resistance to most anthracnose strains in the U.S. and Canada.
The dry bean breeding program at NDSU has grown considerably since originating in 1980. Each year, several hundred unique hybridizations are made to develop improved genotypes in each of nine market classes. In addition, successful collaboration with scientists in several departments at NDSU and scientists at other institutions in the U.S. offer unique opportunities to expand the knowledge base of bean genetics and production.