RAPD markers linked to a block of genes conferring rust resistance to the common bean

Rust, caused by the fungus Uromyces appendiculatus , may cause a significant loss to common bean ( Phaseolus vulgaris L.) yield. RAPD markers tightly linked to the resistance genes may be used in breeding programs to aid the development of rust-resistant b ean cultivars. In this sense, the objective of the present work was to identify RAPD markers linked to a rust resistance gene block present in the cultivar Ouro Negro. Two hundred and fourteen F 2 individuals from a cross between the resistant cultivar Ouro Negro and the susceptible cultivar US Pinto 111 were inoculated with a mixture of eight races of U. appendiculatus . The segregation ratio obtained suggested that resistance is monogenic and dominant. Bulked segregant analysis was used in conjunction with the RAPD technique to search for markers linked to rust resistance genes. Two molecular markers flanking the rust resistance gene block were identifi ed, one at 5.8 ± 1.6 cM (OX11 630 ) and the other at 7.7 ± 1.7 cM (OF10 1,050 ) of the gene. Simulated indirect selection efficiency in the F 2 population using the two markers was 100%. The molecular markers identified in this work are currently being used for the selection of dis easeresistant plants in the commom bean breeding program of the Federal University of Viçosa. *Part of a thesis presented by F.G.F. to the Federal University of Viçosa, in partial fulfillment of the requirements for the P hD degree. Instituto de Biotecnologia Aplicada à Agropecuária (BIOAGRO), Departamentos de Biologia Geral and Bioquímica e Biologia Molecular, Universidade Federal de Viçosa (UFV), 36571-000 Viçosa, MG, Brasil. Send correspondence to E.G.B. E-mail: ebarros@mail.ufv.br


INTRODUCTION
The common bean (Phaseolus vulgaris L.) is susceptible to be attacked by a large number of pathogens which may qualitatively and quantitatively affect plant growth and yield (Vieira, 1983;Stavely and Pastor-Corrales, 1994).In particular, rust caused by the fungus Uromyces appendiculatus may cause a significant loss of yield (Lindgren et al., 1995).
Recommended measures for controlling rust include crop rotation and the elimination of contaminated material (Hall and Nasser, 1996).Fungicides and resistant cultivars are also useful for limiting the progress of this disease (Stavely and Pastor-Corrales, 1994).
The chemical control of rust is rarely employed by small farmers who do not use irrigation and advanced cultivation technologies, mainly because of the need for specialized knowledge in handling these compounds and because of the cost involved.In addition, chemicals contaminate the environment and affect human health.For these reasons, the use of resistant cultivars is preferred by most plant breeders and researchers because they provide an effective, safe and inexpensive method of rust control.
Conventional breeding methods have been widely used to develop new cultivars resistant to U. appendiculatus (Coyne and Schuster, 1975).However, resistance is easily overcome by the extensive pathogenic variability of the fungus (Faleiro, 1997).New strategies to bypass this problem, such as the pyramiding of resistance genes, have been pro-posed.According to Kelly et al. (1994), the pyramiding of three rust resistance genes (Ur-4, Ur-5 and Ur-3) would be enough to confer resistance to 63 of the 65 races of U. appendiculatus identified in the United States.However, the use of this strategy in conventional breeding procedures is extremely difficult, mainly because of the need for multiple inoculations which may lead to erroneous evaluation of the symptoms (Vinhadelli et al., 1997).
Random amplified polymorphic DNA (RAPD) (Williams et al., 1990) markers have been used in breeding programs to help in the production of rust-resistant bean cultivars (Haley et al., 1993;Miklas et al., 1993;Kelly et al., 1994).Markers tightly linked to the resistance genes may be used for the indirect selection of resistant plants in segregating populations, without the need for multiple inoculations.
The objective of the present work was to identify RAPD markers linked to a rust resistance gene block present in the cultivar Ouro Negro, one of the major rust resistance sources used in the bean breeding program of the Federal University of Viçosa, MG, Brazil.races of U. appendiculatus identified in Brazil (Faleiro et al., 1996), whereas cultivar US Pinto 111 is considered to be universally susceptible to rust.
F 2 seeds were sown in a greenhouse in 2.5-liter pots containing substrate.Ten days after sowing, when the primary leaves had reached approximately two-thirds of their full development, the plants were inoculated with a mixture containing equal amounts of spores from U. appendiculatus races 1, 2, 6, 7, 9, 10, 11 and 13 identified by Faleiro et al. (1999b).The spores were suspended in distilled water containing 0.05% Tween 20 and the suspension was diluted to give approximately 2 x 10 4 spores/ml.The inoculum was sprayed onto both leaf surfaces with the aid of a De Vilbiss No. 15 atomizer powered by an electric compressor (Carrijo et al., 1980).After a brief drying period, the plants were transferred to a mist chamber (20 ± 1°C and >95% relative humidity) where they were kept for 48 h, under a 12-h photoperiod.The plants were then returned to the greenhouse where they were evaluated for disease symptoms 15 days after inoculation, using a scale reported by Stavely et al. (1983).
After symptom evaluation, leaf DNA was extracted from each F 2 plant (Doyle and Doyle, 1990).To identify the RAPD markers linked to the resistance gene, DNA bulks from contrasting plants were prepared as described by Michelmore et al. (1991).DNA amplification was done according to Vasconcelos et al. (1996).A total of 1,280 decamer primers were tested (Operon Technologies, Alameda, CA, USA).Amplification products were separated in 1.2% agarose gels containing 10 µg ethidium bromide/ml, visualized under UV light and photographed with an Eagle Eye II photodocumentation system (Stratagene, La Jolla, CA, USA).
To confirm the segregation between the marker and the resistance phenotype, primers which produced polymorphic bands between the two bulks were tested in the progenitors and 214 F 2 plants.The data were analyzed by the chi-square test to verify the segregation ratios.Recombination frequencies and genetic distances between the markers and the resistance gene block were estimated using the software MAPMAKER.EXP 3.0 (Lander et al., 1987;Lincoln et al., 1992).

RESULTS AND DISCUSSION
The segregation analyses suggested that a major dominant gene confers rust resistance in cultivar Ouro Negro (Table I).Preliminary studies on resistance inheritance in this cultivar using one U. appendiculatus race indicated that major genes are indeed the main determinants of bean rust resistance (Faleiro, 1997).
Two RAPD markers linked in coupling-phase to a rust resistance gene block were identified: one with 630 bp (OX11 630 ) and the other with 1,050 bp (OF10 1,050 ).The amplification patterns with the two primers generating these markers were highly reproducible (Figure 1).Segregation analyses of the markers in the F 2 population confirmed their dominant nature (Table I).The number of recombinants showed that markers OX11 630 and OF10 1,050 are located at 5.8 ± 1.6 cM and 7.7 ± 1.7 cM of the gene block, respectively (Figure 2).
That the two markers flank the resistance gene block (Figure 2) is extremely important for the indirect selection of resistant plants.Two cross-over events would be necessary to give rise to a resistant plant bearing no marker.This is particularly evident when the indirect selection efficiency (ISE) is estimated using the markers separately and simultaneously.Preliminary estimates of ISE in 214 individuals of the F 2 population were 96.6 and 95.3% using the markers OX11 630 and OF10 1,050 , respectively.The simultaneous use of the two markers gave an ISE of 100% (Table II).
RAPD markers linked to a resistance gene block have been reported by others.Miklas et al. (1993) identified a marker (OA14 1,100 ) tightly linked to the resistance gene Ur-4 (no recombinants in 84 BC 6 F 2 individuals).Haley et al. (1993) identified a marker (OF10 970 ) linked at 2.15 cM of the rust resistance gene block present in line B-190 and a marker OI19 460 tightly linked to the same gene block (no recombinants in 97 BC 6 F 2 individuals).Haley et al. (1994) identified a marker (OK14 620 ) linked at 2.23 cM of the resistance gene Ur-3.All these markers were used in the pyramiding of these resistance genes in the same genetic background (Stavely et al., 1994(Stavely et al., , 1997)).
These markers were also tested in our segregating population, but none were polymorphic and, in some cases, the corresponding DNA bands were not detected (data not shown).However, as already discussed, primer OF10 produced a 1,050-bp fragment linked to the rust resistance gene block present in cultivar Ouro Negro.This primer also produced a DNA fragment (970 bp) linked to a rust resistance gene block in line B-190 (Haley et al., 1993).The small difference between the two bands is apparently related to the way their sizes were estimated.The resistance gene blocks present in line B-190 and in cultivar Ouro Negro are very similar and probably map to the same location.Both confer the same type of resistance, with immunity to certain U. appendiculatus races and small pustules (< 0.3 mm in diameter) to other races (Haley et al., 1993;Faleiro et al., 1999a).Interestingly both Ouro Negro and B-190 are highly productive black seed cultivars.
The molecular markers identified in this work are currently being used in the selection of disease-resistant plants in the common bean breeding program of the Federal University of Viçosa.

ACKNOWLEDGMENTS
This research was supported by a grant from PADCT/FINEP and FAPEMIG.F.G.F. was the recipient of a CAPES fellowship.

Figure 2 -
Figure 2 -The positions of markers OX11 630 and OF10 1,050 , expressed as genetic distances (cM), relative to the rust resistance gene block.

Table I -
Segregation analyses between rust resistance and the molecular markers OX11 630 and OF10 1,050 in the F 2 population derived from a cross between cultivars Ouro Negro and US Pinto 111.
A B Figure 1 -Electrophoretic analyses of DNA amplification products obtained with primers OX11 (A) and OF10 (B).In both gels, P 1 corresponds to cultivar Ouro Negro, P 2 to cultivar US Pinto 111 and F 2 to 27 individuals from the segregating population.Lane M contains lambda phage DNA digested with EcoRI, BamHI and HindIII (size markers).The arrows indicate markers OX11 630 (A) and OF10 1,050 (B).

Table II -
Indirect selection efficiency (ISE) of rust resistant plants based on the presence/absence of single and associated molecular markers.