Crop Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 19 March 2008
Published in Crop Sci 48:562-570 (2008)
© 2008 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Garzón, L. N.
Right arrow Articles by Blair, M. W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Garzón, L. N.
Right arrow Articles by Blair, M. W.
Agricola
Right arrow Articles by Garzón, L. N.
Right arrow Articles by Blair, M. W.
Related Collections
Right arrow Other Legumes
Right arrow Marker-assisted selection

Molecular Marker-Assisted Backcrossing of Anthracnose Resistance into Andean Climbing Beans (Phaseolus vulgaris L.)

Luz Nayibe Garzóna, Gustavo A. Ligarretoa and Matthew W. Blairb,*

a Universidad Nacional de Colombia, Facultad de Agronomía, Ciudad Universitaria, Bogotá D.C.- Colombia, Av. Carrera 30 N° 45-03
b International Center for Tropical Agriculture (CIAT), Recta Cali-Palmira, Km. 17. A.A.6713, Cali-Colombia

* Corresponding author (m.blair{at}cgiar.org).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Anthracnose, caused by the fungus Colletotrichum lindemuthianum, is considered a major constraint in the production of common bean (Phaseolus vulgaris L.). This study aimed to evaluate, in a backcross plant-breeding program, the efficiency of selecting plants resistant to anthracnose using marker-assisted selection (MAS) for two resistance genes, Co-5 and Co-42 derived from the resistance source G2333 based on the linked PCR based markers SAB3 and SAS13. The amplification of both markers was compared using DNAs extracted with two techniques, alkaline extraction, which is a fast, and inexpensive method for high throughput screening; versus a proteinase K based miniprep extraction, which is more time consuming but provides more DNA. To further evaluate the effectiveness of the markers in selecting for resistance, we compared the marker genotypes and observed phenotypes for 266 plants from eight backcross families inoculated with a field isolate of anthracnose. The Co-5 gene and SAB3 proved to be useful and the markers associated with Co-5 and Co-42 could be pyramided to give added levels of anthracnose resistance.

Abbreviations: AE, alkaline extraction method • BC, backcross • F as in F1, F1:2, etc., filial generation • MAS, marker-assisted selection • PCR, polymerase chain reaction • PK, proteinase K extraction method • SCAR, sequence-characterized amplified region


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
COMMON BEAN (Phaseolus vulgaris L.) is considered one of the world's most important legumes (Broughton et al., 2003), and anthracnose, caused by Colletotrichum lindemuthianum (Sacc. & Magn.) Bri. & Cav., is a major disease of the crop, causing losses of up to 100% when contaminated seed is planted or when prolonged favorable conditions for disease development occur during the growing cycle (Schwartz, 1991). The disease pathogen is distributed worldwide, wherever common bean is cultivated (Beebe and Pastor-Corrales, 1991). Co-evolution is known to have occurred between fungal pathotypes and the common bean's two gene pools so that Andean races of the fungus usually attack Andean bean genotypes and Mesoamerican races attack Mesoamerican genotypes (Balardin and Kelly 1998; Melotto et al., 2000). Congruent with this co-evolution, the best sources of resistance for breeding programs are often found in complementary gene pools and genes for resistance usually must be introgressed through backcrossing especially of Mesoamerican genes into Andean backgrounds (Miklas et al., 2006).

A range of control strategies have been suggested for the management of anthracnose, but varietal resistance is considered to be the most effective, economical, and easy for farmers to adopt (Pastor-Corrales et al., 1995; Schwartz et al., 1982). In a recent review of resistance genes, Kelly and Vallejo (2004) listed ten loci for anthracnose resistance identified as Co-1 to Co-10; however, Co-3 and Co-9 are now known to be allelic. Gonçalves-Vidigal and Kelly (2004) identified a tenth gene, Co-11. Geffroy et al. (2000) have reported other anthracnose resistance genes and quantitative trait loci as well. Except for co-8, all the numbered genes are dominant; and multiple alleles have been described for Co-1, Co-3, and Co-4 (Kelly and Vallejo 2004). An allele of the Co-4 locus, named Co-42, is considered to have a broader spectrum of resistance than other resistance genes and is derived from the differential genotype G2333 (Balardin and Kelly 1998), which also possesses the genes Co-7 and Co-5 (Alzate-Marin et al., 2001; Young et al., 1998). The Co-42 gene has been successfully used in breeding programs conducting marker-assisted selection (MAS) (Alzate-Marin et al., 2004; Miklas and Kelly 2002; Miklas et al., 2003). In contrast, gene Co-5 has not been extensively used, even though it is considered to be highly effective against C. lindemuthianum races from Central America and Mexico (Campa et al., 2005; Kelly and Vallejo, 2004).

Various sequence characterized amplified region (SCAR) markers have been developed for the three known anthracnose resistance genes in G2333. The first to be developed was a dominant marker named SAS13 that was very tightly linked (0.0 cM) to the Co-42 gene (Young et al., 1998). This marker was later determined to probably be part of the resistance gene, which is postulated to encode a protein kinase involved in pathogen recognition (Melotto et al., 2004). The second marker was another dominant SCAR called SAB3 that was linked, but not tightly (12.98 cM), with the Co-5 gene (Vallejo and Kelly, 2001). More recent markers for Co-42 have been developed (SH18 and SBB14) and are more genotype specific but less closely linked to the resistance locus than SAS13 (Awale and Kelly, 2001).

Backcrossing has been suggested as one of the best methods to incorporate anthracnose resistance genes into susceptible materials that are commercially acceptable (Alzate-Marin et al., 2004). Marker-assisted selection (MAS) can improve the process of backcrossing by selecting the resistance gene as a smaller introgression than is possible without MAS, allowing a faster genotypic return to the recurrent parent and reduced linkage with unwanted genes (Tanksley et al., 1989). One important factor that must be considered for MAS is the type of DNA extraction used, with the most adequate methods being those that permit the handling of large numbers of samples, thus reducing time and costs and allowing more segregants to be screened for a given trait (Karakousis and Langridge, 2003; Xu et al., 2005).

Various DNA extraction techniques have been developed, varying in amount of tissue used and quantity and quality of DNA obtained (Hosaka, 2004; Porebski et al., 1997). The most useful, and inexpensive methods are those that do not use expensive laboratory chemical or multiple processing steps and that are easily scalable or adapted to high-throughput evaluation (Francia et al., 2005). For common bean, several "miniprep" methods have been developed that rely on organic solvents (Afanador and Hadley, 1993) or the use of high-salt solutions for DNA precipitation (Mahuku, 2004). Among the most inexpensive methods to have been implemented in common beans is alkaline extraction (Blair et al., 2007), but no study has determined the comparative advantages of different DNA extraction methods for marker-based selection in common bean.

The goal of this study was to compare markers and DNA extraction techniques for the implementation of a marker assisted selection program for anthracnose resistance in backcross breeding for six Andean climbing-bean genotypes using the G2333 source of resistance and two SCAR markers for the Co-42 and Co-5 genes. The specific objectives were (i) to compare the alkaline extraction method for DNA isolation (AE), with a proteinase K "miniprep" method (PK) using differential cultivars for anthracnose in common bean and a set of BC1F1 segregants; (ii) to determine the efficiency of marker assisted selection based on the disease reaction of BC1F1:2 segregants using artificial inoculation compared to evaluation with SAB3 and SAS13 using DNA from the AE method; and (iii) to pyramid the two resistance genes through marker assisted selection in one of the climbing-bean backcrosses analyzed.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Plant Materials
Two sets of genotypes were used for this research. The first group consisted in the twelve differential cultivars described by Pastor-Corrales (1991) for anthracnose resistance in common beans. These differentials were selected because they had been used in previous studies for the markers of interest (Kelly and Vallejo 2004; Melotto et al., 2000). The second group of genotypes was derived from backcross populations produced by crosses between G2333 and six commercial varieties. In this case, four of the commercial varieties were large red, bola roja type, (‘Cabrera’, ‘Moreno’, ‘Pesca’, and ‘Simijaca’), one was a pink mottled type (‘Agrario’) and one was a red mottled type (‘Cargamanto’). In all cases, the commercial varieties were used as recurrent parents while G2333 was used as the donor parent. Seed from F1 and BC1F1 populations was obtained through crosses performed under field conditions using artificial hybridization with each of the commercial varieties as maternal parents and G2333 or the corresponding F1 hybrid as the paternal parent. Individual plants from the first backcross generation were allowed to self to derive BC1F1:2 families from individually harvested plants for the backcross with Pesca. The commercial varieties were representative of grain types popular with consumers in the Andean region (including Colombia and Ecuador) and are adapted to cool climates. All are climbing beans from the Andean genepool that are very susceptible to C. lindemuthianum (Ligarreto, 1997; Santana and Mahuku, 2002). G2333 is a small red seeded climbing-bean genotype from the Mesoamerican gene pool that is highly resistant to anthracnose races in various regions of the world (Pastor-Corrales et al., 1994). Seed of G2333 was provided by International Center for Tropical Agriculture (CIAT) (http://isa.ciat.cgiar.org/urg/ verified 21 Jan. 2008), while the commercial varieties were provided by the National University of Colombia (Ligarreto, 1997).

DNA Extraction
Two DNA extraction techniques were used, one based on alkaline extraction (AE) and the other based on proteinase K extraction (PK). The AE method protocol was as described by Klimyuk et al. (1993) but modified for use in common beans. Briefly, a hole-puncher was used to remove 6-mm disks of tissue from young leaflets of each individual plant, which were transferred individually to the wells of a 96 microtitre plate using sterile forceps. The AE method was begun by adding 40 µL of 0.25 M NaOH and incubating the samples in a water bath at 100°C for 2 min. The tissue was then macerated with a small plastic pestle and neutralized with 40 µL of 0.25 M HCl, which was followed immediately by the addition of 20 µL of a 0.5 M Tris-HCl pH 8.0 solution. The tissue was again incubated in a water bath at 100°C for 2 min then left to cool to room temperature, on which different dilutions of the DNA extract were prepared (1:1, 1:5, 1:10, 1:20, 1:30, and 1:50). The PK method was based on the miniprep procedure from Mahuku (2004), which involved extracting DNA with ammonium acetate, inactivating proteins with SDS/proteinase K and precipitating polysaccharides under high salt conditions. DNA concentrations from the PK method were estimated using a Hoefer DyNA fluorometer (Hoefer Pharmacia Biotech, Inc., San Francisco, CA) (DNA Quant 200). DNA isolated from the AE method, was not measured for concentration given that the dilution series was used instead.

PCR Amplification
Two SCARs, SAB3 linked with the Co-5 gene and SAS13 linked with the Co-4 gene, were used for marker-assisted selection. For SAB3, we initially used the amplification conditions as described by Vallejo and Kelly (2001), while for SAS13, we followed the recommendations of Young et al. (1998). For both amplification protocols, we added an initial cycle at 92°C for 3 min and denaturation in each cycle was performed at 92°C for 30 s. The conditions were also modified with a step-down amplification profile modifying the annealing temperature for each marker, 5°C above and 5°C below that reported for SAB3, and 2°C above and 5°C below that reported for SAS13 with 39 cycles for both markers. All amplification reactions were performed on a PTC-200 thermal-cycler (MJ Research, Watertown MA) with a total reaction volume of 15 µL containing 5 µL of DNA of the appropriate dilution for alkaline extraction DNA, or 5 µL of a 10 ng/µL concentration for the miniprep extraction DNA, 1X PCR buffer (500 mM KCl, 10 mM Tris-HCl pH 8.8, 1% Tritron X-100, 1 mg mL–1 bovine serum albumin), 0.10 µM of each primer (Invitrogen Corp., Carlsbad, CA), 0.25 mM of each dNTP, 2.5 mM MgCl2, 1 unit Taq DNA polymerase, and HPLC quality H2O. After amplification the PCR products were evaluated by adding 2 µL of loading buffer (glycerol at 30% and bromophenol blue at 0.25%) to the samples and analyzing them in horizontal electrophoresis chambers (FB-SB-2025) using agarose gels at 1.5% with 0.5 µL mL–1 of ethidium bromide submerged in 0.5x TBE buffer (90 mM Tris-borate; 2 mM EDTA, pH 8.0) and run at 200–220 V for 30 min. The bands were visualized under ultraviolet light and photographed with a Kodak DC290 gel documentation camera.

Inoculation and Disease Reaction
Disease reaction was evaluated by inoculating BC1F1:2 families and controls as described below with spores of C. lindemuthianum at a concentration of 1.2 x 106 conidia per mL in water at 15 d after planting with seedlings in the primary leaf stage. The control genotypes included G2333 used as a positive control for resistance and the commercial variety and recurrent parent Pesca as well as ‘La Victorie’, as negative controls for susceptibility. La Victoire is a snap bean reported as being highly susceptible to anthracnose by Pastor-Corrales et al. (1994). The inoculum was provided by the legume project at the National University of Colombia (Bogotá, Colombia) and consisted of a mixture of monosporic cultures, originating from infected pods and leaves from Cundinamarca and Boyacá, Colombia, representing race 1 according to the race classification system of Pastor Corrales (1991). After inoculation, the plants were taken to a greenhouse where they were kept at temperatures of 20 to 22°C with a relative humidity of approximately 100%. Ten days after inoculation, the plants were evaluated using the 1 to 9 scale proposed by Schoonhoven and Pastor-Corrales (1987), where plants with disease-reaction scores of 1 to 3 were resistant and plants with scores of 4 to 9 were susceptible.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Standardization of the Alkaline Extraction Procedure
The twelve differential cultivars for anthracnose described by Pastor-Corrales (1991) were used to standardize the AE method as shown in Fig. 1 . The two SCAR markers worked well with DNA from the AE method, presenting the same expected band sizes as with the PK method DNA for the source genotype G2333. For SAB3, the same 400 bp band was observed for G2333 extracted either with AE (lane 5) or PK (lane 13) methods. Similarly, for SAS13 the expected 950 bp band was found for G2333 based on both types of DNA (lanes 5 and 13). The only discrepancy for the amplification of the two markers with AE compared to PK methods was found for SAS13 which showed a positive band for TO, PI 207262 and ‘Widusa’ with PK but not with AE. The amplification of SAS13 for these genotypes agrees with the results of Young et al. (1998) who found that this marker was not specific to G2333 and other genotypes containing Co-42, but also amplified a band in several genotypes that had the dominant Co-4 allele. The lack of amplification of microprep DNA for these three genotypes may be due to allele specificity that is detected with the lower quality DNA of the AE method as compared to the higher quality DNA of the PK method. Meanwhile we also observed that the SAB3 marker amplified a band from TU (lane 12; Fig. 1), which agrees with the results of Vallejo and Kelly (2001) who tested some of the differentials during the marker development process and found TU to be positive for this marker.


Figure 1
View larger version (72K):
[in this window]
[in a new window]

  		Figure 1. PCR amplification of the proteinase K (PK) extraction DNA (panel A) and alkaline extraction (AE) method DNA (panel B) for the SCAR markers SAS13 and SAB3 in the anthracnose differential genotypes from Pastor-Corrales (1991): MDRK (lanes 1 in both panels and both markers), ‘Perry Marrow’ (lanes 2), ‘Kaboon’ (lanes 3), ‘Michelite’ (lanes 4), G2333 (lanes 5), AB136 (lanes 6), Cornell 47–292 (lanes 7), Mexico 222 (lanes 8), PI 207262 (lanes 9), ‘Widusa’ (lanes 10), TO (lanes 11), TU (lanes 12). Lane 13 represents control reaction of proteinase K miniprep DNA for G2333 in panel B with molecular weight ladder ({lambda}-HindIII) indicated with ‘M’ in both panels.

 
The amplification of the two SCAR markers with template DNA obtained through the AE method required the use of a DNA concentration curve that varied the dilution factor from 1:10 to 1:50. The dilution factor of 1:20 enabled us to obtain optimal quality of amplification for both SCARs. The dilution factor is probably important given that for the AE method, no DNA purification steps are performed before its use in the PCR reactions. This means that the DNA may have contaminants, including polysaccharides, polyphenolics, and other secondary compounds. The concentration of these contaminants may be higher in the lower dilutions such as 1:1, 1:5, or 1:10, explaining why PCR reactions fail at these dilutions and only work at higher dilution factors such as 1:20 or more.

In standardizing the amplification conditions fewer modifications were made for the SAB3 marker than for the SAS13 marker. The ease of amplification of the SAB3 product, compared to the SAS13 marker even with the lower quality DNA, could be due to its smaller product size of only 400 bp given that Klimyuk et al. (1993) suggested that the AE method is more efficient when amplifying small products.

Alkaline Extraction versus Proteinase Potassium Extraction
To further compare the amplification of DNAs extracted with AE vs. PK methods in a real-life MAS setting, we evaluated 158 BC1F1 plants derived from crosses between the six Andean climbing varieties as recurrent parents and G2333 as the source of resistance (Table 1 ). Using DNA from the PK method as a control in these 158 plants, we obtained six false negatives with SAB3, which corresponded to a correlation value of r = 0.93 (P = < .0001), and three false negatives with SAS13, which corresponded to a correlation value of r = 0.97 (P = < .0001). In certain cases the AE method DNA was found to be more reliable than the PK method given that amplification failures were also produced with the PK method DNA (Fig. 2 ). A total of 13 samples (seven with SAB3 and six with SAS13) amplified with DNA obtained from the AE method but not with DNA from the PK method. To rule out error, DNA extractions and PCR reactions were repeated and amplification was obtained with the respective SCARs on this second run, confirming that the previous samples had been false negatives. Therefore, although overall the PK method was found to be an effective DNA extraction technique, the cleaning step in the process of decontaminating DNA in this technique may not be sufficient for some tissue samples and may somewhat reduce the repeatability of sample evaluation with this DNA extraction technique. The risk of false negatives using the PK method may also be due to degradation of DNA during the high temperature step needed for denaturing proteinase K which is used to prevent this enzyme from inhibiting the PCR reaction (Dilworth and Frey, 2000; Mahuku, 2004; Steiner et al., 1995).


View this table:
[in this window]
[in a new window]

  		Table 1. Marker amplification for SAB3 and SAS13 SCAR markers in 158 BC1F1 plants from six backcross (BC) populations using DNA prepared by alkaline extraction (AE) or proteinase K (PK) DNA extraction methods.

 

Figure 2
View larger version (76K):
[in this window]
[in a new window]

  		Figure 2. SAB3 marker amplification in a subset of BC1F1 plants from the marker-assisted backcross-breeding program using either DNA from a) the PK method or b) the AE method. Lanes 1– 15 represent individual plants from the cross Pesca x (Pesca x G2333). Molecular weight ladder (M) indicated in left hand lane. Positive (R) and negative (S) control genotypes, G2333 and Pesca indicated in right hand lanes. Arrows indicate false negatives in the PCR amplification of the PK method DNA.

 
The results demonstrate consequently the usefulness of the AE method DNA, which generated almost the same positive and negative results as with PCR amplification of the PK method DNA. Since DNA extraction is an expensive component of molecular breeding, and consumes 30 to 60% of the total time needed for processing samples (Dilworth and Frey, 2000), the achievement of reliable PCR amplification with a high-throughput AE method is very promising. Time-savings for AE compared to other methods arise from the parallel output of 96-well plate processing and the fact that liquid nitrogen is not needed for the initial, individual-tube grinding of leaf tissue, a common step in most miniprep methods. Furthermore, the AE method represents a cost-savings compared to miniprep methods that require organic solvents or commercial proteins such as those of the methods of Afanador and Hadley (1993) and Mahuku (2004), respectively. In terms of laboratory safety and the generation of hazardous waste products, the AE method also compares favorably to those methods involving chloroform or phenol extractions (Porebski et al., 1997). In summary, the AE method minimizes time and use of laboratory materials, increases lab safety, and enables hundreds of samples to be analyzed rapidly.

One important consideration in the implementation of the AE method was the standardization of the amount of plant material used for this method and the amount of DNA homogenate from this protocol used in the PCR reaction. Excess homogenate can inhibit the PCR reaction, most likely because of a high concentration of plant proteins and secondary metabolites in the undiluted AE-method DNA (Dilworth and Frey, 2000). Thus, the quantity of tissue used in the AE method was minimal to reduce the amount of contaminants. In this regard our use of 6-mm leaf disks seemed ideal. At the same time, we estimated that the quantity of DNA extracted per sample with one of these leaf disks in the AE method was sufficient for 380 PCR reactions based on our amplification conditions and optimum dilution factor. An additional advantage of using 6-mm leaf disks was high uniformity in tissue amounts and the resulting DNA extract.

Both procedures used in this study were effective for screening of SCAR markers perhaps because these markers work well with intermediate-quality DNA obtained through rapid extraction procedures (Kelly and Miklas, 1998). The techniques furthermore involved small amounts of tissue, simple procedures, and minimal quantities of chemicals as recommended for MAS by Zhang and Stewart (2000). The AE method especially was extremely fast, and cost-effective, with reagents and supplies costing as little as US $0.06 per sample, and where one technician could process 96 leaf samples in 20 min. The characteristics of this type of extraction therefore permitted an efficient evaluation of large breeding populations.

Efficiency of Marker-Assisted Selection
To determine the efficiency of marker-assisted selection using the AE method on a mass-scale in selecting anthracnose resistance, we decided to evaluate a group of 266 BC1F1;2 plants representing eight BC1F1-derived families from the backcross Pesca x (Pesca x G2333), with disease phenotyping and marker genotyping as shown in Table 2 . After disease reaction screening with C. lindemuthianum inoculation and marker evaluation for markers SAB3 and SAS13, the appropriate chi-square statistics were calculated for both phenotypic and genotypic information based on the expected segregation for the eight BC1F1-derived families and 25 susceptible and resistant check plants.


View this table:
[in this window]
[in a new window]

  		Table 2. Segregation for SAB3 and SAS13 marker genotypes using the alkaline extraction (AE) method and anthracnose disease reaction in eight BC1F1:2 families generated from the Andean climbing bean ‘Pesca’ and the resistance source G2333.

 
Results of this trial showed that all of the plants for the negative checks Pesca and La Victorie were susceptible while the resistant check G2333 did not show any symptoms of disease. Meanwhile, the expected phenotypic:disease and genotypic:marker segregation ratios tested for the BC1F1:2 families were either 3:1 or 15:1, resistant to susceptible, based on the marker genotypes and predicted phenotypes of the original BC1F1 plant and the observed segregation ratios and chi-square values as shown in Table 2.

The observed segregation fit a 3:1 resistant to susceptible expected ratio for the BC1F1:2-1 family while for the BC1F1:2-2 family, a higher number of susceptible plants than expected was obtained. For both families, the only marker that amplified was SAB3, and segregation for this marker adjusted to the expected ratio of 3:1, as expected from the genotype of the original BC1F1 plant. For the BC1F1:2-1 and BC1F1:2-2 families, amplification of SAB3 and resistance screening indicated the presence of the Co-5 gene and that this gene is effective against the anthracnose isolate used in the inoculation. Meanwhile, of the families selected from BC1F1 plants carrying the SAS13 marker, the BC1F1:2-4 family fit a 3:1 resistant-to-susceptible ratio, while for the BC1F1:2-3 family there were a higher number of susceptible plants than expected. The genotypic evaluation showed no amplification of the SAB3 marker while segregation of the SAS13 marker adjusted to the expected 3:1 ratio. The amplification of this marker, which is associated with the Co-42 gene, would explain the resistance observed.

Among the families selected from BC1F1 plants carrying both the SAB3 and SAS13 markers, the observed ratio fit a 15:1 resistant-to-susceptible phenotypic segregation ratio for the BC1F1:2-6 family but not for the BC1F1:2-5 family. These results may indicate that duplicate dominant genes were segregating in the first of these families but not in the second. In the BC1F1:2-6 plants, amplification of SAB3 adjusted to the expected ratio, whereas, with SAS13, bands were observed in fewer individuals than expected. Since phenotypic evaluations adjusted to the expected ratio of 15:1, the smaller than expected number of SAS13 amplifications might be explained by the failure of the marker to amplify with DNA from some of the BC1F1:2 plants from this family. The high level of resistance present in this family (93% of the plants in the family) highlights the importance of the digenic interaction between Co-5 and Co-42, and the advantage obtained on being able to pyramid them in a single population. Meanwhile, in the BC1F1:2-5 family, SAB3 amplified in 42 individuals and SAS13 in 38 individuals showing that both markers had been present in the original BC1F1 plant; however, a 3:1 resistant-to-susceptible ratio was observed, indicating that a single resistance gene had been present in that original plant. An early recombination event between SAB3 and Co-5 may have occurred changing the phase from coupling to repulsion for this marker and separating the resistance gene from the SAB3 marker. This would have left only Co-42 to provide resistance and thus would explain the observation of monogenic resistance in this family. SAB3 and Co-5 are known to be less tightly linked than SAS13 and Co-42 (Kelly and Vallejo, 2004).

Among the families selected from BC1F1 plants carrying neither of the markers, no amplification was observed for either marker in the BC1F1:2-7 family, adjusting to the expected. In contrast, in the BC1F1:2-8 family segregation for SAS13 was observed. The lack of amplification of both SCARs in the first of these families indicated the absence of the Co-5 and Co-42 genes. This was confirmed by the phenotypic evaluation of disease reaction, in which all the individual progeny presented susceptibility. Meanwhile, the BC1F1:2-8 family was expected to be highly susceptible but was shown to segregate for resistance instead. This inconsistency could be explained as a result of a false negative at the time of evaluation for SAS13 in the BC1F1 population. This hypothesis was confirmed in the segregation ratios for the markers themselves where no plants showed amplification for SAB3 but SAS13 amplification was observed in 10 individuals, adjusting to the segregation expected for dominant monogenic resistance. The results confirmed that the BC1F1:2-8 family segregated for SAS13 and the Co-42 gene.

Overall our results on the inheritance of anthracnose resistance agree very well with previous studies. First, our findings of duplicate dominant gene epistasis in one of the G2333 derived backcross families agree with the first inheritance study of anthracnose resistance in this accession by Pastor-Corrales et al. (1994), which analyzed populations with a Colombian isolate (race 521) and found two dominant genes. Young and Kelly (1997) and Young et al. (1998) predicted a third gene but did not define the interaction of this gene with Co-5 and Co-42 while highlighting the importance of two dominant genes. Pathania et al. (2006) also observed a digenic ratio when evaluating populations derived from G2333 for resistance to the Indian race 515 and predicted that the two independent dominant genes each had equal effects.

Breeding Implications
After assessing the overall effectiveness of marker-assisted selection within each BC1F1:2 family as described above, we evaluated the use of the markers across the full set of 266 individuals from this portion of the breeding program. When performed as a joint analysis, the data obtained from the phenotypic evaluation and the amplification of the SCARs established that of the 138 plants that amplified with SAB3 127 were resistant, showing that there was a significant relationship between resistance and the marker (X2 = 81.14, P = .00); however, amplification of the band in 11 susceptible materials can be attributed to a recombination event between SAB3 and the gene of interest, as the linkage between the two was reported to be as loose as 12.98 cM (Vallejo and Kelly, 2001). Finally, of the 124 plants that amplified the SAS13 marker, 32 showed susceptibility. This cannot be explained as a result of recombination because of the tight linkage between the marker and the Co-4 gene as reported by Young et al. (1998) and Melotto et al. (2004), but rather were the result of false negatives that can occur with SAS13. It appears that SAS13 is hampered by poor amplification and multiple banding due to its targeting of sequences within a serine threonine kinase gene that is found in multiple copies at the Co-4 locus evaluated by Melotto et al. (2004). Ideally, new markers should be developed that would resolve these problems of loose linkage in the case of SAB3 and imperfect repeatability of SAS13, however both markers are useful starting points for marker-assisted selection. Furthermore, since SAS13 does not discriminate between Co-4 and Co-42 alleles, it will be useful to evaluate the effectiveness of more loosely linked markers such as SAH18 and SBB14 that are more informative due to their specificity for the G2333 allele (Awale and Kelly, 2001). However, in the meantime, SAS13 has been used to successfully introduce Co-42 into highly susceptible pinto bean lines through marker-assisted backcrossing in the absence of pathogen screening (Miklas and Kelly 2002). Alzate-Marin et al. (2004) used RAPDs from which the SCARs SAB3 and SAS13 were derived (OPAS13 linked to the Co-4 gene and OPABO3 linked to the Co-5 gene) to develop resistant common-bean lines of the "carioca" grain type. Although marker-assisted selection can lead to false positives, it can be considered to be superior to classical phenotypic selection for anthracnose resistance because it is insensitive to pathogen variability or escape and can be combined with agronomic evaluation at almost any plant growth stage. It can also be applied at very early stages of the plant's life cycle to discard plants or rows that do not contain the desired gene.

In conclusion, our results bear out the utility of G2333 as an effective resistance source in breeding programs to develop varieties against anthracnose for the Andean region, and in laying out a clear strategy for marker-assisted selection of the Co-42 and Co-5 genes contained in this valuable genotype. Recent results from Ernest and Kelly (2004) question the strength of Co-42 resistance in certain Andean genetic backgrounds, but this does not seem to have held true for the climbing-bean crosses that we analyzed here. The possibility of complementary genes present in certain Andean or Mesoamerican germplasm but absent in other Andean landrace genotypes could explain the conflicting results. Despite these shortcomings for Co-42, the fact that G2333 also possesses the Co-5 gene makes it of interest for more widespread use and evaluation, especially since we found marker-assisted selection particularly effective in selecting for this gene. Furthermore, G2333 is a useful parent in climbing-bean crosses not only for its anthracnose resistance but also because of its high yield potential and vigorous growth habit. Significantly, we found that even in a single backcross of G2333 to the genotypes used in this study we could produce useful commercial segregants.


   ACKNOWLEDGMENTS
 
This research was part of an MSc program at the Univ. Nacional de Colombia for the first author and was financed by funds from Colciencias, SENA and the International Center for Tropical Agriculture (CIAT). The authors would like to thank Teresa Mosquera, María Isabel Chacòn, and Oscar Oliveros for helpful comments, George Mahuku and Linda Rincon for advice on inoculum preparation, Alcides Hincapie for greenhouse activities, and Hector Fabio Buendia for laboratory assistance.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication August 19, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Garzón, L. N.
Right arrow Articles by Blair, M. W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Garzón, L. N.
Right arrow Articles by Blair, M. W.
Agricola
Right arrow Articles by Garzón, L. N.
Right arrow Articles by Blair, M. W.
Related Collections
Right arrow Other Legumes
Right arrow Marker-assisted selection

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome