Crop Science Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 January 2005
Published in Crop Sci 45:379-387 (2005)
© 2005 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 Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Frei, A.
Right arrow Articles by Dorn, S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Frei, A.
Right arrow Articles by Dorn, S.
Agricola
Right arrow Articles by Frei, A.
Right arrow Articles by Dorn, S.
Related Collections
Right arrow Other Legumes
Right arrow Insect Resistance

QTL Mapping of Resistance to Thrips palmi Karny in Common Bean

A. Freia,b, M. W. Blaira,*, C. Cardonaa, S. E. Beebea, H. Gub and S. Dornb

a Centro Internacional de Agricultura Tropical, CIAT, AA 6713, Cali, Colombia
b Institute of Plant Sciences/Applied Entomology, Swiss Federal Institute of Technology (ETH), CH-8092, Zurich, Switzerland

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


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Host plant resistance in common bean (Phaseolus vulgaris L.) is a promising component in an integrated cropping system for managing thrips (Thrips palmi Karny) infestation. This study was conducted to identify quantitative trait loci (QTLs) for resistance to thrips in common bean, using F5:7 recombinant inbred lines (RILs) as a mapping population. The RILs, derived by single seed descent (SSD) of the cross of two Mesoamerican bean lines, BAT 881 and G 21212, were found to show transgressive segregation for thrips resistance in the field. Correlations between damage and reproductive adaptation (RA) scores were significant within and between seasons. The QTLs for both traits were located based on single interval mapping (IM) and joint interval mapping (JIM) analysis using a genetic map constructed with microsatellite and random amplified polymorphic DNA (RAPD) markers. Eight of eleven resulting linkage groups (LGs) were shown to be homologous to chromosomes of the integrated linkage map of common bean. A major QTL for thrips resistance located on LG b06 explained up to 26.8% of variance for resistance in a single season and was named Tpr6.1. The JIM across several seasons revealed various QTLs on LGs b02, b03, b06, and b08, some of which were located at regions of genes encoding for disease resistance. The identification and mapping of thrips-resistance genes is expected to facilitate the development of resistant bean cultivars by using molecular marker-assisted selection.

Abbreviations: BCMV, Bean common mosaic virus • IM, interval mapping • JIM, joint interval mapping • LG, linkage group • LR, likelihood ratio • PCR, polymerase chain reaction • QTL, quantitative trait locus • RA, reproductive adaptation • RAPD, random amplified polymorphic DNA • RIL, recombinant inbred line • SSD, single seed descent


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE COMMON BEAN was domesticated from a wild-growing vine distributed in the highlands of tropical America and has become a major legume food crop grown worldwide in a broad range of environments and cropping systems (Gepts and Debouck, 1991). Common bean consists of two major gene pools, the Mesoamerican or small-seeded bean type, and the Andean or large-seeded type (Debouck, 1999). Wide adaptation, good consumer acceptance, and high protein content make common bean the most important food legume in developing countries, especially in Latin America, which is the main bean producer and consumer in the tropical world. Several genetic maps have been constructed for this crop, including a core linkage map integrated by RFLP and microsatellite markers (Freyre et al., 1998; Blair et al., 2003).

The melon thrips, T. palmi, a pest species native to Sumatra (Karny, 1925), has recently been found in several American countries and regions, including Florida, Cuba, Puerto Rico, Venezuela, Brazil, and Colombia (Cardona et al., 2002). This pest can cause severe damage to plants not only by the feeding of adults and larvae, but also due to their capacity to vector virus diseases. Since its presence in Colombia was confirmed in 1997, T. palmi has become a serious pest of legumes (mainly dry bean and snap bean), solanaceous crops [potato (Solanum tuberosum L.), pepper (Capsicum annuum L. var. annuum), eggplant (Solanum melongena L.), tobacco (Nicotiana tabacum L.)], and cucurbits [melon (Cucumis melo L. subsp. melo), watermelon (Citrullus lanatus var. lanatus), cucumber (Cucumis sativus var. sativus), pumpkin (Cucurbita spp.)] (Cardona et al., 2002). Estimated yield losses from thrips in snap bean, for example, average 30%, and the entire harvest can be lost in susceptible bean genotypes (Rendón et al., 2001). Damage is visible on leaf ribs, foliage appearance becomes silvery, and reproductive organs are deformed and reduced in size and number (Frei et al., 2003, 2004). Management of thrips is challenging because of its short life cycle and high reproductive rate. It is not surprising that repeated insecticide applications are necessary for effective control of the pest. This management practice, however, has led to high levels of resistance to conventional insecticides in the insect (Lewis, 1997; Bueno and Cardona, 2003).

Host plant resistance is considered a promising component of an integrated crop management system, especially for an insect such as T. palmi, though studies on crop resistance to this species remain scarce (Parrella and Lewis, 1997). In a recently initiated program, 1138 dry bean genotypes were screened for thrips resistance, and 60 of the genotypes displayed a certain degree of resistance. This program included germplasm accessions, commercial genotypes, elite breeding lines, and RILs from the populations BAT 881 x G 21212 and DOR 364 x BAT 477 (Cardona et al., 2002).

The current study aimed to identify QTLs responsible for thrips resistance in the full population of RILs from the cross BAT 881 x G 21212. Genetic mapping and single IM and JIM QTL analysis was used to identify genetic markers linked to QTLs that would be useful in molecular marker-assisted selection and in genomic positioning of the significant thrips-resistance loci. Although molecular markers have been used to elucidate the genetic factors underlying disease resistance in common bean (Kelly et al., 2003), this study represents one of the first attempts to scrutinize QTLs for insect resistance for this important crop.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plant Material
A population of 94 RILs derived from the cross BAT 881 x G 21212 was used in field experiments to assess resistance to T. palmi. These RILs were originally bred for soil fertility experiments and were developed by SSD from the F2 to the F5 generation, followed by seed advance to the F7 generation. BAT 881 is an advanced breeding line with small light brown seeds with dark brown hilum rings derived from accessions originating from Guatemala, Mexico, and Venezuela. Its pedigree is (G 3834/G 2045//G 3627/G 5481). All of these accessions are landraces belonging to the Mesoamerican gene pool, except for G 5481, which is from Mexico but belongs to the Andean gene pool as defined by Debouck (1999). More information on these genotypes is available from the genetic resource unit at CIAT (www.ciat.cgiar.org; verified 11 Aug. 2004). BAT 881 is resistant to Bean common mosaic virus (BCMV) and intermediately resistant to the leafhopper Empoasca kraemeri Ross and Moore, but susceptible to several other diseases. G 21212 is a Mesoamerican landrace of Colombian origin (El Tambo, Nariño, 1360 masl) with small black seeds. It shows tolerance to low soil fertility and drought, but it is susceptible to BCMV (CIAT, 2002, unpublished data).

Field Trials and Thrips Resistance Phenotyping
Field trials for thrips resistance were conducted in three growing seasons: 1999A (planted on 15 April), 1999B (planted on 19 July), and 2000B (planted on 19 October) in Pradera, Valle del Cauca, a region with intensive commercial vegetable production in Colombia. This region is located at an elevation of 987 m, has a mean annual temperature of 24°C and relative humidity of 80%, and has abundant endemic thrips populations. The field trials were composed of a first preliminary trial (1999A) with 94 unreplicated RILs, a second trial (1999B) with 94 RILs in three replicates, and a third trial (2000B) with 48 selected intermediately and highly thrips-resistant RILs in three replicates. In each of these trials, the RILs were planted in randomized complete blocks, along with their parents and the susceptible control PVA 773 (‘ICA Caucaya’), a common commercial variety in Colombia, belonging to the Andean gene pool. Within each plot, each bean genotype was planted in a single 2-m row, with 50 cm between rows and 10 cm between plants. Natural thrips infestation was enhanced by planting next to highly infested fields of snap bean and by controlling spontaneous infestation of the greenhouse whitefly Trialeurodes vaporariorum (Westwood) with buprofezin [2-(tert-butylimino)-3-isopropyl-5-phenylperhydro-1,3,5-thidiazin-4-one; Oportune 25SC (as labeled in Colombia) or Applaud WP (as known elsewhere), Aventis Crop Science, Bogotá, Colombia], a selective insecticide that does not affect thrips (Frei et al., 2003). As in previous studies (Cardona et al., 2002), thrips resistance in bean plants was assessed by measuring (a) visual damage and (b) RA scores based on 1-to-9 nominal ranking scales. Damage was scored as: (1) no visible damage, (3) initial damage to leaf ribs, (5) obvious damage to leaf ribs and leaf deformation, (7) heavy damage along leaf ribs, heavy leaf deformation, silvery appearance of the foliage and stunting of the plant, and (9) severe damage, all leaves deformed, buds killed and with defoliation. Genotypes were classified as resistant (1–5), intermediate (5.1–7), or susceptible (>7). Reproductive adaptation, a visual estimate of pod and seed set under high insect pressure, was scored as (1) no pod setting or very few, deformed, and empty pods, 80% or more yield reduction, (3) many pods are small, some are empty and deformed, 60% yield reduction, (5) some pods are small, a few are deformed and empty, 40% yield reduction, (7) most pods look normal, very few are deformed and empty, 20% yield reduction, and (9) all pods look normal, normal pod setting. Genotypes were classified as susceptible (1–3), intermediate (3.1–5), or resistant (>5). Damage was determined at the V6 growth stage, approximately 50 d after planting, while RA was determined at the R9 growth stage, approximately 70 d after planting in each trial (Fernandez and Gepts, 1984). Additionally, thrips infestation was assessed counting numbers of adult thrips per leaflet at 40 d after planting.

DNA Isolation and Marker Analysis
Genomic DNA was isolated from unexpanded cotyledonary leaves of three bean seedlings from each parent and each RIL line according to the protocol of Afanador and Haley (1993) with a modified CTAB extraction procedure (extraction buffer: 1.5% CTAB, 1.0 M NaCl, 15 mM EDTA, 150 mM Tris-HCl, 2% β-mercaptoethanol). DNA quality was checked in a 0.8% agarose gel and concentration was measured with a fluorometer (DyNA Quant 200, Hoefer Scientific Instruments, San Francisco, CA). Template DNA was diluted to a final concentration of 10 ng µL–1 for use in the polymerase chain reaction (PCR). In a preliminary study, a total of 466 RAPD primers were evaluated to screen polymorphism in the parents and in the 94 RIL lines (CIAT, 2000). The PCR amplification consisted of 38 cycles of 91°C for 15 sec, 42°C for 15 sec, and 72°C for 1 min, and PCR products were separated by electrophoresis in 1.5% agarose gels using 0.5 x TBE buffer. Microsatellite analysis was done with 108 markers developed by Yu et al. (2000), Gaitán-Solis et al. (2002), and Blair et al. (2003); also, CIAT, 2003, unpublished data. In this case, PCR amplification was done according to Blair et al. (2003), with some modifications (final volume of 12.5 µL, with 25 ng of bean genomic DNA, 200 mM of total dNTP, annealing temperature based on specific primer Tm) and PCR products were mixed with loading dye, denatured, and separated by electrophoresis on polyacrylamide gels according to the procedure of Gaitán-Solis et al. (2002), with some modifications (4% gels, 29:1 acrylamide to bisacrylamide, 7 M urea). Gels were silver stained according to the manufacturer's instructions. The microsatellites were evaluated using parental DNA controls and 10 and 25 bp size standards (Promega, Madison, WI) for allele confirmation.

Linkage Analysis
Standard chi-square ({chi}2) tests (threshold P < 0.01) were employed to test for segregation distortion at each marker locus. A linkage map was constructed with all RAPD and microsatellite markers. The software package MAPMAKER 3.0 (Lander et al., 1987) was used to generate the genetic map, whereby genetic mapping was first done by grouping markers at LOD > 5.0 and then by ordering them at LOD > 3.0, using three-point analysis with a maximum intermarker distance of 37.2 cM. Microsatellite markers were anchored according to the integrated microsatellite map for common bean prepared by Blair et al. (2003). After determination of LGs and correct linear arrangement of marker loci, recombination frequencies between markers were estimated and transformed to centimorgan distances using the Kosambi mapping function. Linkage groups that were assigned to the 11 chromosomes of common bean were numbered and oriented, as in the integrated map by Freyre et al. (1998), while LGs, which did not contain mapped microsatellites and were therefore unidentified, were numbered consecutively for use in the QTL analysis.

Data Analysis
Means, standard errors, skewing, kurtosis, and Pearson's correlations were calculated, and the ANOVA was conducted with field data from each season using the software programs SAS (SAS Institute, 1989) and Statistix (Analytical Software, 2000). Broad-sense heritabilities (h2) within and across seasons were calculated for both resistance traits based on mean square ANOVA results, genotypic variance Formula, phenotypic variance Formula, genotype x season interactions Formula, and error variance Formula. All of these variances were calculated on an entry-mean basis. Arbitrary linear (orthogonal) contrasts were conducted using the Scheffe's F test to compare among means of selected genotypes within season and within trait for: (i) the five most resistant RILs, (ii) the five most susceptible RILs, (iii) the BAT 881 parent, (iv) the G 21212 parent, and (v) the susceptible PVA773 check (Statistix, Analytical software, 2000). The QTL analyses were performed with mean RIL scores from each season and the genetic map was constructed, using IM and JIM analysis with the software package QTLCartographer V2.0 (Basten et al., 2001). In the IM as well as the JIM analysis, LOD threshold were derived from 1000-fold permutations. Resulting LOD thresholds lay between 2.6 and 2.8 (likelihood ratio, LR, of 11.82–12.83), depending on season and trait. For JIM, the threshold was equal to the highest threshold found among the traits that were analyzed together. A window size of 10 cM and a 1 cM walking step were used to determine the presence and location of QTLs and whether there was evidence for more than one QTL on LGs with multiple LOD peaks.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Thrips Resistance Phenotyping
Significant differences were found among genotypes for thrips resistance as measured by damage and RA scores in both the 1999B and 2000B seasons (Table 1, Fig. 1) . Most of the differences were among RILs, with parents not significantly different in either season. For damage and RA scores, the coefficient of variation ranged from 11.1 to 20.4% for RILs and from 5.3 to 24.5% for parents across the two seasons. The combined ANOVA for the data during both seasons showed significant differences in damage among replicates, seasons, and genotypes, while in RA only among replicates and genotypes (Table 2). There was no significant genotype x season interaction for either damage (F = 1.11; df = 47, 47; P = ns) or RA score (F = 1.07; df = 47, 47; P = ns). Broad-sense heritabilities based on mean squares calculated for each season separately were higher for damage scores than for RA scores in both seasons. Similarly, heritability estimates based on the combined analysis of variance for the 1999B and 2000B trials were higher for damage (63.4%) than for RA (43.6%) (Table 3).


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

  		Table 1. Analysis of variance and CV for Thrips palmi damage and reproductive adaptation (RA) scores (both measured on 1–9 scales) conducted in two seasons in Pradera, Colombia (1999B and 2000B).

 

Figure 1
View larger version (54K):
[in this window]
[in a new window]

  		Fig. 1. Frequency histograms of damage and reproductive adaptation scores of recombinant inbred lines (RILs) tested for resistance to Thrips palmi in three seasons (1999A, 1999B, and 2000B). Repetitions: 1, 3, 3; number of lines: 94, 94, 48, respectively. The overall mean is indicated by a solid line. Parents BAT 881 (B) and G 21212 (G) are indicated by black triangles; the five most susceptible lines (5S), and five most resistant lines (5R) are indicated by white triangles. Resistant, intermediate, and susceptible intervals are indicated at the bottom of the figure. Means and standard errors are given in Table 3.

 

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

  		Table 3. Estimates of entry-mean basis variance components and heritabilities for RILs grown in two seasons in Pradera, Colombia (1999B, 2000B) for Thrips palmi damage and reproductive adaptation (RA) scores (both measured on 1–9 scales).

 

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

  		Table 2. Combined analysis of variance and CV for Thrips palmi damage and reproductive adaptation (RA) scores (both measured on 1–9 scales) for recombinant inbred lines (RILs) grown in two seasons in Pradera, Colombia (1999B, 2000B).

 
The analysis of orthogonal contrasts showed that the five most resistant RILs compared with the parents in each season had significantly lower damage scores in all but one case and significantly higher RA scores (Table 4). Similarly, the resistance scores for the five most susceptible RILs were significantly different from their parents in most cases, that is, they showed higher damage scores and lower RA scores than their parents. In the parental comparisons, BAT 881 had higher mean damage scores and lower mean RA values than G 21212 in two of the three seasons, though these differences were not statistically significant (Fig. 1). The damage scores of the five most resistant RILs in the three trial seasons ranged from 3.8 ± 0.1 to 5.5 ± 0.3 (mean values ± SE), as compared with the scores of the five most susceptible RILs that ranged from 7.3 ± 0.9 to 8.7 ± 0.0. The RA scores ranged from 4.5 ± 0.1 to 6.6 ± 0.1 for the five most resistant RILs and from 1.0 ± 0.0 to 1.9 ± 0.1 for the five most susceptible RILs. The ranges of both resistance traits for all RILs changed between seasons, being larger in the first than in the two subsequent seasons (Fig. 1). Thrips infestation, as measured by numbers of adult thrips per leaflet at 40 d after planting, was lower in the first season (3.5 in 1999A) than in the subsequent seasons (7.2 in 1999B; 6.6 in 2000B). In the second season (1999B), where the highest infestation levels were recorded, the distributions were skewed toward high damage (skewness = –0.3271, P < 0.05; kurtosis = –0.0477, P = ns), and lower RA (skewness = 0.3052, P < 0.05; kurtosis = 0.7621, P < 0.01), while in the other two seasons kurtosis and skewness were not significant. Despite seasonal variability in the level of thrips infestation and the resistance response, these RILs exhibited a greater range in both damage and RA scores compared with the parents (Table 4), indicating possible transgressive segregation.


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

  		Table 4. Means of parents (BAT 881 and G 21212), recombinant inbred lines (all RILs, five most resistant, five most susceptible), and susceptible check (PVA 773) tested for Thrips palmi damage and reproductive adaptation (RA) scores (both measured on 1–9 scales) for experiments conducted in three seasons (1999A, 1999B, 2000B) in Pradera, Colombia.

 
Damage and RA scores correlated significantly and negatively with each other within all trial seasons (Table 4). Significant positive correlations were found between 1999A and 1999B both for damage (r = 0.259, P < 0.05) and RA scores (r = 0.380, P < 0.001), and between 1999B and 2000B both for damage (r = 0.497, P < 0.001) and RA scores (r = 0.300, P < 0.05), though correlations between 1999A and 2000B were not significant for either damage or RA.

Genotyping
Polymorphic microsatellite markers were amplified on the whole RIL population, and mapping and QTL analysis were performed to determine linkage and association of specific markers with resistance phenotypes. Of 108 microsatellite markers tested in total for parental polymorphism, 33 were polymorphic, 70 were monomorphic, and five did not amplify. The resulting rate of parental polymorphism was 30.5%, which is slightly higher than for other Mesoamerican x Mesoamerican populations (CIAT, 2003, unpublished data). Linkage analysis with the RAPD markers alone identified 15 LGs, while genetic mapping with the selected microsatellites coalesced these into eight LGs (b01, b02, b03, b04, b05, b06, b08, and b09) that were homologous to the chromosomes of the integrated map of Freyre et al. (1998). Microsatellite marker order was colinear with the previous published maps that contain this marker type (Blair et al., 2003). A total of 115 markers (26 microsatellites and 89 RAPDs) were placed on the eight identified LGs. The four groups containing QTLs are shown in Fig. 2 . The average size of the identified LGs was 65.5 cM. The average number of markers identified per LG was 13.1. Three LGs were made up entirely of RAPDs and could not be assigned to the known chromosomes. The total coverage of the genetic map was 611.2 cM, with an average interval length of 4.3 cM between markers. Marker intervals > 20 cM were detected on LGs b05, b06, and b08. A total of 35 RAPDs and seven microsatellite markers remained unlinked to the LGs described above.


Figure 2
View larger version (44K):
[in this window]
[in a new window]

  		Fig. 2. Genetic linkage map for the BAT 881 x G 21212 population, showing linkage groups that contained significant QTLs for Thrips palmi resistance. The QTLs were identified by single interval mapping (black boxes, D99B = damage 1999B, RA99B = reproductive adaptation 1999B), by joint interval mapping for resistance traits within the same season (gray boxes, J99B = joint for damage and RA in 1999B) and by joint interval mapping for each resistance trait across three seasons (white boxes, JD3s = joint for damage across three seasons, JRA3s = joint for RA across three seasons). The LOD peaks are indicated by black horizontal bars, and their values are shown below the QTL box. Linkage group identification is according to Freyre et al. (1998) and Blair et al. (2003). The cumulative distance, in centimorgans, is shown relative to the topmost marker. Marker loci with a significant distorted segregation are marked by a black triangle.

 
Significant segregation distortion (P < 0.01) occurred in 25 of the 151 RAPD (16.6%) and two of the 33 microsatellite marker loci (6.1%). These loci are located on the ends of LGs b01 (1 marker), b03 (1 marker), b04 (1 marker), and b05 (2 markers), as well as on the lower half of LG b06 (7 markers). Segregation distortion in the present map was less frequent than in other common bean maps (CIAT, 2003, unpublished data). The frequency of the G 21212 allele was higher than that of the BAT 881 allele in 59.2% of the loci.

QTL Analysis for Thrips Resistance Traits
In the IM analysis, LOD thresholds of 2.6 to 2.8 (LR 11.82–12.83) were calculated based on the result of a 1000-fold permutation test for each trait analyzed and were used to declare a putative QTL as significant. One major thrips-resistance QTL was identified on LG b06 for both damage and RA in the season 1999B (Fig. 2 and 3) . This QTL was named Tpr6.1, and the highest LOD at the terminal marker BMc128 was 5.8 (LR 27.2) for damage in 1999B and 4.6 (LR 21.6) for RA in 1999B. This QTL explained 26.8% of variance for damage in 1999B (total R2 of 0.278), and 21.9% of variance for RA in 1999B (total R2 of 0.230). The Tpr6.1 QTL was also associated with RA and damage in 2000B, but at LOD values slightly lower (<2.8 and <2.1, respectively) than the calculated threshold for this season (Fig. 3).


Figure 3
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 3. Interval mapping of a major QTL on linkage group b06 found to condition resistance to Thrips palmi in the population BAT 881 x G 21212, in the seasons 1999A, 1999B, and 2000B, using microsatellite and RAPD markers. Resistance traits are explained by letters close to the corresponding curve. Microsatellite markers are indicated in bold and the significance threshold is indicated by a line at LOD = 2.8.

 
As the correlation between seasons was significant, JIM analysis was conducted and revealed QTLs on LGs b02, b03, and b08 (Fig. 2). Joint analysis for damage and RA within the same season confirmed the QTL for season 1999B on LG b06 in the BMc128-O1505 interval and identified a peak LOD (6.3) that was also found at the terminal end of this interval. Meanwhile, joint analysis for both damage and RA scores analyzed jointly across three seasons also revealed a QTL for both damage and RA at this site (peak LOD of 3.0 and 3.7, respectively). Apart from the QTL on LG b06, additional QTLs were detected with the joint analysis of two traits analyzed from 1999B revealing two QTLs on LG b03 (Fig. 2), one between the markers AL401 and AE 101 (peak LOD of 2.8), and one between the marker E402 and BMc 179 (peak LOD of 3.0). Joint analysis for damage and RA scores across three seasons also revealed two less significant, minor QTLs on b02 for each damage (LOD of 3.1) and RA (LOD of 2.9), at the end of the LG, linked to the microsatellite marker BM143. On LG b08, one QTL was identified at the Q1801 marker, for damage across three seasons (LOD of 3.2).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Thrips resistance was identified in the BAT 881 x G 21212 population that was heritable and consistent across seasons despite variability in thrips infestation in the first season compared with other seasons. This was confirmed by the lack of significant genotype x season interactions in the two seasons evaluated in replicated trials and suggested that the resistance reaction of each genotype was consistent regardless of seasonal variation in the infestation level, climate, or growing conditions. However, we do recognize that such environmental variables may have contributed to seasonal differences in the broad-sense heritability estimated for the two traits evaluated. Variability was also observed within seasons as shown by the significant effects of replication. It was notable, however, that heritability levels were low to moderate (32.4 to 61.9%), indicating that at least in some seasons, natural infestation is a useful screening technique. Aggregated distribution of the insect has been observed previously (Kirk, 1997) and may explain some of the variability.

Correlations were always high between damage and RA scores of the same season while between-season correlations were higher for damage than for RA. It was also notable that the heritability estimates were consistently higher for damage score (52.6–61.9%) than for RA score (32.4–47.8%). Both of these results suggest that damage is a more consistent indicator of thrips resistance than RA score. Significant correlations between damage and RA within and across seasons have been observed before (Cardona et al., 2002), and the two indicators have been used somewhat interchangeably. However, our results show that the earlier evaluation of damage at the V6 growth stage is better than the evaluation of RA, which can only be done at the R8 (pod fill) growth stage.

Field results further demonstrated that the RILs exhibited a greater range of phenotypic variation in either resistance trait than their parents, which were intermediately resistant. This indicates transgressive segregation of thrips resistance. In particular, in the 1999B trial, the distribution of damage and RA scores was significantly skewed toward susceptibility, probably because of the higher thrips infestation in this season compared with the other two seasons. The analysis of orthogonal contrasts comparing the five most resistant lines with the five most susceptible lines revealed significant differences between these groups in comparison with their parents and the susceptible control. This also suggests transgressive segregation.

The QTL analysis revealed that the inheritance of thrips resistance was relatively simple. One major QTL named Tpr6.1 was found with interval analysis, while joint interval analysis confirmed this QTL and revealed additionally only three minor QTLs on other LGs. The allele associated with resistance at the major QTL originated from the BAT 881 parent and the locus was located at the end of LG b06 linked to the microsatellite markers BMc128 and BM218. This QTL was consistent for damage and RA in a single season (1999B). Compared with a previously determined map for resistance genes in common bean (Kelly et al., 2003), the major thrips-resistance QTL was found in a region where a BCMV resistance gene (bc-3) and rust resistance gene (Ur-4) are also found.

Meanwhile, the three minor QTLs identified by joint interval analysis affected one or both traits during the three seasons and were located on LGs b02, b03, and b08. The alleles contributing to resistance at these minor QTLs were derived from both parents, BAT 881 and G 21212, which may explain the occurrence of transgressive segregation as discussed above. Two of these QTLs were located in the areas of genes associated with disease response that have been mapped in common bean (Kelly et al., 2003). Specifically, the thrips-resistance QTL on LG b03 was linked to a gene for a pathogen related protein (PvPR-1) and to several genes for glycine-rich proteins (GRP1.8–1 and GRP1.8–2). Glycine-rich proteins are localized in the cell walls of vascular tissues of bean (Keller et al., 1988) and dicot glycine-rich proteins are known to be expressed in response to a variety of environmental stresses, such as viral infection, drought stress, and wounding (Showalter, 1993; Harrak et al., 1999). Plant response to drought stress is manifested by various changes in physiological and metabolic processes, including drying, browning, and leaf loss; some of these processes are similar to severe thrips-feeding symptoms. It can be speculated that a thrips attack might trigger the expression of such genes in a way similar to wounding or drought stress. The other minor QTLs were detected in regions possibly close to race-specific disease resistance genes or resistance gene clusters on LG b08 (Kelly et al., 2003).

Although plant resistance in bean is known for leafhoppers, bruchids [Zabrotes subfasciatus (Say) and Acanthoscelides obtectus (Boheman)], and pod borers (Apion godmani Wagner) (Cardona and Kornegay, 1999; Schmale et al., 2003), few gene tagging studies for insect resistance have been performed so far on bean compared with other crops (Yencho et al., 2000). This is in contrast to the fact that several QTLs are known in reference to stress tolerance and many have been determined for disease resistance in bean (Kelly et al., 2003). It will therefore be interesting if QTLs for thrips resistance are associated with previously studied abiotic or biotic resistance traits. In a few cases, disease resistance gene clusters have been shown to control insect resistance, as in the noted example of the Mi/Meu-1 gene in tomato (Lycopersicon esculentum Mill.) for resistance to both nematodes and potato aphids controlled by an LRR-type (leucine-rich repeat) resistance gene analog (Rossi et al., 1998). Further study will be needed to determine if thrips resistance is controlled in bean by resistance gene analogs or by some other class of genes.

Major, monogenic resistance genes are attractive to the breeder because they are easy to manipulate, and can be rapidly introgressed into susceptible materials through simple backcrossing (Kelly and Miklas, 1999). Nonspecific, polygenic resistance would be more durable, but its deployment creates a major challenge for the breeder since epistasis and environmental variability often mask this type of resistance. The disadvantage of major genes is that the resultant resistance can easily be overcome by new, virulent insect biotypes (Yencho et al., 2000). These race-specific genes are recognized as the least durable source of genetic resistance to highly variable plant pathogens or insects. Although major resistance genes have short lifetimes when used one at a time, the opportunity to pyramid genes [using marker assisted selection (MAS)] will make major gene resistance more useful than at present (Duvick, 1996).

To date, various types of thrips resistance, such as antixenosis, antibiosis, and tolerance, or their combinations, have been found in common bean germplasm (Cardona et al., 2002; Frei et al., 2003) so it will be interesting to determine which of these mechanisms underlies the QTL found in this study. Furthermore, most determinate, large-seeded Andean genotypes are generally susceptible to the infestation of T. palmi, and all resistant genotypes have been found in the Mesoamerican gene pool so far (Cardona et al., 2002). Thus, it would be interesting to test whether and how QTLs for thrips resistance found in the two Mesoamerican genotypes studied here can be introgressed into the Andean gene pool. For this it may be necessary to develop additional RIL populations that can confirm the QTL identified in this study. Finally, it would be interesting to analyze whether thrips resistance QTLs affect other insect pests of common bean. The identification and mapping of the thrips-resistance genes and QTLs should facilitate the development of resistant bean cultivars and the pyramiding of thrips resistance with other traits, particularly through the use of marker-assisted selection.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge J.M. Bueno, H.F. Buendia, O. Checa, and Dr. Yvan Fracheboud, who helped with field trials and/or data analysis, and M. Muñoz and A. Velasco, who provided technical assistance with this study. This work was supported by a grant of the Swiss Center for International Agriculture (ZIL) to SD and HG. USAID funds to MB supplied marker development.

Received for publication April 13, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 


Related articles in Crop Science:

THIS ISSUE IN CROP SCIENCE

Crop Science 2005 45: xi. [Full Text]  



This article has been cited by other articles:


Home page
 Crop Sci.Home page
J. J. Maxwell, M. A. Brick, P. F. Byrne, H. F. Schwartz, X. Shan, J. B. Ogg, and R. A. Hensen
Quantitative Trait Loci Linked to White Mold Resistance in Common Bean
Crop Sci., November 7, 2007; 47(6): 2285 - 2294.
[Abstract] [Full Text] [PDF]

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 Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Frei, A.
Right arrow Articles by Dorn, S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Frei, A.
Right arrow Articles by Dorn, S.
Agricola
Right arrow Articles by Frei, A.
Right arrow Articles by Dorn, S.
Related Collections
Right arrow Other Legumes
Right arrow Insect Resistance

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