| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Institute for Agronomy and Plant Breeding, Christian-Albrechts-Universität Kiel, Am Botanischen Garten 1-9, D-24118 Kiel, Germany
b Dep. of Plant Pathology, Univ. of Wisconsin-Madison, Madison, WI 53706-1598
c Dep. of Entomology, Univ. of Wisconsin-Madison, Madison, WI 53706-1598
d Dep. of Crop Sciences, Univ. of Illinois, Urbana, IL 61801
* Corresponding author (f.kopisch{at}plantbreeding.uni-kiel.de).
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: AMV, Alfalfa mosaic virus • ELISA, enzyme-linked immunosorbent assay • LG, linkage group • OD, optical density • PBS, phosphate-buffered saline • QTL, quantitative trait locus • SSR, simple sequence repeat
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Symptoms caused by AMV on soybean may appear as mosaic or mottle patterns of contrasting mixes of vivid yellow and dark green leaf tissue (Clark and Perry, 2002; Tolin, 1999), however, AMV-infected plants are frequently asymptomatic (Lee-Burrows et al., 2005). Furthermore, soybean plants infected by AMV produce only minor amounts of seed with mottled seed coats (Mueller and Grau, 2007) which frequently is the more apparent symptom caused by other viruses. Yield loss estimates specific to AMV are 32 to 48% once the incidence of AMV-infected plants exceeds 30% in induced epidemics in Wisconsin (Mueller and Grau, 2007).
Resistance to AMV has been reported in several host species including members of Fabaceae such as alfalfa (Medicago sativa L.) (Crill et al., 1971), snap or common bean (Phaseolus vulgaris L.) (Wade and Zaumeyer, 1940), pea (Pisum sativum L.) (Ford and Baggett, 1965), and soybean (Almeida et al., 1982). Gene action of AMV resistance was identified as completely dominant in bean (Wade and Zaumeyer, 1940), button medic [Medicago orbicularis(L.) Bartalini] (Pathipanawat et al., 1996), and tomato (Solanum lycopersicum L.) (Parrella et al., 2004), while recessive gene action was observed in alfalfa (Crill et al., 1971). In soybean, resistance to AMV was reported in the two Brazilian soybean cultivars, Pérola and Planalto, and their common ancestor, Hood, a cultivar from the southern United States (Almeida et al., 1982). Segregation of AMV reaction fit a 1:3 symptomatic/asymptomatic ratio in three F2 populations derived from crossing Pérola to the susceptible Brazilian cultivar Dourados, and in one F2 population derived from crossing Pérola to ‘Coker 136’, which also is AMV susceptible. This suggests that resistance observed in Pérola is controlled by one dominant gene.
In an effort to find additional sources of AMV resistance in soybean, 400 soybean accessions were evaluated in a greenhouse for reaction to an AMV field isolate collected in Wisconsin (C. Grau, unpublished data, 2002). In this screening, plant introduction (PI) 153282 was the only accession that expressed resistance to AMV.
To investigate the genetic basis of AMV resistance in PI 153282, populations derived from crossing PI 153282 with the AMV susceptible cultivar Syngenta S19-90 were tested for resistance to AMV. The objectives of our research were (i) to estimate the number of genes underlying AMV resistance in soybean PI 153282, (ii) to study their gene action, and (iii) to locate these gene(s) on a genetic map.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To confirm results from this population in early generations, a second population of 220 F2 plants was developed from crossing the same parental genotypes. All F2 plants were tested for AMV resistance reactions in a greenhouse assay. Forty resistant and 40 susceptible F2 plants were selected based on enzyme-linked immunosorbent assay (ELISA) tests and grown to maturity to produce seed for F2:3 lines. These lines were subsequently evaluated for AMV resistance reactions.
Greenhouse Experiments
In the greenhouse assays, plants were grown in a 1:1 mixture of sterilized sandy soil to Metro-Mix 366-P growing medium (Scotts-Sierra Horticultural Products Co., Marysville, OH) in 10-cm2 plastic pots. Seeds were dusted with Bradyrhizobium japonicum (Nitragin Inoculants, Liphatech Inc., Milwaukee, WI) and fertility was supplemented with 5 mL of controlled release Osmocote Plus fertilizer (Scotts-Sierra Horticultural Products Co.). Plants were grown with a daily 16-h photoperiod and watered once a day.
The initial experiment with the F4 population was conducted as a randomized complete block design with two replications. All 174 F4:7 lines were evaluated in each replication. The first replication was planted on 19 Aug. 2004 and inoculated on 2 Sept. 2004. Each experimental unit consisted of three 10-cm2 pots each with three plants. The second replication was planted on 26 Apr. 2005 and inoculated on 6 May 2005. In this replication, the experimental unit size was increased to four pots each planted with three plants. In both replications the population was tested along with the parents and the susceptible control cultivars A2506 and Conrad 94.
The F2 population, plus parents, were evaluated for AMV reactions in a test planted on 4 Mar. 2005. Four plants were grown per 10-cm2 pot and pots were arranged in random order. Both inoculated and noninoculated plants of the parents PI 153282 and S1990 were included as check entries.
F2:3 lines derived from selected F2 plants were evaluated for AMV reactions along with PI 153282 and S19-90. Seeds were planted on 19 Aug. 2005 and plants were thinned to one plant per 10-cm2 pot on 29 Aug. 2005. Plants were arranged in randomized complete block design with five replications each containing one plant from every family along with PI 153282 and S1990. This experiment was inoculated with AMV as described below and noninoculated PI 153282 and S19-90 plants were included. As negative control, a sixth replication without AMV inoculation was added.
Virus Inoculation
The resistance assays were conducted with AMV-C, a field isolate recovered from plants collected in a commercial soybean field in Wisconsin during 2001. The isolate was maintained in source plants and mechanically transmitted in frequent succession to the susceptible soybean cultivar Colfax (Graef et al., 1994) to maintain high virus titer in the greenhouse. The Phaseolus vulgaris cultivar Bountiful was used as a local lesion host to assay for the presence of AMV and a determinant of the relative viral infectivity of inoculum sources before inoculation (Crill et al., 1970). For preparation of inoculum, infectious plant sap was prepared by grinding the youngest fully expanded trifoliate leaf from AMV-C source plants in 1 mL of phosphate-buffered saline (PBS; pH 7.2).
Plants were inoculated at the V1 stage (Fehr et al., 1971) by dusting each fully expanded unifoliate leaf with carborundum and rubbing both leaves gently with cotton swabs soaked in AMV-C sap. During inoculation, inoculum was frequently replaced due to instability and rapid degradation of AMV in buffer solution. Inoculated soybean plants were incubated for 10 to 14 d and rated for AMV reaction at the V3 growth stage (Fehr et al., 1971). When visual ratings were taken on the F2 plants and F2:3 lines, the youngest fully expanded trifoliate leaf was collected and evaluated with ELISA to test for the presence or absence of AMV.
AMV Symptom Rating
Individual plants from the populations of F2 plants, F2:3 lines, and F4:7 lines were rated for response to AMV based on visual symptoms. F2 plants and F4:7 lines were rated for symptom severity on a 0 to 3 scale by comparing inoculated plants to the noninoculated control plants with 0 = asymptomatic; 1 = slightly smaller leaves or shorter stature; 2 = combination of either shorter stature, smaller leaves, or slight mosaic discoloration of the leaves; 3 = high level of symptom expression of stunting, smaller leaves, and mosaic leaf discoloration. Premature plant death was observed occasionally and included in category 3. F2:3 lines were only rated for disease incidence with 0 = asymptomatic and 1 = foliar symptoms including chlorotic mottling or mosaic, which may or may not have been accompanied by plant stunting. The 0 and 1 score was used to rate the F2:3 lines because an initial analysis of data from previous tests suggested that the resistance was qualitative and because the visual ratings were supplemented with ELISA tests.
ELISA
AMV-C was detected in the F2 and F2:3 lines with an indirect ELISA using a polyclonal AMV antibody (Agdia Inc., Elkart, IN). From each plant, the youngest fully expanded trifoliate leaf was individually placed into a 7.6 by 7.6 cm grinding bag (Agdia Inc., Elkart, IN) approximately 14 d post inoculation. The leaves were ground in 1 mL PBS (pH = 7.2) and 100 µL of the solution was loaded onto a 96-well ELISA plate (Agdia Inc.) precoated with carbonate buffer (pH 9.6). Blank wells were filled with 100 µL carbonate buffer. Sample absorbance was measured by an EL800 Universal MicroPlate Reader (Biotek Instruments Inc., Winooski, VT) at a wavelength of 405 nm and expressed as optical density (OD). Absorbance readings equal to or above OD 0.1 were considered positive for AMV. Occasionally plants were ELISA negative, but exhibited typical symptoms caused by AMV. For these situations, the plant was phenotypically classified as susceptible. Background, as measured in buffer and negative control wells, was minimal for all ELISA assays. All reactions were performed at room temperature.
ELISA tests were used to confirm the reaction of asymptomatic plants to AMV. In the absence of symptoms, absorbance readings equal to or above OD 0.1 were considered positive for AMV. Presence of AMV symptoms were usually confirmed by absorbance readings of OD 0.1 and greater. In cases where plants were symptomatic for AMV but ELISA negative, plants were considered positive once the possibility of a second virus was eliminated. False negatives to AMV may be due to the variability inherent in ELISA assays, plant age, and plant tissue collected.
DNA Marker Analysis
In the F4:7 and F2:3 mapping populations, DNA was isolated from samples of 10 plants per line using a modified CTAB method (Kisha et al., 1997; Kabelka et al., 2006). The DNA from each line was genotyped with simple sequence repeat (SSR) markers developed by P.B. Cregan (USDA-ARS, Beltsville, MD) according to methods described in Cregan and Quigley (1997). Polymerase chain reaction products were analyzed in 3% metaphor (FMC BioProducts, Rockland, ME) agarose gels, 6% (w/v) nondenaturing polyacrylamide gels (Wang et al., 2003), or on an ABI Prism 377 DNA Sequencer (ABI-PEC, Foster City, CA) as outlined in Patzoldt et al. (2005). DNA from the F2 population was extracted using a quick extraction method described by Bell-Johnson et al. (1998).
Statistical Analysis
An analysis of variance (ANOVA) using PROC MIXED in SAS (SAS Institute, 2000) was calculated for the AMV reaction data collected in the replicated experiment for the F4:7 population. Means for each line were estimated with the LSMEANS statement. Segregation ratios for AMV response observed in all three populations were tested for deviation from expected segregation using a chi-square test. Spearman correlations between AMV symptom severity and ELISA were calculated in the F2 population using PROC CORR in SAS.
A linkage map was constructed with the marker data from the F4:7 population using Joinmap 3.0 (Van Ooijen and Voorrips, 2001). Quantitative trait loci controlling AMV leaf symptoms were mapped with PlabQTL 1.2 (Utz and Melchinger, 1996). Due to the advanced generation of the F4:7 lines, dominance effects were not included in the genetic model used for quantitative trait locus (QTL) detection. Support intervals for QTL positions were determined by LOD drop-offs of 1.0 as applied by Helgadottir et al. (2004). Experiment-wise LOD thresholds for the QTL analysis were determined empirically with 1000 permutations (Churchill and Doerge, 1994) for an experiment-wise alpha level of 
E = 0.10. Unlinked markers were analyzed with simple regression at an experiment-wise alpha level of E = 0.10 using PROC GLM in SAS.
For confirmation, QTL were mapped with PlabQTL in the F2 and F2:3 populations using only markers in proximity of previously mapped QTL. This confirmation mapping was done using the visual AMV severity as well as ELISA data.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
QTL Analysis
A major QTL for visual symptom severity was detected on LG J between Sat_228 and Satt_339 with a LOD score of 45.5 (Fig. 2
). The peak was 3 cM from Sat_228 and 6 cM from Sat_339. The allele from the resistant parent PI 153282 had an additive effect of –1.02 units for AMV symptom severity. This QTL explained 71% of the phenotypic variation (R2) or 79% of the genotypic variation. No significant association with AMV symptom severity was detected on any other linkage group nor among the 54 unlinked markers using simple regression (E = 0.10).
The combining of the segregation of Sat_228 with the distribution of the lines for visual ratings shows that underlying the QTL peak is a major gene that controls AMV resistance (Fig. 1). Almost all lines rated as resistant (visual symptom severity <1.8) were homozygous for the allele from PI 153282, the resistant parent. In addition, almost all lines classified as susceptible (visual symptom severity >2.6) were homozygous for the allele from S19-90, the susceptible parent. The few exceptions were mostly lines that were segregating for the gene containing region. There were three lines with susceptible ratings that were classified as homozygous for the resistance allele based on the marker. These lines may have resulted from recombination between the marker and gene during the line development.
F2 Population
Visual Symptoms
The distribution of AMV visual symptom severity in the F2 population (Fig. 3
) was strongly skewed toward low symptom severity. As expected, the susceptible parent S19-90 had the highest symptom severity of 3 while the resistant parent PI 153282 and both noninoculated parents showed no AMV symptoms. If plants with ratings of 0 and 1 are classified as resistant and plants with ratings of 2 and 3 as susceptible, the observed phenotypic segregation perfectly fit a 3:1 resistant (165 plants)/susceptible (55 plants) ratio as expected for monogenic resistance with complete dominance.
|
2 = 0.218, P = 0.64). This further supports our hypothesis that AMV resistance is a monogenic trait with complete dominance. The correlation between absorbance readings and visual symptom severity was highly significant (P < 0.001) but quite low with r = 0.29.
|
|
The marker genotypes of F2 plants for Sat_228 further support dominant gene action of the AMV resistance gene (Fig. 3). Over 90% of plants predicted to be homozygous or heterozygous for the AMV resistance gene based on Sat_228 were indeed classified as resistant (rating 0 or 1). However, the segregation of this marker was strongly distorted with 28 plants homozygous for the S19-90 allele, 57 heterozygous and 59 homozygous for the PI 153282 allele. This distortion is probably caused by missing data for 75 F2 plants due to using quick-extracted DNA for the marker analysis. In contrast, there was no segregation distortion and only few missing data points observed for Sat_228 in the F4:7 population, where high quality CTAB extracted DNA was used for the marker analysis.
F2:3 Populations
Based on the ELISA results, 40 AMV negative F2 plants ranging from OD –0.002 to 0.143 and 40 AMV positive F2 plants ranging from OD 0.676 to 2.249 were selected to produce F2:3 lines for a progeny test.
Visual Symptoms
Because plants within the F2:3 lines were rated as asymptomatic (0) or symptomatic (1), the lines were classified based on the proportion of plants observed with AMV symptoms. For 14 F2:3 lines derived from AMV-negative F2 plants, all plants were asymptomatic, and for 26 F2:3 lines derived from AMV-positive F2 plants, all plants were symptomatic. This high frequency of lines that were not segregating for visual AMV incidence was expected due to the development of these lines from the most and least symptomatic F2 plants.
ELISA
The absorbance readings among the F2:3 lines ranged from OD 0 to 2.25. Like with the visual symptoms, the lines were classified based on the proportion of plants within lines that were symptomatic (OD > 0.1). All plants were AMV negative for 19 lines and for seven lines all plants were AMV positive based on the ELISA tests. The correlation calculated on a single plant basis between ELISA and visual AMV incidence was r = 0.65 (P < 0.001).
DNA Marker Analysis
The proportion of plants within F2:3 lines that were symptomatic based on visual incidence and ELISA was used to confirm the map position of the resistance gene with marker scores from Sat_228, Sat_339, and Sct_065. Resistance based on visual incidence was mapped to position 30 on LG J (LOD = 8.0) with a support interval ranging from position 27 to 34 (Fig. 6
) and an R2 of 41.8%. Based on ELISA, the gene mapped to position 27 (LOD = 4.6) with an upper end support boundary at position 32 and an R2 of 26.8%. The QTL allele from PI 153282 had an additive effect of –0.4 units and a dominance effect of –0.07 units for visual AMV incidence and an additive effect of OD –2.4 and a dominance effect of OD –2.2 for ELISA.
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The resistance gene was mapped on LG J of the soybean map to a 9 cM interval surrounding Sat_228. To our knowledge, no resistance to viruses or any other pathogen has been previously mapped to this interval (SoyBase, 2007). The markers mapping close to the gene, such as Sat_228 would be suitable for marker-assisted selection, although better positioning of the gene relative to this and other markers would be beneficial.
In the F4:7 population, AMV response was measured by visually scoring AMV symptoms of the inoculated plants. In the F2 and F2:3 populations, AMV response also was scored with ELISA. The quite low correlation between ELISA and visual scoring in the F2 population (r = 0.29) compared to the moderate correlation in the F2:3 families (r = 0.65) likely reflects the single plant basis of the F2 results, which would have been subject to larger experimental error than the results for the F2:3 lines. We obtained the best mapping results in the F4:7 population due to the inbred nature of the lines and the large number of plants within experimental units. There were up to 12 plants in each experimental unit, which reduced our experimental error as indicated by a high heritability of 90%.
The single dominant gene hypothesis for AMV resistance agrees with previously reported ratio of 3:1 resistant/susceptible F2 plants in four populations segregating for Hood-derived AMV resistance (Almeida et al., 1982). At this time we do not know whether AMV resistance in Hood and PI 153282 is controlled by the same allele or locus. To address this objective, we have already started to develop appropriate plant material.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Received for publication August 15, 2007.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
