Identification of Soybean Aphid Resistance in Early Maturing Genotypes of Soybean Siddhi J. Bhusal, Guo-Liang Jiang,* Kelley J. Tilmon, and Louis S. Hesler
Abstract Soybean aphid (SA) (Aphis glycines Matsumura) has been an important pest of soybean [Glycine max (L.) Merr.] in the United States since 2000. Identification and genetic characterization of SA resistance in early maturing soybean germplasm will facilitate development of aphid-resistant cultivars in the northern region. To identify new sources of SA resistance in early maturing soybeans, a total of 334 soybean genotypes including resistant and susceptible checks were tested in the greenhouse and field. Caged (no-choice) and noncaged tests were used in greenhouse screening under artificial inoculation of SA, and field evaluations were performed relying on natural aphid infestation with or without artificial SA inoculation. In the greenhouse, four genotypes (PI 603712, PI 464911, PI 430491, and PI 603432B) of maturity group (MG) 0 or 00 exhibited low levels of aphid colonization similar to resistant checks, with 17 to 52 aphids per plant 2 wk after inoculation, and three genotypes (PI 612759B, PI 200595, and PI 603426D) of MG 0 were moderately resistant. In the field, however, only PI 603712 and PI 430491 exhibited a resistance reaction with fewer than 100 or 100 to 200 aphids per plant at peak infestation. Plant Introduction 603712 was the only genotype that consistently exhibited resistance to SA in all tests—even higher than that of other known sources of SA resistance in the field. This suggests that PI 603712 might be a new source of SA resistance. In addition, the relatively high levels of SA colonization on a Rag1 genotype (PI 548663 or ‘Dowling’) in greenhouse tests suggest that the colony used in greenhouse tests might be virulent on Rag1 and therefore might be biotype 2. High levels of SA infestation on Rag1 and Rag2 genotypes in field tests also imply that biotypes 2 and 3 may have been present in the eastern South Dakota field.
crop science, vol. 53, march– april 2013
S.J. Bhusal, G.-L. Jiang, and K.J. Tilmon, Plant Science Department, South Dakota State University, Brookings, SD, 57007; and L.S. Hesler, USDA-ARS North Central Agricultural Research Laboratory, Brookings, SD, 57006. Received 29 June 2012. *Corresponding author ([email protected]
). Abbreviations: FC, Forage Collection; MG, maturity group; MR, moderately resistant; R, resistant; RCBD, randomized complete block design; S, susceptible; SA, soybean aphid(s); SDSU, South Dakota State University.
oybean, a major crop in the United States, is grown in more than 30 states and produces about 83.2 Tg of grain on 30 million ha yr-1, with an average yield of 2.79 Mg ha-1 (41.5 bushels per acre). South Dakota grows 1.66 million ha of soybean acreage that yields 4.1 Tg annually (SOYSTATS, 2012). However, soybean yield has been limited by various abiotic and biotic factors including drought, insects, and pathogens. Among them, soybean aphid (SA) (Aphis glycines Matsumura), a native pest of soybean in Asia, has been an important pest of soybean in the United States. It was first reported in the United States in 2000 and had spread to 80% of the U.S. soybean-production area by 2002 (Hartman et al., 2001; Venette and Ragsdale, 2004). It became established in 30 states of the United States and three provinces of Canada by 2009 (Ragsdale et al., 2011). Soybean aphids feed by piercing leaves and stems with their proboscis and sucking plant sap, thereby causing symptoms of leaf curling, wilting, yellowing, and abscission and stunting plants. Aphid feeding also reduces the number and size of pods and the number of seeds set. Severe infestation of SA may cause over 50% soybean yield loss (Ostlie, 2002) and increase production costs for Published in Crop Sci. 53:491–499 (2013). doi: 10.2135/cropsci2012.06.0397 © Crop Science Society of America | 5585 Guilford Rd., Madison, WI 53711 USA 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.
its management (Ragsdale et al., 2011). It can reduce the annual value of soybean yield by US$2.4 billion (Song et al., 2006). Soybean aphids also causes crop damage indirectly by transmitting plant pathogenic viruses (e.g., Soybean mosaic virus, Alfalfa mosaic virus, and Bean yellow mosaic virus) and facilitating sooty mold development deposition of honeydew on leaf surfaces (Hill et al., 2001; Mensah et al., 2005). Soybean aphid grows and reproduces well at temperatures of 20 to 25°C (McCornack et al., 2004). Management of SA requires regular scouting to determine if populations reach the treatment threshold (Ragsdale et al., 2007) as their populations can double within 1.5 d under favorable conditions (McCornack et al., 2004). Although some pyrethroid and organophosphate insecticides are available (Hodgson et al., 2012), chemical spray is not always an ideal option as it may increase environmental contamination, kill beneficial insects, and trigger frequent pest outbreaks (Mensah et al., 2005). Therefore, resistant varieties may be the most effective, economical, and ecofriendly means to protect soybean from SA, by providing a preventive measure against pest outbreaks. Resistance to SA exists in some soybean Plant Introductions. Rag1, a single dominant gene, is the first gene of SA resistance identified in PI 548663 (Dowling) (Hill et al., 2006a). Soybean aphid resistance in ‘Jackson’ is also controlled by a single dominant gene, which is allelic to Rag1 in Dowling (Hill et al., 2006b, 2012; Li et al., 2007). Another single dominant resistance gene, Rag2, was identified in PI 243540 (‘Sennari’) (Mian et al., 2008b) and PI 200538 (‘Sugao Zarai’) (Kim et al., 2010). Zhang et al. (2010) reported Rag3 in PI 567543C. Two recessive genes account for SA resistance in PI 567541B and PI 567598B (Mensah et al., 2008; Zhang et al., 2009). The Soybean Genetics Committee (2009) provisionally designated recessive resistance genes as rag4 and rag1c in PI 567541B, rag1b in 567598B, and rag3b in PI 567543C. Several other genotypes that possess SA resistance have also been identified (Hill et al., 2004; Mensah et al., 2005; Diaz-Montano et al., 2006; Hesler et al., 2007; Hesler and Dashiell, 2007, 2008; Mian et al., 2008a). Most of the previously reported aphid-resistant genotypes belong to mid-late and late maturity groups (MGs). Only a few genotypes from MG 0 and 00 with resistance to SA are available (Hesler and Dashiell, 2009), and their resistance has not been confirmed by subsequent tests. Therefore, this study aimed to explore the SA resistance sources in early MGs of soybean. The objectives were to evaluate and identify SA resistance in soybean germplasm accessions from MG 00 and 0. The biotypes of SA in eastern South Dakota were also briefly explored and discussed.
MATERIALS AND METHODS Five experiments were conducted to evaluate early-maturity soybean genotypes against SA: two greenhouse tests and three field tests. In all, 334 soybean germplasm accessions, including 492
33 from MG 00 and 293 from MG 0 (test entries), six resistant checks, and two susceptible checks, were evaluated. Seeds were obtained from the National Soybean Research Center (Urbana, IL). All 334 genotypes were evaluated in the greenhouse on South Dakota State University (SDSU) campus using a caged (no-choice) test and a noncaged test. A subset of 330 of these genotypes was evaluated in a field test in Brookings, SD (four genotypes were not included due to lack of seeds). Two additional field tests were conducted in Aurora, SD. A subset of 323 soybean genotypes was included in one test. In the other test, 64 selected genotypes, six resistant checks, and two susceptible checks were evaluated to confirm the results of the greenhouse tests. The resistant checks used in the experiments were PI 548663 (Rag1), PI 243540 (Rag2), PI 567543C (Rag3), PI 567541B (rag4), PI 71506, and PI 567597C (Hill et al., 2006a; Mian et al., 2008b; Zhang et al., 2009, 2010; Van Nurden et al., 2010; Mensah et al., 2005). Susceptible checks were PI 567167 and PI 597386 (Hill et al., 2004; Hesler and Dashiell, 2009).
Greenhouse Screening Aphid Culture Virus-free adult SA were procured from the Soybean Entomology Laboratory, SDSU Plant Science Department, and multiplied on susceptible soybean genotype S19-R5 (Syngenta, Wilmington, DE) in insect-rearing cages in the greenhouse. The colony originated from several SA collected from a soybean field near Brookings, SD, in 2009. Temperature and relative humidity in the greenhouse were set at 22 to 25°C and 70% with slight variability. The light period was set at 16 h (5:00 AM–9:00 PM) per day with supplemental lights from high intensity filament lamps (430 W).
Caged (No-Choice) Test The caged (no-choice) experiment was conducted in a randomized complete block design (RCBD) with three replications. For each replication, five seeds per germplasm accession were planted in a plastic pot (10 by 10 cm for top and 10 cm deep). After emergence, three seedlings per pot were retained by removing extra seedlings and covered with a specifically designed cage, that is, a clear plastic cup with aphid-proof screen windows on opposite sides and on the top cover. Each plant was inoculated at the unifoliolate (VC) stage (McWilliams et al., 1999) with two wingless adult SA. Inoculated plants in each pot were then covered immediately with the cage. Plants were irrigated by adding water to holding trays to avoid interfering with SA infestations. The total number of aphids per plant was counted 1 and 2 wk after inoculation. The average number of aphids per plant at each counting time was calculated for each replicate or pot and used as the unit for statistical analysis. The data collected 2 wk after inoculation are presented. Replications of this caged test were conducted subsequently rather than simultaneously due to limitation of space and cage availability. A total of three replications were performed one at a time during spring, early fall, and mid fall in 2011, respectively. Due to logistical limitations of evaluating large number of test lines on the same day by counting all aphids per plant, the test lines were planted in different batches (subsequent days) within each replication, that is, four batches for the first
crop science, vol. 53, march– april 2013
Table 1. Description of scores used to evaluate soybean aphid infestations on soybean genotypes in the greenhouse noncaged test and field screening tests. Description Score
Number of aphids per plant for greenhouse noncaged test
Number of aphids per plant and severity of plant damage for field tests
1 2 3 4
≤25 26 to 100 101 to 200 201 to 500
≤25 aphids per plant; plant appears normal and healthy. 26 to 100 aphids per plant; plant appears normal and healthy. 101 to 200 aphids per plant; older leaves may be slightly yellowing 201 to 500 aphids per plant; plant appears slightly stunted with yellowing older leaves and slight curling young leaves and slight appearance of sooty mold and cast skins on stem and leaves.
>500 aphids per plant, without obvious plant damage (e.g., curling, yellowing and sooty mold development) >500 aphids per plant, with obvious plant damage (e.g., curling, yellowing and sooty mold development)
NA, not applicable.
replication and three batches for the second and third replication. The resistant and susceptible checks were included in all batches for comparison. Before the above experiment, a preliminary screening (or a one-replicate preparatory experiment) was also conducted in the spring of 2011 with the same set of entries and in the same manner as described above, but the plants were not caged before inoculation. The results are not presented here but were referred to for selection of the entries for a field experiment (i.e., confirmation test described below).
Noncaged Test In the fall of 2011 and early spring of 2012, a noncaged test was conducted in the greenhouse in a RCBD with three replications using the same planting and SA inoculation methods used in the caged test while the inoculated plants were not caged in this experiment. For each replication, all test entries and checks were planted at the same time. Two weeks after inoculation, the SA population per pot was scored using a 1 to 5 scale modified after Mian et al. (2008a) (Table 1).
Field Screening Aurora Field Test A subset of 323 soybean genotypes (322 test entries and one susceptible check) was planted in a three-replicate RCBD with one-row plots of 3 m length at the SDSU Aurora Research Farm in the summer of 2011. When most of the genotypes were at R4 to R5 developmental stages, natural infestation of SA was scored using a 1 to 5 scale (Table 1).
Brookings Field Test Three hundred thirty soybean genotypes were planted in 1.5-m-long single-row plots (50 seeds per plot) and replicated three times in a RCBD at the USDA-ARS Eastern South Dakota Soil and Water Research Farm near Brookings in the summer of 2011. Artificial inoculation supplemental to natural infestation was performed by placing one SA-infested soybean stem in the upper canopy of each plot on 19 July. Infested stems were obtained from separate aphid-colony plants (91B91; Pioneer Hi-Bred International, Inc., Johnston, IA) that were grown at the USDA-ARS North Central Agricultural Research Laboratory, Brookings. When susceptible plants in plots and border areas were heavily infested with SA, all experimental lines were scored in a 1 to 6 scale modified after Mian et al. (2008a) (Table crop science, vol. 53, march– april 2013
1). Most of the test plants were at R4 to R6 developmental stages when scored.
Confirmation Test of Selected Genotypes from Greenhouse Tests Based on the greenhouse observations performed in the spring of 2011, that is, the first replication of caged test and the preliminary observation (preparatory experiment), 64 of the genotypes that had fewer than 100 aphids per plant were selected and reevaluated in a three-replicate RCBD at SDSU Aurora Research Farm in the summer of 2011 along with the same six resistant and two susceptible checks. One-row plots of 0.6 m length were planted with 10 seeds per plot. To supplement natural infestation, experimental lines were inoculated by putting a leaflet or a part of a leaflet with 20 to 25 mixed-aged SA on the central plant of each plot at the V2 and/or V3 stage on 3 July. Soybean aphid-infested leaves were obtained from the greenhouse where SA were multiplied on susceptible genotype S19-R5 (Syngenta, Wilmington, DE) in insect-rearing cages. Two weeks later, the total number of aphids per plant was counted on first five plants for each plot. Genotypes with fewer than 200 aphids per plant at the first count (and also resistant and susceptible checks) were counted again on the same five plants at 4 wk after inoculation. Six weeks after inoculation, by which time most of the genotypes were at R5 to R6 stage, all experimental plots were scored using the 1 to 6 scale as described above (Table 1).
Data Analysis Analysis of variance was performed separately for individual tests on a basis of plot or pot means. Counts were square-root transformed to meet assumptions in ANOVA. Transformed counts were analyzed by using PROC GLM whereas scores were analyzed by using PROC GLIMMIX (SAS Institute, 2011). The SA counts in the resistant and susceptible checks from the caged test replicated over different times of the year (spring, early fall, and mid fall of 2011) were analyzed as multi-environment experiments to determine the effects of experimental times. The data taken from resistant and susceptible checks from different batches within each replication in the caged test were also analyzed separately to determine if there was a significant difference among groups within the same replication. Because the different batches within each replication were not significantly different, all the batches within each replication were combined and analyzed accordingly. Least significant
Table 2. Number of soybean germplasm accessions identified in different categories of soybean aphid resistance reaction to aphid infestations.
Category † R MR S †
Number of classified germplasm accessions and number of aphids per plant or aphid score‡ Greenhouse tests Field tests Caged Noncaged With inoculation (Brookings) Without inoculation (Aurora) 9 (5); 130 aphids per plant
9 (5); score 1.0–1.9 8 (0); score 2.0–2.9 317 (3); score ≥3.0
2 (1); score 1.0–2.9 3 (3); score 3.0–3.9 235 (4); score ≥4.0
1 (NA§); score 1.0–1.9 4 (NA); score 2.0–2.9 317 (1); score ≥3.0
R, resistant; MR, moderately resistant; S, susceptible.
The figure in parenthesis indicates the number of checks included.
NA, not applicable.
difference was used to examine the significance of differences among genotypes.
RESULTS A highly significant difference in resistance to SA infestation was observed among the tested soybean germplasm accessions in all the experiments (p < 0.0001). For a simplified description and easy reference to breeding use, the test lines were classified into three categories of resistance to aphid infestation, that is, resistant (R), moderately resistant (MR) and susceptible (S). On a case-by-case basis, the categorization was performed according to the reaction of a test line to aphid colonization (i.e., the number of aphids per plant or aphid score) and a comparison with the checks (Table 2). Briefly, in the greenhouse caged test, a reaction similar to resistant checks (p > 0.05) with fewer than 60 aphids per plant was regarded as R, a reaction different than susceptible checks (p < 0.05) with 60 to 130 aphids per plant was regarded as MR, and a reaction different from resistant checks (p < 0.05) with more than 130 aphids per plant was regarded as S. In other tests, aphid scores were used in categorization instead of number of aphids per plant. For the greenhouse noncaged test, R, MR, and S referred to a reaction similar to resistant checks (p > 0.05) and score of 1.0 to 1.9, a reaction similar to resistant checks (p > 0.05) and score of 2.0 to 2.9, and a reaction different from resistant checks (p < 0.05) and score of 3.0 or higher, respectively. In the field test with artificial inoculation, R, MR, and S applied to a reaction similar to resistant checks (p > 0.05) and score of 1.0 to 2.9, a reaction similar to resistant checks (p > 0.05) and score of 3.0 to 3.9, and a reaction with score of 4.0 or higher, respectively. For the field test without inoculation and no resistant checks included, R, MR, and S referred to a reaction different than susceptible checks (p < 0.05) and score of 1.0 to 1.9, a reaction similar to susceptible checks (p > 0.05) and score of 2.0 of 2.9, and a reaction with score of 3.0 or higher, respectively. The numbers of test lines identified with different reactions to SA infestation are presented in Table 2 and the detail is addressed below separately for individual tests. In view of limited pages, only the results of selected test lines (Plant Introductions) that exhibited R or MR reaction to SA in one test at least are presented (Tables 3 and 4). A complete set of data for all lines tested both in greenhouse and field has been published in the Germplasm Resources 494
Information Network website (http://www.ars-grin.gov/ cgi-bin/npgs/html/eval.pl?495135 and http://www.arsgrin.gov/cgi-bin/npgs/html/eval.pl?495136).
Greenhouse Caged (No-Choice) Test Analysis of variance with six resistant and two susceptible checks indicated that there was no significant difference in number of aphids per plant between replications performed at different times (p = 0.13), and no significant interaction existed between times and genotypes (p = 0.20). Genotypes significantly differed in SA numbers per plant (p < 0.0001). Plant Introduction 548663 was readily colonized by SA, with counts of SA per plant similar to susceptible check PI 597386 and even higher than susceptible check PI 567167 (Table 3). Other SA-resistant genotypes PI 243540, PI 567543C, PI 567541B, PI 71506, and PI 567597C exhibited resistance. Analysis of variance for combined data with 326 test entries and eight checks indicated that the expression of SA resistance was significantly different among genotypes (p < 0.0001), with a range of average numbers of aphids per plant from 12.3 to 507.4 2 wk after inoculation. Out of 326 test entries in the caged test, four genotypes (PI 603712, PI 464911, PI 430491, and PI 603432B) exhibited an average of fewer than 60 aphids per plant. Similar to PI 567543C, PI 603712 exhibited higher resistance than PI 567541B although the difference was not statistically significant. Plant Introduction 464911, PI 430491, and PI 603432B were statistically similar to all the resistant checks against SA. In addition, three genotypes (PI 612759B, PI 200595, and PI 603426D) averaged from 75 to 97 aphids per plant (Table 3) and were statistically similar (p > 0.05) to the resistant checks PI 567541B, PI 567597C, and PI 71506 but were less resistant (p < 0.05) than PI 567543C. The numbers of aphids per plant in PI 578388A, PI 612759A, PI 603698C, PI 603301B, and PI 437583 were relatively fewer (102–129) and significantly different than those in the susceptible checks. These five lines were also regarded as MR to SA.
Greenhouse Noncaged Test Soybean aphid infestation differed among genotypes (p 4) (Table 4), and there was no significant difference in SA colonization between PI 548663 or PI 243540 and the two susceptible checks. Resistant checks PI 567543C (source of Rag3) and PI 71506 exhibited significantly lower colonization (p < 0.05) of SA than the susceptible checks PI 597386 and PI 567167.
DISCUSSION Although some accessions of SA resistant germplasm have been identified, germplasm lines with SA resistance are still very limited. Exploring SA resistant germplasm in early MGs is not only a practical requirement for breeding early maturing cultivars but is also helpful for enhancement and diversification of the SA resistant gene pool. Therefore, early maturing (MG 0 and 00) soybean genotypes were evaluated for SA resistance in this study by caged and noncaged greenhouse screening and multiplelocation field screening. Four early maturing genotypes, PI 603712, PI 603432B, PI 464911, and PI 430491, were identified as resistant genotypes against the greenhouse SA colony. They consistently exhibited resistance similar to resistant checks in both caged and noncaged tests in greenhouse. The performance of PI 603712 (MG 0) was especially consistent for SA resistance. In addition, PI 200595, PI 578388A, and PI 603301B were MR against the greenhouse SA colony in both greenhouse tests. Under natural conditions in field screening, PI 603712 consistently exhibited a higher level of resistance in both locations (Brookings and Aurora, SD). It had smaller numbers of SA per plant and lower SA scores than those of 496
resistant checks although the difference was not significant in Aurora (Tables 3 and 4). Plant Introduction 430491 (MG 00) also had a similar level of resistance compared to resistant checks in both greenhouse and field tests. Therefore, we speculate that PI 603712 and PI 430491 may be new sources of SA resistance with genes different from Rag1 and Rag2. Genetic characterization of their resistance will be useful for soybean breeding for SA resistance. Plant Introduction 603432B and PI 464911 were resistant against the greenhouse SA colony. However, they did not show resistance in the field. Similarly, PI 200595, PI 578388A, and PI 603301B exhibited a moderate resistance to SA in greenhouse tests but were defeated in the field. In this study, a single SA colony was used in the greenhouse tests, but multiple genotypes or biotypes apparently existed in the field under natural conditions as discussed above. The presence of different genotypes or biotypes in the aphid population in the field might be the major reason for the differences in resistance responses of these genotypes between greenhouse and field tests (Hesler et al., 2007). The rapid evolution of SA biotypes has led to widespread virulence of soybean aphid (Michel et al., 2011). Another interesting phenomenon was that the test genotypes PI 612759B, PI 603426D, PI 612759A, and PI 603698C exhibited susceptibility in the noncaged test but exhibited moderate resistance in the caged test in the greenhouse. Except for PI 612759B in Aurora, they were susceptible in both field tests also. When tested in the field without inoculation, PI 612759B was scored as an intermediate reaction to SA (score of 2.7), statistically similar to both the susceptible check PI 567167 and resistant
crop science, vol. 53, march– april 2013
line PI 603712 (Table 3). Although the rate of SA growth and multiplication on this genotype in a field test with inoculation was slower (