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


     

Published online 19 March 2008
Published in Crop Sci 48:517-526 (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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saghai Maroof, M. A.
Right arrow Articles by Tolin, S. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Saghai Maroof, M. A.
Right arrow Articles by Tolin, S. A.
Agricola
Right arrow Articles by Saghai Maroof, M. A.
Right arrow Articles by Tolin, S. A.
Related Collections
Right arrow Soybean
Right arrow Plant Disease
Right arrow Crop Genetics

Pyramiding of Soybean Mosaic Virus Resistance Genes by Marker-Assisted Selection

M. A. Saghai Maroofa,*, S. C. Jeongb, I. Gunduzc, D. M. Tuckera, G. R. Bussa and S. A. Tolind

a Dep. of Crop & Soil Environmental Sciences, Virginia Tech, Blacksburg, VA 20461
b BioEvaluation Center, Korea Research Institute of Bioscience and Biotechnology, Chungbuk 363-883, Republic of Korea
c Phillip Morris USA, Research Center, Richmond, VA 23261
d Dep. of Plant Pathology, Physiology & Weed Science, Virginia Tech, Blacksburg, VA 24061

* Corresponding author (smaroof{at}vt.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Soybean mosaic virus (SMV) causes a disease of soybean [Glycine max (L.) Merr.] that is prevalent throughout the United States. The disease can be effectively managed through the deployment of single-dominant resistance genes known as Rsv genes that confer resistance to different strains of SMV. Pyramiding respective Rsv genes from different loci (Rsv1, Rsv3, and Rsv4) through marker-assisted selection (MAS) is an ideal method for creating durable and wide spectrum resistance to all strains of SMV. In this study, simple sequence repeat markers were used to create isogenic lines of the susceptible cultivar Essex containing one, two, or three Rsv loci for observing background and epistatic effects of Rsv1, Rsv3, and Rsv4 on inoculation with six strains of SMV. Results indicate that an Essex background or modifier genes from the donor source had effects on reactions of Rsv3 and Rsv4 genes, causing the isogenic lines to be more susceptible than the Rsv donor parents. Two-gene and three-gene isolines of Rsv1Rsv3, Rsv1Rsv4 and Rsv1Rsv3Rsv4, acted in a complementary manner, conferring resistance against all strains of SMV, whereas isolines of Rsv3Rsv4 displayed a late susceptible reaction to selected SMV strains. We demonstrate with MAS and three near-isogenic lines, each containing a different SMV-resistance gene, that pyramided lines can be generated in a straightforward manner into two- or three-gene–containing lines with high levels of resistance to SMV.

Abbreviations: BYDV, barley yellow dwarf virus • dpi, days postinoculation • ELISA, enzyme linked immunosorbent assay • LS, late susceptible • MAS, marker-assisted selection, MLG, molecular linkage group • NIL, near-isogenic line • RFLP, restriction fragment length polymorphism • SMV, soybean mosaic virus • SSR, simple sequence repeat


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
INCORPORATING DURABLE DISEASE RESISTANCE into susceptible host backgrounds has often been difficult using conventional breeding methods. Conventional breeding programs have been limited to monogenic, race-specific resistance genes since they are easy to introgress into susceptible backgrounds through simple backcrossing techniques (Kelly and Miklas, 1998). However, single gene resistance has often proven ephemeral and highly vulnerable to dynamic and diverse plant pathogen populations. Therefore, breeders are endeavoring to shift to breeding durable forms of resistance by pyramiding race-specific genes into a single cultivar. Conventional gene pyramiding requires extensive disease screenings with several races of the pathogen due to the race specificity of many of these genes after each cycle of crossing. Further complicating a pyramiding effort is the frequent absence of an effective selection method due to a lack of differentiating races. Testcrossing to susceptible genotypes is required in each cycle to detect the presence (or absence) of the masked genes due to epistatic interactions between many resistance genes. In addition, as highly effective resistance genes that are effective to many races of the pathogen are incorporated into breeding populations, valuable hypostatic genes are easily lost in each backcross cycle. Marker-assisted selection (MAS) of hypostatic genes would facilitate their maintenance in pyramiding populations.

Molecular markers tightly linked to resistance genes have facilitated their pyramiding into single elite cultivars, with the possibility of creating more durable or broad spectrum resistance. Liu et al. (2000) were able to successfully integrate three powdery mildew (Blumeria graminis) resistance gene combinations Pm2 + Pm4a, Pm2 + Pm21, and Pm4a + Pm21 into wheat (Triticum aestivum) cultivar Yang158. Double homozygote plants were selected from small F2 populations with the assistance of restriction fragment length polymorphism (RFLP) markers flanking the mildew resistance genes. Since Pm21 confers resistance to all known mildew isolates in China and Europe, conventional breeding strategies for pyramiding resistance would be difficult. Using similar pyramiding schemes in rice (Oryza sativa), four bacterial blight (Xanthomonas oryzae) resistance genes—Xa-4, xa-5, xa-13, and Xa-21 (Huang et al., 1997; Singh et al., 2001)—and three genes for blast resistance, pi1, Piz5, and Pita (Hittalmani et al., 2000), were incorporated into single rice cultivars using RFLPs and polymerase chain reaction (PCR)–based markers.

A few studies have reported pyramiding virus resistance genes for the potyviruses bean common mosaic virus in Phaseolus vulgaris (Kelly et al., 1995; Miklas et al., 2000) using molecular markers and pepper veinal mottle virus in Capsicum annuum (Caranta et al., 1996) through conventional breeding techniques. Werner et al. (2005) pyramided rym4, rym5, rym9, and rym11, all with different modes of action, into a single winter barley (Hordeum vulgare) using doubled haploid populations and molecular markers to select for F1 plants containing two, three, and eventually four resistance genes. Selection of lines carrying all four resistant genes without the aid of molecular markers was not feasible due to lack of differentiating virus strains. The resulting cultivar conferred resistance against barley mild mosaic virus (BaMMV) and two different strains of barley yellow mosaic virus (BaYMV and BaYMV-2).

Soybean mosaic virus (SMV, Potyvirus; Potyviridae) is one of the most common viral diseases of soybean [Glycine max (L.) Merr.], occurring in all areas where soybeans are grown (Thottapilly and Rossel, 1987) and capable of causing severe yield losses (Irwin and Goodman, 1981). Inoculation of plants at or before floral development results in extensive yield losses of nearly 40% (Ren et al., 1997) and causes reductions in seed quality due to seed coat mottling (Hobbs et al., 2005). Single dominant genes (Rsv) have been deployed in soybean cultivars for managing SMV incidence (Buss et al., 1988; Smith, 1968).

The majority of Rsv resistance alleles (Rsv1-y, Rsv1-m, Rsv1-t, Rsv1-k and Rsv1-s) are located at the Rsv1 locus and exhibit resistance to lower-numbered SMV strain groups (G1–G3) but susceptibility, often characterized by mosaic or systemic necrotic reactions, to higher-numbered strains (SMV G5–G7) (Chen et al., 1991). In contrast, lines with the Rsv3 locus are susceptible to lower-numbered strains but resistant to higher-numbered strains (Gunduz et al., 2002). Initially, it was reported that Rsv4 conferred resistance to all known SMV strains (Ma et al., 1995). However, Ma et al. (2002) and Gunduz et al. (2004) observed that SMV-G1 and -G7 are capable of long-distance vascular spread and limited local movement, restricting the virus to cells along veins of later-developing leaves in heterozygous, Rsv4-containing plants. The appearance was unlike a typical susceptible mosaic reaction and was termed a late susceptible (LS) phenotype (Ma et al., 2002).

In the current study, PCR-based flanking markers for Rsv1, Rsv3, and Rsv4 were used to pyramid all three resistant genes into a single southern-adapted line, ‘Essex’, to create a durable form of resistance to SMV. Possible epistatic effects among and between the three respective Rsv genes were also investigated in an Essex background on inoculation with SMV-G1, -G2, -G3, -G5, -G6, and -G7. By screening these Rsv-pyramided lines with six SMV strains in greenhouse studies, we were able to demonstrate that reaction of the pyramided lines with differing Rsv gene combinations often varied from predictions based on single gene screening results.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plant Materials
Three isogenic lines were developed that each carried a single dominant SMV resistance gene in a susceptible Essex recurrent parent background and were used for pyramiding Rsv genes. V94-3972, a BC6 Essex isoline, carries SMV resistance gene Rsv1 derived from ‘Epps’, which is an isoline carrying the Rsv1 allele derived from PI96983 (Hartwig, 1984). V229 is a BC6 Essex isoline with SMV resistance gene Rsv3 derived from L29 (Bernard et al., 1991). V97-9003 is a BC4 Essex isoline with SMV resistance gene Rsv4 derived from the donor parent, D26. D26 is an F3:4 line from an Essex x PI86355 cross that was selected for uniform resistance to SMV (Ma et al., 1995). Additional advanced backcross and sister lines were used in the disease screening experiments. V00-7266 is an advancement of the Rsv1 isoline V97-9003. V239, a sister line of V229, is a BC6 Essex isoline containing Rsv3. V00-7339 is an advancement backcross of V97-9003. Disease screening was also conducted on V94-5152, an F6:7 selection from D26 population (Buss et al., 1997).

The pyramiding scheme to incorporate the three Rsv genes into Essex is shown in Fig. 1 . Pairwise crosses were first performed in spring 2000 to facilitate the construction of three possible two-gene pyramid combinations including Rsv1Rsv3, Rsv1Rsv4, and Rsv3Rsv4. F1 plants were grown at Warsaw, VA, in summer 2000 to produce F2 seeds. All F1 plants were screened with molecular markers linked to Rsv1, Rsv3, and Rsv4, as described below, to confirm that they were true crosses. The following winter in the greenhouse, 100 F2 seeds from each combination were planted in plastic pots and inoculated with SMV strains G1 and G7. F2 individual plants from Rsv1Rsv3 and Rsv1Rsv4 populations were inoculated with SMV-G7, and Rsv3Rsv4 populations were inoculated with SMV-G1. Susceptible plants, which showed a mosaic reaction 15 d postinoculation (dpi), were discarded, and genomic DNA from resistant F2 individual plants was screened with molecular markers. Molecular marker screening was repeated for the plants determined to be homozygous dominant for the two respective Rsv genes, for confirmation purposes. Disease data for all the remaining resistant F2 plants were collected at additional intervals of 4 and 8 wk after inoculation.


Figure 1
View larger version (35K):
[in this window]
[in a new window]

  		Figure 1. Schematic diagram showing marker-assisted selection for pyramiding the three major genes (Rsv1, Rsv3, and Rsv4) for soybean mosaic virus (SMV) resistance.

 
On the basis of the results of disease and marker screenings, F2 individual plants possessing two homozygous alleles at two Rsv loci were crossed to each other to pyramid three SMV resistance genes into single lines (Fig. 1). The F1 seeds from the crosses were planted in the greenhouse in winter 2001 to produce F2 seeds. DNA samples from F1 plants were screened with molecular markers closely linked to Rsv1, Rsv3 and Rsv4 that identified three true F1 plants designated F1-42, -43, and -44 (Fig. 2 ), from which F2 seeds were harvested. Forty F2 seeds from each F1 plant, a total of 120 seeds, were planted in pots in summer 2001 in the greenhouse. Genomic DNA from resistant F2 individual plants were screened with molecular markers, as described below, for the presence of each of the three genes. Molecular marker screening of the 120 plants was repeated for confirmation purposes. A majority of F2 plants lost one of the three genes and were discarded. Seeds from three F2 plants (F2-42-5, F2-43-12 and F2-44-21) that contained homozygous resistance alleles at the three gene loci were harvested.


Figure 2
View larger version (51K):
[in this window]
[in a new window]

  		Figure 2. Polyacrylamide gel showing the separation of the alleles of polymerase chain reaction–based molecular markers. The microsatellite marker genotyping of three F1 plants obtained by the crosses between ‘Essex’ isolines containing two soybean mosaic virus resistance genes is shown. The expected genotypes of F1 plants and their parental plants are listed in the inset.

 
DNA Isolation and Molecular Marker Analysis
Soybean leaf tissue samples were prepared following two DNA isolation protocols. The procedure by Doyle and Doyle (1987) was used to isolate large amounts of quality DNA from bulked tissues. To obtain crude DNA suitable for PCR analysis, a simplified procedure by Edwards et al. (1991), with some modifications for marker analysis of plant tissue, was used.

Simple sequence repeat (SSR) markers Satt114, Satt510, Satt120, and 64A8C, are closely linked to Rsv1 by 3 cM or less on molecular linkage group (MLG) F (Gore et al., 2002; Yu et al., 1994). Simple sequence repeat markers Satt063 and A519F/R are linked to Rsv3 at distances of 3.2 cM and 0.8 cM, respectively, in a ‘Tousan 140’/‘Lee 68’ F2 population and map to mLG-B2 (Jeong et al., 2002). A comparison of the Rsv3 linkage map from the Tousan 140/Lee 68 population (Jeong et al., 2002) and the integrated soybean map from Cregan et al. (1999) indicated that Satt560 is linked to Rsv3 at a distance of approximately 6 cM. Simple sequence repeat markers Satt558 and Satt542 map to mLG-D1b and are linked to Rsv4 at distances of 7.8 and 4.7 cM, respectively (Hayes et al., 2000). Each F2 individual plant was screened by two flanking markers for selection of segregating Rsv gene(s). Primer sequences of 64A8C were kindly provided by R. Innes (Indiana University, Bloomington). Primer sequences of A519F/R were previously reported by Jeong et a al. (2002). All other microsatellite primer sequences are publicly available at Soybase (http://soybase.agron.iastate.edu) or previously published (Cregan et al., 1999). Amplification using SSR primer sets was performed essentially as described by Yu et al. (1994) and Cregan et al. (1999). Polyacrylamide gel electrophoresis was performed as previously described by Saghai Maroof et al. (1994).

Virus Cultures and Inoculation
Strains of SMV in this study were SMV-G1, -G2, -G3, -G5, -G6, and -G7 (Cho and Goodman, 1979) and were the same strains used in previous work (Chen et al., 1991; Hayes et al., 2000). Cultures were maintained in the greenhouse by periodic transfer to Essex for SMV-G1, Lee 68 for SMV-G2, and -G3, and ‘York’ for SMV-G5, -G6, and -G7. Greenhouse inoculations were performed as described by Chen et al. (1991) by grinding infected trifoliolate leaflets in 0.01 M neutral sodium phosphate buffer (1:10; w/v) in a mortar and pestle and using the pestle to rub unifoliolate leaves (V-1 stage) previously dusted with 600 mesh carborundum (Buehler, Lake Bluff, IL). Plants were rinsed with tap water immediately thereafter. The pathotype of all SMV strains was confirmed by inoculation to the standard differential cultivars (Chen et al., 1991; Cho and Goodman, 1979, 1982) (Table 1 ) before screening potentially pyramided lines.


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

  		Table 1. Effect of six soybean mosaic virus (SMV) strains on parental lines, ‘Essex’ isolines, and pyramided isolines and differential cultivars. Results are a composite of weekly observations of plants 7 wk postinoculation.

 
Four parental lines, five differential cultivars, and 12 Essex isolines containing one, two, or three Rsv genes were tested for reaction to six SMV strains (Table 1). For each strain–line combination, one plastic pot (7.6 cm diam., 7.6 cm depth) containing six seeds was planted in two sets 6 wk apart, one for SMV-G1, -G2 and -G3 and a second for SMV-G5, -G6 and -G7. Pots were filled with MetroMix360 (Scotts-Sierra Horticultural Products, Marysville, OH), and Osmocote 18% N, 6% P, 12% K (Scotts-Sierra Horticultural Products) was used as a supplemental fertilizer as needed. A mock-inoculated control pot of each entry was included in each set. Plants in each set were inoculated as described above in 1 d and then were randomly distributed on two greenhouse benches for each set. An automated watering system was used to reduce the chance for strain contamination among neighboring pots.

Experiments were conducted over a period of 7 wk, with detailed scoring at 7, 10, 14, 21, 28, and 35 dpi. Digital images of selected pots were taken over time for each treatment. At 35 dpi, the relative virus accumulation in leaves of selected treatments was assessed by enzyme linked immunosorbent assay (ELISA). Three upper trifoliolate leaflets from each pot tested were selected at random, imaged digitally, weighed, and placed in a grinding bag (Agdia, Inc., Elkhart, IN). An amount of buffer to give a final dilution of 1:10 (w/v) was added to the bag, and leaves were completely macerated with a ball-bearing grinding tool (Agdia, Inc.). A final dilution to 1:50 was made in grinding buffer (carbonate) for indirect ELISA, and 200 µL of each treatment was placed in each of two wells of a microtiter plate. Following incubation with 1:1000 dilution of antiserum to SMV (Hunst and Tolin, 1982), anti-rabbit IgG (whole molecule), F(ab)')2 fragment-Alkaline Phosphatase antibody produced in goat (Sigma-Aldrich, St. Louis, MO), and 4-nitrophenyl phosphate (pNPP) substrate (Sigma-Aldrich), absorbance at 405 nm was recorded with a Spectramax (Molecular Devices, Sunnyvale, CA) plate reader. Leaf immunoprints were also taken of selected treatments according to Gunduz et al. (2004).


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Pairwise Combination of Soybean Mosaic Virus Resistance Genes
To pyramid Rsv1 and Rsv3 genes, V94-3972 (Rsv1-near isogenic line NIL])was crossed with V229 (Rsv3-NIL) (Fig. 1). True crosses were determined by using Rsv1 and Rsv3 molecular markers as described above. Eighty-four F2 plants were inoculated with SMV-G7. Plants exhibiting mosaic or necrotic symptoms were assumed to be lacking Rsv3. Twenty-seven such plants were identified and discarded. Fifty-seven symptomless plants were retained. Genomic DNA of resistant plants was assayed with SSR markers 64A8C (Rsv1) and A519F/R (Rsv3). Five of the 57 resistant plants were determined to be homozygous dominant for both Rsv1 and Rsv3. Observed disease and molecular data fit a 12R:3N:1S ratio. Seeds from F2 plants that were homozygous for Rsv1 and Rsv3 were planted to produce F2:3 families. Progeny from two of five F2 plants, V02-8900-1 and V01-7393-1, underwent greenhouse screenings with six SMV strain discussed in further sections.

V94-3972 (Rsv1-NIL) was crossed with V97-9003 (Rsv4-NIL) (Fig. 1) to pyramid Rsv1 and Rsv4 genes. Similarly, as described above, genomic DNA from F1 plants of this combination was screened with Rsv1 and Rsv4 molecular markers to detect true crosses. Disease screening of 86 F2 plants confirmed that 32 were necrotic or susceptible to SMV-G7. The necrotic reaction on these plants originated from the Rsv1 source as the gene exhibits a necrotic reaction to SMV-G7. Genomic DNA from 54 resistant plants was screened with SSR markers 64A8C (for Rsv1) and Satt558 (for Rsv4), and 6 plants were determined to be homozygous for both genes. Observed disease and molecular data are in complete agreement with expected Mendelian ratios (12R:3N:1S ratio) when the F2 population was inoculated with SMV-G7. Progeny of two of six F2 plants that remained symptomless, V02-8889-1 and V01-7364-1, were screened with six SMV strain to examine reactions types conferred by the pyramid.

To pyramid Rsv3 and Rsv4 genes, V229 (Rsv3-NIL) was crossed with V97-9003 (Rsv4) (Fig. 1) as described above. Forty-one of 92 F2 plants were susceptible on inoculation with SMV-G7 and were discarded. Screening genomic DNA from 51 resistant plants with SSR markers Satt560 (Rsv3) and Satt558 (Rsv4) indicated that five individual F2 plants were homozygous resistant at both gene loci. Three of five double homozygous individuals exhibited an LS reaction, while the other two lines remained symptomless. The F2 population did not segregate to fit a normal 12R:4S ratio when inoculated with SMV-G1 or -G7 probably because plants exhibiting an LS reaction were classified as susceptible. Rsv3 is susceptible to SMV-G1, and some heterozygotes of Essex and the Rsv4-containing PI88788 often exhibited LS reactions (Gunduz et al., 2004). In genetic studies of PI88788 and ‘Columbia’, which contains genes considered to be Rsv3 and Rsv4, LS plants were classified as resistant (Ma et al., 2002; Gunduz et al., 2004). Molecular marker assays of the F2 population clearly indicated alleles segregated to fit expected ratios. Similar to the other two gene combinations, seed from two Rsv3Rsv4 homozygous F2 plants which remained symptomless (V02-8908-1 and V01-7341-1), was harvested and screened with six SMV strains in a greenhouse experiment.

Selection of Plants with Three Rsv Genes
The 2000–2001 winter study confirmed a total of five Rsv1Rsv3, five Rsv1Rsv4, and five Rsv3Rsv4 F2 plants with Rsv gene combinations that imparted resistance to SMV and contained homozygous resistance alleles at molecular marker loci. These homozygous, two-gene lines were crossed to each other to pyramid three Rsv genes. Screening of F1s was done as described for the two-gene pyramid development. Marker analysis indicated plant F1-43 from crosses between homozygous Rsv1Rsv3 and Rsv3Rsv4 combinations and F1-42 and F1-44 from crosses between homozygous Rsv1Rsv4 and Rsv3Rsv4, contained the expected genotypes (Fig. 2).

Progeny from three F1 plants were screened with Rsv1-linked molecular markers to detect F2 plants containing homozygous dominant alleles at all three SMV resistance loci. A total of 40 plants from each of three separate F2 populations (F2-42, -43, and -44) were analyzed to examine Rsv gene combinations. The F2-42 population contained four plants (F2-42-5, -14, -15, -20) that were homozygous dominant at all three Rsv loci. From 40 plants analyzed in the F2-43 population, two plants, F2-43-12 and F2-43-21, were homozygous dominant at all three Rsv loci. The F2-44 population contained 2 plants, F2-4, -21, and -30, that were homozygous dominant for all three Rsv loci, and 13 plants that contained only two homozygous dominant loci. Seed from F2-42-5, F2-43-12, and F2-44-21 plants was harvested and screened for reaction to six strains of SMV in the greenhouse experiment.

Responses of and Virus Accumulation in Parental Lines to Virus Inoculation
The results of inoculation of parental lines and differential cultivars verified the expected pathotype of the SMV strains, as shown in Table 1. Briefly, within 6 to 7 dpi, all Essex and Lee 68 (rsv) plants inoculated with each of six virus strains exhibited vein clearing and systemic mosaic symptoms, indicative of a susceptible reaction. All mock-inoculated controls remained symptomless throughout the course of the study. York was resistant to SMV-G1, -G2 and -G3 but susceptible to SMV-G5, -G6, and -G7. Reactions of ‘Kwanggyo’, ‘Marshall’, and ‘Ogden’ were as expected (Table 1), with systemic necrosis restricted to minor veins (SMV-G2 and -G3 on Marshall; SMV-G3 on Ogden) or systemic necrosis of leaves, petioles, and the primary meristem (SMV-G5, -G6, and -G7 on Kwanggyo; SMV-G7 on Marshall and Ogden). The source of Rsv1, PI96983, developed symptoms after inoculation only with SMV-G7, beginning as localized small necrotic lesions 6 dpi and necrotic veins by 7 dpi. Veinal necrosis and interveinal necrotic spots developed on upper, noninoculated trifoliolates by 10 dpi. Necrosis of the primary meristem, second and third trifoliolate leaves, and petioles was evident on PI96983 by 21 dpi on all SMV-G7 inoculated plants, but not after inoculation with any other SMV strain.

The source of Rsv3, L29, developed mosaic symptoms in all plants inoculated with SMV-G1, -G2, and -G3. Plants inoculated with SMV-G3 showed a mild mosaic early, but later developed severe leaf curl. A few 1- to 2-mm necrotic spots were observed on upper trifoliolate leaves 21 to 28 dpi after inoculation with SMV-G5 and -G6. V94-5152, an F6:7 selection from D26, the source of Rsv4 (Buss et al., 1997), was resistant to all six SMV strains and developed no symptoms during the time observed, except for transient chlorotic flecking with SMV-G6 at 28 dpi (Table 1).

Virus accumulation in Essex was essentially equivalent in plants infected with SMV-G1, -G2, or -G3 and showing similar mosaic symptoms and in SMV-G3-infected L29 (Table 2 ). However, relative virus accumulation was about 30% lower in SMV-G1-infected L29. No virus was detected by ELISA in V94-5152 inoculated with any of the six virus strains.


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

  		Table 2. Relative accumulation of six soybean mosaic virus (SMV) strains in selected soybean cultivars and ‘Essex’ isolines with resistance genes, at 35 d after inoculation, in an indirect enzyme linked immunosorbent assay (ELISA).

 
Reaction of and Virus Accumulation in Essex Isolines of Rsv1, Rsv3, or Rsv4
Results of inoculation of three Essex isolines with six strains of SMV demonstrated that these lines had the same qualitative response to virus strains as did the corresponding parental line, but there were some differences in symptom severity (Table 1). The Rsv1 isoline V00-7266, an advancement of V97-3972, was resistant to the five lower-number strains, but when inoculated with SMV-G7, it developed severe leaf crinkle and rugosity. Necrosis developed later on inoculated leaves of the isoline than on PI96983, and systemic necrosis on the isoline was confined to leaf lamina and was not observed in meristematic and petiole tissues (Fig. 3B, 3C ). Comparing the Rsv1 isoline to the recurrent parent, Essex, symptoms caused by SMV-G7 were less pronounced at 14 dpi than the isoline exhibiting only a mild mosaic (Fig. 3A).


Figure 3
View larger version (84K):
[in this window]
[in a new window]

  		Figure 3. (A) Essex Rsv1 donor source, (B) PI96983, and (C) Rsv1 Essex isoline V00-7266 inoculated with SMV-G7 14 d postinoculation (dpi). Reaction of (D) Essex, (E) L29, parental source of Rsv3, and (F) isoline V239 inoculated with SMV-G2 14 dpi. (G) Rsv3Rsv4 pyramided Essex isoline V01-7341-1 characterized late susceptible reaction inoculated with SMV-G1 35 dpi. (H) Immunoprint of the leaf pictured in Fig. 3G showing distribution exhibiting nonrandom distribution of virus indicated by dark areas on the leaf blot.

 
Isoline V239, an advancement of V229, reacted similarly to L29 but with some differences. SMV-G1 caused nearly identical systemic mosaic symptoms in the Rsv3-containing parent and isoline, with virus accumulation at a level reduced from that in Essex by 30 and 50% (Table 2), respectively. SMV-G2 caused much more severe mosaic symptoms and stunting in V239 than in Essex and L29 (Fig. 3D, 3E). SMV-G3 was also more severe and caused leaf curling on the isoline, but virus accumulation was equivalent to that in L29 and in Essex. SMV-G5 caused no mosaic symptoms on L29, but faint necrotic flecks were observed on some trifoliolate leaves of V239. SMV-G6 caused local chlorotic and necrotic lesions on inoculated leaves of isoline V239, but not on L29, indicating some local infection by virus but no systemic symptoms. SMV-G7 caused no symptoms in any Rsv3-containing entries, and absence of virus in V239 isoline was confirmed by ELISA (Table 2).

The Rsv4 isoline V00-7339, an advancement of V97-9003, showed no symptoms in the first 2 wk following inoculation with all strains, similar to V94-5152. However, the isoline began to develop late symptoms at 21 and 28 dpi, with SMV-G2 and -G5. The upper trifoliolate leaves began to show chlorotic spots and line patterns, with vein banding and islands of green tissue surrounded by chlorotic veins in fourth and later trifoliolates. Faint lesions were also observed at 28 dpi on SMV-G6- and SMV-G7-inoculated unifoliate leaves. No virus was detected at 35 dpi with any of the SMV strains inoculated to V94-5152 leaves. In contrast, virus accumulated in the Rsv4 isoline to levels in SMV-G2- and SMV-G5-inoculated plants was nearly equivalent to that of susceptible Essex (Table 2).

Reaction of and Virus Accumulation in Isolines with Pyramided Rsv Genes
Inoculation of SMV strains to Essex isolines having two Rsv genes confirmed that the two genes were functional and often imparted greater resistance than the corresponding single genes, particularly for the Rsv1Rsv3 and Rsv1Rsv4 gene pyramids. Rsv1Rsv3 lines V02-8900-1 and V01-7393-1 were uniformly resistant to inoculation with all six SMV strains, showing complementary gene action (Table 2).

Rsv1Rsv4 lines, V02-8889-1 and V01-7364-1, were also uniformly resistant to all SMV strains, including SMV-G7. However, SMV-G7 produced a few late-developing (28 dpi) necrotic lesions in V02-8889-1 but no chlorois or mosaic symptoms, while disease symptoms were entirely absent in the V01-7364-1 isoline (Table 2).

Two lines of Rsv3Rsv4 also responded nearly identically, showing no symptoms following inoculation with SMV-G5, -G6, and -G7. No symptoms were observed on V02-8908-1 and V01-7341 inoculated with SMV-G1, -G2, and -G3 until 21 dpi. At that time, chlorotic spots began to appear on second and third trifoliolate leaves of plants inoculated with SMV-G1 and -G2. Later leaves showed vein banding, green islands, and chlorotic spotting by 28 d, typical of the LS phenotype (Fig. 3G). Plants inoculated with SMV-G3 showed some chlorotic spotting on third and fourth trifoliolate leaves at 21 dpi, but by 28 and 35 dpi, no symptoms were observed, suggesting that Rsv4 continued to be dominant over Rsv3 in SMV-G3-inoculated plants.

The Rsv3Rsv4 isolines inoculated with SMV-G1 and SMV-G2 had relatively moderate to high levels of virus titer at 35 dpi (Table 2). Rsv4 isoline, V00-7339, had no detectable SMV-G1 but a moderate level of SMV-G2 (ELISA = 2.111) and SMV-G5 (ELISA = 2.640) at 35 dpi. Immunoprints of the Rsv3Rsv4 isolines indicated the virus was not uniformly distributed throughout the leaf in LS reactions but was localized to chlorotic areas (Fig. 3H). Three Rsv1Rsv3Rsv4 lines resulting from crosses among the backcross lines having two Rsv genes were similarly inoculated with the six SMV strains (Table 1). In all cases, no symptoms were observed during the 35 d of observation.


   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Allelism studies of SMV-resistant soybean accessions such as PI486355 (Chen et al., 1993), Columbia (Ma et al., 2002), Tousan 140, and ‘Hourei’ (Gunduz et al., 2002) have shown that each of these lines carry two Rsv genes: Rsv1+Rsv4, Rsv3+Rsv4, and Rsv1+Rsv3, respectively. Resistance in these lines is likely due to interactions of the independent dominant resistance genes conditioning different types of resistance (Bowers et al., 1992; Hayes et al., 2004; Ma et al., 1995). In this experiment, we obtained similar results for Rsv1Rsv3 and Rsv1Rsv4 gene combinations. The Rsv1Rsv3 combination provides complete resistance to all SMV strains. Pairwise pyramid combinations of Rsv1Rsv3 and Rsv1Rsv4 have complementary effects, resulting in an extreme resistance response to all strains.

Essex isoline V00-7339 carrying Rsv4 exhibited late susceptibility to SMV-G2 and -G5 but not to other strains. Two separate Rsv3Rsv4 isolines also exhibited LS reaction to SMV-G1 and -G2, but not to SMV-G5. On these isolines, SMV-G3 caused a new phenotype of only faint chlorotic spots after 2 to 3 wk but no symptoms on later emerging leaves. A selection from a D26 population, V94-5152, showed no late susceptibility to any strains of SMV. Unpredicted effects of combining resistance genes in the current study may be explained by contributions of tightly linked genes flanking the R gene as reported by high-resolution mapping studies (Gore et al., 2002) and eventual cloning of resistance gene candidates (Hayes et al., 2004). PI96983 confers resistance to a wider spectrum of SMV strains than do other Rsv1-carrying cultivars (Chen et al., 1994) and contains one candidate Rsv1 gene (3gG2) along with additional "complement" genes in the gene cluster conditioning its extreme resistance to strains SMV-G1 through SMV-G6. Hayes et al. (2004) reported that the presence or absence of other complementary genes along with the 3gG2 gene in the cluster conditioned unique reactions to SMV in rare recombinants of a PI96983 (Rsv1) by Lee 68 (rsv) population. Lines missing these complementary gene(s) displayed reaction types similar to Ogden (Rsv1-t) and Marshall (Rsv1-m), which show a necrotic response to a greater number of strains than does PI96983. Therefore, multiple members and combinations of these genes at the Rsv1 locus condition different responses to SMV strains. Repeated backcrossing to the recurrent parent, Essex, in this study may have eliminated unidentified complementary genes at an Rsv locus, particularly Rsv4, causing the isogenic lines to become more susceptible than the original R gene donor parent. Wang et al. (2006) found L92-8580, an isogenic line of Williams, to be susceptible to SMV-G5 whereas the donor parent ‘Suweon 97’ was resistant to SMV-G5. They concluded that a possible rare recombination event at the Rsv locus between the two lines may have occurred, similar to reports by Hayes et al. (2004) in the PI96983 by Lee 68 population. We conclude that, background effects must be considered when pyramiding resistance genes either conventionally or through transformation methods, as not only the effectiveness of the resistance gene may vary, but the reactions conditioned in different backgrounds to different virus strains or pathotypes may differ.

Results from the current study clearly demonstrate that combining Rsv3Rsv4 created a late susceptibility reaction which results from a delay in virus movement throughout plants. The term tolerance has been used to characterize similar reactions of barley yellow dwarf virus (BYDV) on some members of the Graminea family. Cooper and Jones (1983) defined tolerance as the situation in which the virus is allowed to replicate in host tissues but that tolerant cultivars do not develop severe disease symptoms. Oat (Avena sativa) lines also have been shown to have lower amounts of BYDV compared with susceptible lines without a hypersensitive reaction at initial infection sites (Gray et al., 1993). Chain et al. (2005) concluded that wheat lines tolerant to BYDV reduce the development rate of virus in the first days of infection and reduce virus load in infected leaves.

Late susceptibility likely exerts less selection pressure on virus populations and decreases the probability of mutations arising in virus populations to overcome host resistance (Burdon, 1993), as has occurred for alleles at the Rsv1 locus (Cho and Goodman, 1979). Deployment of cultivars possessing Rsv3Rsv4 genes may contribute to broad, effective, and long-term resistance since the pyramided line allows some replication but prevents rapid long-distance movement of virus in the plant and thus decreases virus inoculum in a field. Late susceptibility can be viewed similarly to tolerance as it allows pathogens to infect and colonize hosts but does not place extensive pressure on pathogen populations, thus creating a durable form of resistance.

Cho and Goodman (1979) observed that the most common SMV strains were SMV-G1, -G2 and -G3 in the United States. Breeding for SMV resistance is even more important in Korea and Japan, where several recent reports of resistance-breaking strains have been reported (Choi et al., 2005; Koo et al., 2005; Saruta et al., 2005). Because the virus is seedborne, seed distribution serves to distribute the virus to different environments and other areas of the world. Presumably, SMV host resistance that reduces initial success of establishing infection and rate of within-plant spread may provide greater durability (Bar-Zur and Salomon, 1995; Dintinger et al., 2005; Gunduz et al., 2004; Padgett et al., 1990), which is necessary even in the absence of resistance-breaking strains.

Pyramiding SMV resistance genes into an elite line is predicted to increase resistance against additional strains of SMV by epistatic interaction among genes. Breeding programs have had difficultly in selecting homozygous plants with multiple resistance genes on the basis of disease reaction of haplotype alone, partially from one gene masking the presence of another. In the current study, we have demonstrated, with MAS and three NILs each containing a different SMV resistance gene, that pyramided lines can be generated in a straightforward manner. Availability of isolines carrying the major genes Rsv1, Rsv3, and Rsv4 in the common Essex genetic background allowed us to combine all three resistance genes into a single line in only two cycles of crossing. DNA markers have no masking effect on each other and aid in identifying homozygous pyramided lines from F2 generations without tedious inoculations and/or progeny testing or the need for pathogen strains to identify the presence of specific resistance genes. However, reactions conditioned in the Essex isolines sometimes differed from the original donor parental source of the Rsv gene. Pyramided lines also displayed uncharacteristic symptoms to some virus strains that differed from the reaction exhibited by the parental lines. For example, Rsv4 was not dominant over Rsv3, as a late susceptible reaction was found in the two-gene pyramid. The three-gene pyramid remained resistant to all strains and exhibited no late susceptible reaction during the course of the experiment proving the effectiveness of MAS in developing highly resistant cultivars. These isogenic lines provide unique genetic materials that could facilitate global gene expression studies with the aid of microarray technology in future experiments. Isogenic lines that possess one, two, or all three of the identified SMV resistance genes, in the same susceptible background, should help our understanding of the epistatic interactions and genetic mechanisms between these distinct loci from different sources (Saghai Maroof et al., 2008). Identifying additional genes or modifiers for SMV resistance will be necessary to create a durable, broad-spectrum resistance in future cultivars.


   ACKNOWLEDGMENTS
 
This study was supported in part by the USDA NRICGP Grant no. 96-35300-3648, United Soybean Board and Virginia Soybean Board. The authors would like to express gratitude to Moss Baldwin for her technical assistance inoculating plants and photographic images.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 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 October 19, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Crop Sci.Home page
D. M. Tucker, M. A. S. Maroof, S. C. Jeong, G. R. Buss, and S. A. Tolin
Validation and Interaction of the Soybean Mosaic Virus Lethal Necrosis Allele, Rsv1-n, in PI 507389
Crop Sci., June 26, 2009; 49(4): 1277 - 1283.
[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 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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saghai Maroof, M. A.
Right arrow Articles by Tolin, S. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Saghai Maroof, M. A.
Right arrow Articles by Tolin, S. A.
Agricola
Right arrow Articles by Saghai Maroof, M. A.
Right arrow Articles by Tolin, S. A.
Related Collections
Right arrow Soybean
Right arrow Plant Disease
Right arrow Crop Genetics

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