Pyramiding multiple genes for resistance to soybean mosaicvirus in soybean using molecular markers

Ainong Shi Æ Pengyin Chen Æ Dexiao Li ÆCuiming Zheng Æ Bo Zhang Æ Anfu Hou

Received: 19 December 2007 / Accepted: 20 August 2008

� Springer Science+Business Media B.V. 2008

Abstract Seven strains of Soybean mosaic virus

(SMV) and three independent resistance loci (Rsv1,

Rsv3, and Rsv4) have been identified in soybean. The

objective of this research was to pyramid Rsv1, Rsv3,

and Rsv4 for SMV resistance using molecular

markers. J05 carrying Rsv1 and Rsv3 and V94-5152

carrying Rsv4 were used as the donor parents for gene

pyramiding. A series of F2:3, F3:4, and F4:5 lines

derived from J05 9 V94-5152 were developed for

selecting individuals carrying all three genes. Eight

PCR-based markers linked to the three SMV resis-

tance genes were used for marker-assisted selection.

Two SSR markers (Sat_154 and Satt510) and one

gene-specific marker (Rsv1-f/r) were used for select-

ing plants containing Rsv1; Satt560 and Satt063 for

Rsv3; and Satt266, AI856415, and AI856415-g for

Rsv4. Five F4:5 lines were homozygous for all eight

marker alleles and presumably carry all three SMV

resistance genes that would potentially provide

multiple and durable resistance to SMV.

Keywords Gene pyramiding �Marker-assisted selection � Resistance gene �Soybean � Soybean mosaic virus

Abbreviations

MAS Marker-assisted selection

MLG Molecular linkage group

PCR Polymerase chain reaction

PI Plant introduction

SMV Soybean mosaic virus

SSR Simple sequence repeat

Introduction

Soybean mosaic virus (SMV) is one of the most

destructive viral diseases in soybean [Glycine max

(L.) Merr.]. Pathogenic variability among SMV

A. Shi � P. Chen (&) � D. Li � C. Zheng �B. Zhang � A. Hou

Department of Crop, Soil, and Environmental Sciences,

University of Arkansas, Fayetteville, AR 72701, USA

e-mail: pchen@uark.edu

Present Address:A. Shi

Institute of Genomic Diversity, Cornell University,

Ithaca, NY 14853, USA

Present Address:D. Li

College of Agriculture, Northwest Agriculture

and Forestry University, Yangling, Shaanxi

712100, People’s Republic of China

Present Address:C. Zheng

Syngenta Seeds Inc, 3 Ultraway Dr, Highland,

IL 62249, USA

Present Address:A. Hou

Agriculture and Agri-Food Canada, Unit 100-101 Route

100, Morden, MB, Canada R6M 1Y5

123

Mol Breeding

DOI 10.1007/s11032-008-9219-x

isolates has been widely observed (Hill 1999).

Various SMV strain groups were reported in different

regions around the world, and the pathotypic rela-

tionships among these groups have not been

determined. There were 17 strains (SC1–SC17)

reported in China (Guo et al. 2005) and five strains

(A–E) reported in Japan (Takahashi et al. 1980). In

Korea, Choi et al. (2005) recently reported the

emergence of SMV isolates capable of overcoming

all of the known resistance genes. In the U.S., Cho

and Goodman (1979, 1982) classified 98 isolates

collected from soybean germplasm into seven strain

groups (G1–G7) and recently this classification

system was updated by Chen and Choi (2007)

(Table 1).

Various soybean germplasm have been screened

for SMV resistance, and inheritance studies have

shown that resistance is controlled, in most cases, by

a single dominant gene. Three independent loci,

Rsv1, Rsv3, and Rsv4, have been reported for SMV

resistance (Kiihl and Hartwig 1979; Buss et al. 1997,

1999). Rsv1 confers resistance to some strains from

G1 to G6, but not G7; Rsv3 conditions resistance to

G5 through G7, but not G1 through G4; Rsv4

provides resistance to all seven strains only at the

seedling stage, but gives a delayed susceptible

reaction at a later stage. Nine alleles have been

identified at the Rsv1 locus: Rsv1 in PI 96983, Rsv1-m

in ‘Marshall’, Rsv1-y in ‘York’, Rsv1-r in ‘Raiden’,

Rsv1-k in ‘Kwanggyo’, Rsv1-t in ‘Ogden’, Rsv1-h in

‘Suweon 97’, Rsv1-s in LR1, and Rsv1-n in PI

507389 (Chen and Choi 2007). Three alleles at Rsv3

locus have been reported in OX686, L29, and

‘Harosoy’ (Buzzell and Tu 1989; Buss et al. 1999;

Gunduz et al. 2001). In addition, HLS was assumed

to carry an allele at Rsv3 locus (Bowers et al. 1992),

but this assumption has not been confirmed. V94-

5152 and PI 88788 were shown to carry Rsv4

conferring resistance related to restriction of viral

movement in plant (Buss et al. 1997; Gunduz et al.

2004). ‘Peking’ displayed a latent infection to SMV

(Shigemori 1991) and carried an allele at the Rsv4

locus (Gunduz et al. 2004). Some soybean genotypes

that were resistant to G1 through G7 were reported to

carry two resistance genes, such as ‘Hourei’

(Rsv1Rsv3) (Gunduz et al. 2002), PI 486355

(Rsv1Rsv4) (Chen et al. 1993; Ma et al. 1995), and

‘Columbia’ (Rsv3Rsv4) (Ma et al. 2002). However,

soybean containing all three genes (Rsv1Rsv3Rsv4)

has not been found.

Rsv1 was mapped on soybean molecular linkage

group (MLG) F (Yu et al. 1994). Two RFLP markers,

pA 186 and pK 644a, and one SSR marker, SM176,

were found to be tightly linked to Rsv1 with distances

of 1.5, 2.1, and 0.5 cM, respectively (Yu et al. 1994).

A RAPD marker OPN11980/1070 and its derived

SCAR marker SCN11980/1070 were also identified to

be linked to Rsv1 with a distance of 3.03 cM (Zheng

et al. 2003). In another study, a high-resolution map

of the Rsv1 region was constructed with 38 loci by 24

markers including one RAPD, four SSRs, and 19

RFLPs; and Rsv1 was closely linked to the SSR

marker Satt510 (\2.4 cM) (Gore et al. 2002).

Recently, a PCR-based primer Rsv1-f/r was devel-

oped from the Rsv1 candidate gene 3gG2 (Shi et al.

2008). Rsv3 was flanked by A519F/R at 0.8–0.9 cM

Table 1 Differential reaction of soybean genotypes to seven SMV strains identified in the US

Genotype Gene (s) Reaction to SMV straina References

G1 G2 G3 G4 G5 G6 G7

Lee68/Essex rsv S S S S S S S Chen et al. (1991)

York Rsv1-y R R R N S S S Roane et al. (1983); Chen et al. (1991)

PI 96983 Rsv1 R R R R R R N Kiihl and Hartwig (1979)

PI 507389 Rsv1-n N N S S N N S Ma et al. (2003)

L29 Rsv3 S S S S R R R Buss et al. (1999)

V94-5152 Rsv4 ER ER ER ER ER ER ER Buss et al. (1997)

Hourei Rsv1Rsv3 R R R R R R R Gunduz et al. (2002)

PI 486355 Rsv1Rsv4 R R R R R R R Chen et al. (1993); Ma et al. (1995)

Columbia Rsv3Rsv4 R R R R R R R Ma et al. (2002)

a R = resistant (symptomless); ER = early resistance at seedling stage; N = necrosis (systemic necrosis); S = susceptible (mosaic)

Mol Breeding

123

and M3Satt at 0.8 cM, and mapped on MLG B2

(Jeong et al. 2002). Rsv4 was mapped on MLG D1b,

and flanked by two SSR markers, Satt542 at 4.7 cM

and Satt558 at 7.8 cM (Hayes et al. 2000). Two ESTs

(expressed sequence tags) markers, AI856415-g or -S

and BF070293-S, were mapped at 2.8 cM on one side

of Rsv4, and additional two EST markers,

AW307114A (3.3 cM) and AW471852A (2.4 cM),

were mapped on the other side of Rsv4 (Hwang et al.

2006). These identified markers may serve as a good

molecular tool for marker-assisted selection (MAS)

and gene pyramiding for SMV resistance.

The use of genetic resistance is the most effective

and economical way of controlling plant diseases.

Gene pyramiding is a practical approach to achieving

multiple and durable resistance (Schaffer and Roelfs

1985; Mundt 1990; Singh et al. 2001; Castro et al.

2003). Gene pyramiding has been successfully

applied in combining multiple genes not only for

qualitative disease resistance such as bacterial blight

resistance (Huang et al. 2004) and blast resistance

(Hittalmani et al. 2000) in rice, powdery mildew

resistance in wheat (Liu et al. 2000), but also for

quantitative resistance such as stripe rust resistance in

barley (Castro et al. 2003). In addition, Collier et al.

(1997) successfully applied a gene pyramiding strat-

egy for Hessian fly resistance in rice. However, there

is a lack of information on gene pyramiding using

MAS strategy for disease resistance in soybean. The

soybean 9 SMV interaction system has been well

studied at the molecular level and, therefore, is an

excellent system to implement MAS in pyramiding

multiple resistance genes.

Breeders frequently face complex choices in

designing efficient crosses and selection strategies

aimed at combining desired genes into a single target

genotype (Wang et al. 2007). It is difficult to select

plants with multiple resistance genes based on

phenotype alone due to the epistatics (Mohler and

Singrun 2004). MAS has been widely used in

selection for disease resistance by applying genetic

markers to identify and select specific genes or

combine multiple resistance genes (Brahm et al.

2000; Fjellstrom et al. 2004). For example, MAS for

resistance to soybean cyst nematode (SCN) has been

cost-effective compared with greenhouse screening

and has increased the efficiency and speed of

developing SCN resistant cultivars (Concibido et al.

2004).

Marker-assisted selection becomes very important

and useful in the effort of combing resistance genes

for SMV. Gene combinations, Rsv1 ? 3, Rsv1 ? 4,

Rsv3 ? 4, and Rsv1 ? 3 ? 4, can not be distin-

guished by host reactions to SMV strains G1–G7

because they all give the same resistance reaction to

all seven SMV strains. Pyramiding of all three genes

Rsv1, Rsv3, and Rsv4 can be accomplished through

MAS using tightly linked gene-specific molecular

markers. The objective of this research was to

pyramid Rsv1, Rsv3, and Rsv4 with the aid of

molecular markers in order to develop new soybean

lines with multiple genes for potentially durable

resistance to SMV.

Materials and methods

Plant materials and SMV tests

Two soybean lines, ‘J05’ and V94-5152, were used as

donor parents for pyramiding three SMV resistance

genes, Rsv1, Rsv3, and Rsv4. J05 is a Chinese cultivar

containing two genes (Rsv1 and Rsv3) and resistant to

seven SMV strains G1–G7 (Zheng et al. 2006). V94-

5152 was derived from PI 486355 9 Essex and

carries Rsv4 for seedling resistance to SMV strains

G1–G7 (Chen et al. 1993; Ma et al. 1995; Buss et al.

1997). Seed of J05 and V94-5152 were provided by

Dr. Glenn Buss of Virginia Polytechnic Institute and

State University and they were maintained in a cold

storage or by growing out in the field and harvesting

in bulk. J05 was crossed with V94-5152 and the F1

plants were grown in the field and harvested individ-

ually. An average of 200 F2 plants and 84 and 92

random F2:3 lines with adequate seed were inoculated

with G1 and G7 in the greenhouse, respectively

(Zheng et al. 2006). Forty-six F3 plants were selected

from eight F2:3 lines (five or six plants from each line)

that were homogeneously resistant to both G1 and G7

and were screened with molecular markers linked to

the three target genes. Those F3 plants that had the

three genes of interest in the homozygous or hetero-

zygous state were propagated for additional

generations and further confirmed for the presence

of the three resistance genes by molecular markers

and SMV reactions. The scheme for gene pyramiding

was shown in Fig. 1.

Mol Breeding

123

Progeny from each selected plant in each genera-

tion were grown and screened with SMV G1 and G7

under the greenhouse conditions (20–25�C tempera-

ture and 15 h photoperiod) at Harry R. Rosen

Alternative Pest Control Center of University of

Arkansas, Fayetteville, AR. Virus inoculum was

prepared by grinding systemically infected leaves

from virus-maintaining stock plants with a mortar and

pestle in 0.05 M potassium phosphate buffer at an

approximate dilution of 1:10 (w/v) at pH 7.2. Inoc-

ulation was performed by rubbing the inoculum with a

pestle onto both unifoliolate leaves that had been

previously dusted with carborundum (Chen et al.

1994; Zheng et al. 2005). SMV strains G1 and G7

were kindly provided by Dr. Sue Tolin of Virginia

Polytechnic Institute and State University. Virus

strain identity was verified by inoculation of a set of

differential genotypes consisting of ‘Essex’ (rsv), PI

96983 (Rsv1), PI 507389 (Rsv1-n), ‘York’ (Rsv1-y),

L29 (Rsv3), and V94-5152 (Rsv4) (Table 1) (Chen

et al. 1991; Gunduz et al. 2001; Ma et al. 2003). Plants

of each genotype were monitored for SMV symptoms

on a regular basis and were classified as resistant (R,

symptomless), necrotic (N, stem-tip necrosis), or

susceptible (S, mosaic).

Marker selection and PCR-based assay

A total of 33 molecular markers surrounding three

SMV resistance loci (12 for Rsv1, 7 for Rsv3, and

14 for Rsv4) were selected to screen J05 and V94-

5152 (Table 2). The polymorphic markers between

the two parents (Fig. 2) were further used to trace

SMV resistance genes in each plant from the

selected F2:3 lines and their advanced generations.

Among the 12 selected markers around the Rsv1

locus on MLG F, HSP176L, which was later

renamed as SOYHSP176 on the soybean composite

genetic map (Song et al. 2004), was selected as it

was first found to be tightly linked to Rsv1 (Yu

et al. 1996). Satt510 and Sat_120 with a genetic

distance of 2.4 and 3.8 cM to the Rsv1 locus,

Fig. 1 Procedure of

pyramiding Rsv1, Rsv3, and

Rsv4 genes for SMV

resistance

Mol Breeding

123

respectively, were also selected according to Gore

et al. (2002). A new PCR-based marker (Rsv1-f/r),

recently developed by Shi et al. (2008) based on the

sequence of the candidate gene 3gG2 at the Rsv1

locus (Hayes et al. 2004), was used to differentiate

individuals with an amplified 341-bp DNA fragment

Table 2 PCR-based

markers and their position

in relation to three SMV

resistance loci on soybean

linkage map

a The Rsv1 locus is flanked

by Sat_154 and Satt510,

and closely linked to

SOYHSP176 (0.5 cM apart)

(Yu et al. 1996); Rsv3 is

located between A519 and

M3Satt with 0.8–0.9 cM

(Jeong et al. 2002); Rsv4 is

flanked by AW471852R

(2.4 cM apart) and Satt634

(2.2 cM apart)b Sat_254 is mapped using

an F2 population from

‘Sowonkong’ 9 V94-5152

by Hwang et al. (2006), but

located at a different

location on the integrated

map (Song et al. 2004)

Marker Linkage

group

Position

(cM)

Resistance

locusaReferences

Sat_297 F 59.6 Rsv1 Song et al. (2004)

Sat_229 F 62.8 Rsv1 Song et al. (2004)

Satt114 F 63.7 Rsv1 Song et al. (2004);

Shi et al. (2008)

Sat_234 F 66.6 Rsv1 Song et al. (2004)

SOYHSP176 F 68.4 Rsv1 Yu et al. (1996);

Song et al. (2004)

Sat_154 F 68.9 Rsv1 Song et al. (2004)

Rsv1-f/r F 69.1 Rsv1 Shi et al. (2008)

Satt510 F 71.4 Rsv1 Gore et al. (2002);

Song et al. (2004)

Sat_317 F 73.0 Rsv1 Song et al. (2004)

Sct_103 F 74.1 Rsv1 Song et al. (2004)

Sat_120 F 76.0 Rsv1 Gore et al. (2002);

Song et al. (2004)

Satt334 F 78.1 Rsv1 Song et al. (2004)

Satt063 B2 93.5 Rsv3 Jeong et al. (2002);

Song et al. (2004)

A519 B2 96.7 Rsv3 Jeong et al. (2002)

M3Satt B2 97.5 Rsv3 Jeong et al. (2002)

Satt560 B2 97.9 Rsv3 Song et al. (2004)

Sat_424 B2 100.1 Rsv3 Song et al. (2004)

Satt726 B2 100.6 Rsv3 Song et al. (2004)

Satt687 B2 113.6 Rsv3 Song et al. (2004)

Satt558 D1b 43.9 Rsv4 Hayes et al. (2000);

Song et al. (2004)

Sat_254b D1b 46.9b Rsv4 Hwang et al. (2006)

BF070293-S D1b 46 Rsv4 Hwang et al. (2006)

AI856415-g D1b 46 Rsv4 Hwang et al. (2006)

AI856415-S D1b 46 Rsv4 Hwang et al. (2006)

BI470504 D1b 46.5 Rsv4 Song et al. (2004);

Hwang et al. (2006)

Satt634 D1b 46.6 Rsv4 Song et al. (2004);

Hwang et al. (2006)

BF070293 D1b 47.3 Rsv4 Song et al. (2004);

Hwang et al. (2006)

AI856415 D1b 50.1 Rsv4 Song et al. (2004)

AW417852R D1b 51.2 Rsv4 Hwang et al. (2006)

AW307114A D1b 52.1 Rsv4 Hwang et al. (2006)

Satt296 D1b 52.6 Rsv4 Song et al. (2004)

Satt542 D1b 53.0 Rsv4 Hayes et al. (2000); Song et al. (2004)

Satt266 D1b 59.6 Rsv4 Hayes et al. (2000); Song et al. (2004);

Hwang et al. (2006)

Mol Breeding

123

specific for Rsv1. Another marker used for Rsv1

screening was Satt114, which was linked to the

PCR-based marker Rsv1-f/r with a distance of

5.42 cM (Shi et al. 2008). In addition, seven SSR

markers on MLG F were selected from the region of

18.5 cM spanning both sides of the Rsv1 locus

according to soybean genetic composite map to

ensure polymorphism (Song et al. 2004). Among the

seven markers surrounding the Rsv3 locus on MLG

B2, Satt063, A519 and M3Satt were selected based

on their close linkage (3.2–6.3 cM, 0.8–0.9 cM, and

0.8 cM, respectively) to Rsv3 (Jeong et al. 2002).

Additional four SSR markers, Satt560, Sat_424,

Satt726, and Satt687, on MLG B2 near the Rsv3

locus within a region of 15.7 cM were also selected

from the soybean composite genetic map to ensure

polymorphism (Song et al. 2004) (Table 2). Among

the 14 selected markers for Rsv4 on MLG D1b, two

flanking markers, Satt542 (4.7 cM) and Satt558

(7.8 cM), were selected according to Hayes et al.

(2000). Ten additional markers linked to Rsv4 on

MLG D1b including Sat_254, BF070293-S,

AI856415-g, AI856415-S, BI470504, Satt634,

BF070293, AW417852R, AW307114A, and Satt266

were selected based on the report by Hwang et al.

(2006). Also included in the maker selection for

Rsv4 were AI856415 and Satt296 selected from the

soybean composite genetic map to ensure polymor-

phism (Song et al. 2004).

Genomic DNA was extracted from fresh leaves of

greenhouse-grown plants using the CTAB (hexade-

cyltrimethyl ammonium bromide) method (Kisha

et al. 1997). PCR amplification was performed in

an iCycler Thermal Cycler (Bio-Rad Laboratories

Inc., Hercules, CA) following standard PCR proce-

dures with minor modifications. Briefly, each 50 ll

PCR reaction mixture consisted of 36 ll sterilized

ddH2O, 5 ll 109 PCR buffer (Promega, Madison,

WI), 3 ll MgCl2 (25 mM), 1.5 ll dNTP (2.5 mM),

1.5 ll each primer (20 ng/ll), 0.2 ll Taq polymerase

(Promega, Madison, WI) (5 U/ll), and 1.3 ll tem-

plate DNA (20 ng/ll). PCR procedure consisted of an

initial denaturation step at 94�C for 5 min, 38 cycles

of 45 s at 94�C, 45 s at 45–55�C depending on the

proper primer Tm (52�C for Rsv1-f/r), and 1 min at

72�C followed by an extension step at 72�C for 5 min

and a 4�C soak. The PCR products were separated on

6% non-denaturing polyacrylamide gel or 1.0–1.5%

agarose gel in 0.59 TBE, and visualized by staining

with ethidium bromide.

Results and discussion

Specific molecular markers and polymorphism

Eight of the 33 markers screened, Satt510, Sat_154,

and Rsv1-f/r around Rsv1 locus, Satt063 and Satt560

around Rsv3, as well as Satt266, AI856415, and

AI856415-g around Rsv4, showed polymorphism

between J05 and V94-5152. Since Satt510 and

Sat_154 did not amplify DNA fragment associated

with the Rsv1 allele in J05, multiple primers

Satt510 ? Sat_317 and Sat_154 ? Sat_317 were

designed to detect the Rsv1 gene (Table 3; Fig. 2).

Sat_317 amplified a same size DNA fragment

between J05 and V94-5152 (Table 2). Therefore, if

only one band was amplified by the two-primer

combination in a plant, the plant should contain the

homozygous alleles Rsv1Rsv1; if two bands were

amplified in a plant, the plant should contain either

homozygous alleles rsv1rsv1 or heterozygous alleles

Rsv1rsv1. J05 showed only one band amplified from

the two-primer combination whereas V94-5152

showed two bands, one of which was amplified from

Fig. 2 DNA fragments amplified by eight molecular markers

in soybean line J05 (lanes 1, 3, 5, 7, 9, 11, 13, and 15) and V94-

5152 (lanes 2, 4, 6, 8, 10, 12, 14 and 16): Satt510 ? Sat_317

(lanes 1 and 2), Rsv1-f/r (lanes 3 and 4, no band in lane 4 due

to the lack of Rsv1, and Sat_154 ? Sat_317 (lanes 5 and 6) for

Rsv1; Satt063 (lanes 7 and 8) and Satt560 (lanes 9 and 10) for

Rsv3; Satt266 (lanes 11 and 12), AI856415-g (lanes 13 and 14),

and AI856415 (lanes 15 and 16) for Rsv4. Lanes 15 and 16

were attached from a different gel where a better separation

was achieved. Lane M is a 100 bp DNA molecular ladder

Mol Breeding

123

Sat_317 and the other from Satt510 or Sat_154

(Table 2; Fig. 2). Rsv1-f/r was a dominant PCR-

based marker and, therefore, could easily detect the

presence of the Rsv1 candidate gene 3Gg2 (Shi et al.

2008). As expected, Rsv1-f/r gave rise to the present

of one band in J05 but absence of the band in V94-

5152 (Table 3; Fig. 2). Therefore, Rsv1-f/r was a

very effective marker for detecting the presence of

Rsv1. However, Rsv1-f/r could not distinguish

between homozygous alleles Rsv1Rsv1 and hetero-

zygous alleles Rsv1rsv1. With the use of two-primer

combinations Satt510 ? Sat_317 and Sat_154 ?

Sat_317, it was possible to identify plants containing

the Rsv1 gene in homozygous state (Rsv1Rsv1).

Satt063 and Satt560 gave rise to different bands

between J05 and V94-5152 and served as excellent

markers for detecting the Rsv3 gene (Table 3; Fig. 2).

The Rsv3 locus was 3.0-6.0 cM from Satt063 and less

than 2 cM from Satt560 on MLG B2 (Jeong et al.

2002; Song et al. 2004). Although no flanking

markers for the Rsv3 gene were available, Satt063

and Satt560 were efficient in identifying plants

carrying Rsv3 because of the close linkage. The

effectiveness of Satt063 and Satt560 for Rsv3 selec-

tion was confirmed by SMV inoculation in single

gene situations where the presence of Rsv3 gave rise

to resistance to G7 and the absence of Rsv3 was

associated with susceptibility to G7.

Satt266, AI856415, and AI856415-g gave rise to

different bands between J05 and V94-5152 (Table 3;

Fig. 2), therefore, proved useful for detecting specific

Rsv genes. Rsv4 in V94-5152 is flanked by AI856415

(AI856415-g) and Satt266, and AI856415 is very

closely linked to the Rsv4 gene (Hayes et al. 2000;

Song et al. 2004; Hwang et al. 2006). The combina-

tion of these three markers was efficient to identify

plants carrying the Rsv4 gene. With all eight poly-

morphic markers, we were able to find plants derived

from J05 9 V94-5152 that contain all three genes in

the homozygous state (Rsv1Rsv1Rsv3Rsv3Rsv4Rsv4).

Such plants produced, as expected, the same DNA

fragments as J05 (Rsv1Rsv3) amplified by

Satt510 ? Sat_317, Sat_154 ? Sat_317, Rsv1-f/r,

Satt063, and Satt560, and the same fragments as

V94-5152 (Rsv4) when amplified by Satt266,

AI856415, and AI856415-g (Figs. 3–5).

Gene pyramiding scheme and marker-assisted

selection

The procedures for pyramiding the three genes, Rsv1,

Rsv3, and Rsv4, were shown in Fig. 1. Based on SMV

reactions of 84 and 92 F2:3 lines to G1 and G7,

respectively (Zheng et al. 2006), eight F2:3 lines

resistant to G1 and G7 were selected and presumed to

Table 3 Scheme of gene pyramiding strategy using molecu-

lar-assisted selection and polymorphism between J05 and V94-

5152 detected by eight selected markers linked to three SMV

resistance genes

Marker Linked

gene

Banding patterna

J05

(Rsv1 ? 3)

V94-5152

(Rsv4)

Sat_317 Rsv1 1 1

Satt510 ? Sat_317 Rsv1 0 ? 1 2 ? 1

Sat_154 ? Sat_317 Rsv1 0 ? 1 2 ? 1

Rsv1-f/r Rsv1 1 0

Satt063 Rsv3 1 2

Satt560 Rsv3 1 2

Satt266 Rsv4 1 2

AI856415 Rsv4 1 2

AI856415-g Rsv4 1 2

a 0 = absence of the DNA fragment; 1 = presence of a band

in J05; 2 = presence of a band in V94-5152; 0 ? 1 = presence

of one band in J05 amplified from primer Sat_317, but absence

of a band amplified from the other primer; 2 ? 1 = presence of

two bands amplified from both primers

Fig. 4 DNA fragments amplified from Sat_154 ? Sat_317

(Rsv1) in 37 F3 plants derived from J05 9 V94-5152 and two

parents J05 (lane left 2) and V94-5152 (lane left 3). Lane left 1

is a 100 bp molecular ladder. The upper fragment was

amplified from Sat_317 and lower fragment was amplified

from Sat_154 (Rsv1)

Fig. 3 DNA fragments amplified by Satt510 ? Sat_317 in 36

F3 plants derived from J05 9 V94-5152 and two parents J05

(lane left 3) and V94-5152 (lane left 2). Lane left 1 is a 25 bp

molecular ladder. The upper fragment was amplified from

Sat_317 and lower fragment was from Satt510 (Rsv1)

Mol Breeding

123

carry Rsv1 ? 3, Rsv1 ? 4, Rsv3 ? 4, or Rsv1 ?

3? 4 gene combinations. Forty-six F3 plants from

these eight F2:3 lines were grown in the greenhouse

and were genotyped by using the eight selected

molecular markers specific for Rsv1, Rsv3, or Rsv4.

None of the eight F2:3 lines were shown to contain

homozygous alleles for all eight markers based on the

molecular data from the 46 F3 plants genotyped

(Table 4), but some lines contained homozygous

alleles at a single locus. For example, JV-06G3-5 was

homozygous for Satt063 and Satt266 and presumably

contained homozygous alleles Rsv3Rsv3; JV-06G 3-7

and JV-06G 3-8 were homozygous for Satt266,

AI856415-g, and therefore, contained homozygous

Rsv4Rsv4 alleles. Plants that contained one or two

genes of interest based on the marker data were

eliminated from the MAS process, some of which

might have been resulted from possible double

recombinations between the marker and the resis-

tance gene.

Five out of 46 F3 plants were selected to produce

next generation as the candidates carried all three

genes, Rsv1, Rsv3, abd Rsv4, based on genotypic data

from all eight markers (Table 5). Plants JV-06G3-1-1

and JV-06G3-3-1 were amplified by Rsv1-r/f, indi-

cating that these two plants contained the Rsv1 gene.

But they showed two bands when amplified by the

two-primer combinations, indicating these plants

were heterozygous at Rsv1 locus. However, these

plants contained homozygous alleles for Rsv3-linked

markers (Satt063 and Satt560) and Rsv4-linked

markers (Satt266, AI856415, and AI856415-g).

Fig. 5 DNA fragments

amplified from a Satt063

(Rsv3), b AI856415 (Rsv4),

c Satt266 (Rsv4), and d

Satt560 (Rsv1) in 30 F4

plants derived from

J05 9 V94-5152 and two

parents J05 (right lane 2)

and V94-5152 (right lane 1)

Table 4 Segregation of eight molecular markers for SMV resistance genes in selected plants representing eight F2:3 lines derived

from J05 9 V94-5152

F2:3 line Rsv1 locus Rsv3 locus Rsv4 locus

Satt510

? Sat_317

Sat_154

? Sat_317

Rsv-f/r Satt063 Satt560 Satt266 AI856415 AI856415-g

JV-06G3-1 Sa S S S S S Ho-V94 Ho-V94

JV-06G3-2 S S S S S S S S

JV-06G3-3 S S S S S S S S

JV-06G3-4 S S S S S S S S

JV-06G3-5 S S S Ho-J05 Ho-J05 S S S

JV-06G3-6 S S S S S S S S

JV-06G3-7 S S S S S Ho-V94 Ho-V94 Ho-V94

JV-06G3-8 S S S S S Ho-V94 Ho-V94 Ho-V94

a S = segregating for the marker; Ho-J05 = homozygous for the marker allele and same size of band as present in J05 for Rsv1 or

Rsv3; Ho-V94 = homozygous for the marker allele and same band as present in V94-5152 for Rsv4

Mol Breeding

123

Ta

ble

5P

oly

mo

rph

ism

of

eig

ht

mo

lecu

lar

mar

ker

sin

sele

cted

F3

pla

nts

and

sele

ctio

no

fp

rog

enie

sw

ith

targ

eted

SM

Vre

sist

ance

gen

esd

eriv

edfr

om

J05

9V

94

-51

52

Gen

erat

ion

Sel

ecte

dp

lan

tsR

sv1

Rsv

3R

sv4

Ex

pec

ted

gen

oty

pe

Sat

t51

0?

Sat

_3

17

Sat

_1

54

?

Sat

_3

17

Rsv

1-f

/rS

att0

63

Sat

t56

0S

att2

66

AI8

56

41

5A

I85

64

15

-g

F3

JV-0

6G

3-1

-11

?2

a1

?2

11

12

22

Rsv

1rs

v1R

sv3

Rsv

3R

sv4

Rsv

4

JV-0

6G

3-3

-11

?2

1?

21

11

22

2R

sv1

rsv1

Rsv

3R

sv3

Rsv

4R

sv4

JV-0

6G

3-5

-11

11

1?

21

22

2R

sv1

Rsv

1R

sv3

_R

sv4

Rsv

4

JV-0

6G

3-7

-11

11

11

1?

22

2R

sv1

Rsv

1R

sv3

Rsv

3R

sv4

_

JV-0

6G

3-8

-11

11

11

22

2R

sv1

Rsv

1R

sv3

Rsv

3R

sv4

Rsv

4

F4

JV-0

6G

3-5

-1-3

11

11

12

22

Rsv

1R

sv1

Rsv

3R

sv3

Rsv

4R

sv4

JV-0

6G

3-8

-1-1

11

11

12

22

Rsv

1R

sv1

Rsv

3R

sv3

Rsv

4R

sv4

JV-0

6G

3-8

-1-5

11

11

12

22

Rsv

1R

sv1

Rsv

3R

sv3

Rsv

4R

sv4

F5

JV-0

6G

3-5

-1-3

-11

11

11

22

2R

sv1

Rsv

1R

sv3

Rsv

3R

sv4

Rsv

4

JV-0

6G

3-8

-1-1

-11

11

11

22

2R

sv1

Rsv

1R

sv3

Rsv

3R

sv4

Rsv

4

JV-0

6G

3-8

-1-1

-31

11

11

22

2R

sv1

Rsv

1R

sv3

Rsv

3R

sv4

Rsv

4

JV-0

6G

3-8

-1-5

-11

11

11

22

2R

sv1

Rsv

1R

sv3

Rsv

3R

sv4

Rsv

4

JV-0

6G

3-8

-1-5

-51

11

11

22

2R

sv1

Rsv

1R

sv3

Rsv

3R

sv4

Rsv

4

Par

ents

J05

11

11

11

11

Rsv

1R

sv1

Rsv

3R

sv3

V9

4-5

15

22

22

22

22

2R

sv4

Rsv

4

a1

=p

rese

nce

of

ab

and

inJ0

5;

2=

pre

sen

ceo

fa

ban

din

V9

4-5

15

2;

1?

2=

pre

sen

ceo

ftw

ob

and

sam

pli

fied

fro

mb

oth

pri

mer

s

Mol Breeding

123

Therefore, these two plants were presumed to carry

Rsv1rsv1Rsv3Rsv3Rsv4Rsv4 based on the linked SSR

markers. JV-06G3-5-1 was amplified by Rsv1-r/f and

contained heterozygous alleles for Satt063 and

homozygous alleles for other six markers, indicating

that this plant carries Rsv1Rsv1Rsv3rsv3Rsv4Rsv4.

The heterozygosity for the Satt063 allele in JV-06G3-

5-1 was likely resulted from a crossover between the

marker Satt063 and Rsv3. JV-06G3-7-1 was ampli-

fied by Rsv1-r/f and contained heterozygous alleles

for Satt266, and homozygous alleles for other six

markers, indicating that this plant carried Rsv1Rsv1

Rsv3Rsv3Rsv4rsv4. The heterozygous alleles for the

Satt266 were probably a result of recombination

between the marker Satt266 and Rsv4. JV-06G3-8-1

was amplified by Rsv1-f/r and contained homozygous

alleles for other seven markers, indicating that this

plant carries Rsv1Rsv1Rsv3Rsv3Rsv4Rsv4.

Five out of 46 F3 plants representing five F2:3 lines

were selected as potentially having the targeted genes

(Table 5) to produce F4 generation. In the F4

generation, 30 F4 plants representing five original

F2:3 lines were screened with eight selected markers.

Three plants (JV-06G3-5-1-3, JV-06G3-8-1-1, and

JV-06G3-8-1-5) consistently showed homozygous

alleles for all selected markers, indicating the pres-

ence of all three SMV resistance genes, Rsv1, Rsv3,

and Rsv4 in these plants. Subsequently, progeny from

the three selected F4 plants did not segregate for the

three SMV resistance loci based on the marker data.

Similarly, five F5 plants were selected as having

homozygous alleles for all eight markers. As a

phenotypic confirmation, SMV reactions to G1 and

G7 were evaluated for all selected lines at each

generation. All selected lines of each generation

showed resistance to both G1 and G7 (Table 6). The

advanced lines representing two original F2:3 lines

will be evaluated for yield and other agronomic traits

in the field and genetic confirmation of stacked

resistance genes is under way.

The significance of this research includes two

important implications: the feasibility of pyramiding

multiple genes using MAS and the possibility of

durable resistance for ever-changing SMV strains.

Selection for plants having multiple genes by clas-

sical breeding approach is extremely difficult and

nearly impossible in the case of SMV resistance

based on phenotype. The MAS approach using PCR-

based markers used in this study has made it possible

to stack-up three SMV resistance genes. The multiple

genes will play an important role in preventing

breaking-down of genetic resistance due to occur-

rence of new virulent strains.

Resistance breaking (RB) cases of plant disease

frequently occur in nature, especially, when the

resistance to a specific disease is conditioned by

single genes. New RB SMV isolates that infected

soybean lines possessing the single gene at Rsv1,

Rsv3 or Rsv4 locus were recently reported in Korea

(Choi et al. 2005). However, cultivars with three

resistance genes may have a broad and durable

resistance to the changing SMV isolates. In this study

using MAS strategy, soybean lines with all three

SMV resistance genes in homozygous state

(Rsv1Rsv1Rsv3Rsv3Rsv4Rsv4) have been developed

from the cross of J05 (Rsv1Rsv3) 9 V94-5152

(Rsv4). These lines, once released, may serve as

valuable germplasm for breeder to use a parents in

Table 6 Summary of marker genotyping and SMV phenotyping of selected plants of lines derived from J05 9 V94-5152 through

different generations

Phenotyping progeny lines with SMV inoculation Genotyping selected individual plant with eight molecular markers

Generation No. of lines SMV reactiona Generation No. of plantsb No. of plants with homozygous alleles at locus

G1 G7 Total Selected Rsv1 Rsv3 Rsv4

F2:3 84–92 40R 39R F2:3 – 8 lines 0 1 2

F2:3 8 8R 8R F3 46 5 3 4 4

F3:4 5 5R 5R F4 30 3 3 3 3

F4:5 3 3R 3R F5 46 5 5 5 5

F5:6 5 5R 5R

a No. of lines that are homogeneously resistantb No. of plants that are homozygous for the gene

Mol Breeding

123

breeding programs in which broad and durable

resistance to SMV is an objective.

References

Bowers GR Jr, Paschal EHII, Bernard RL, Goodman RM

(1992) Inheritance of resistance to soybean mosaic virus

in ‘Buffalo’ and HLS soybean. Crop Sci 32:67–72

Brahm L, Rocher T, Friedt W (2000) PCR-based markers

facilitating marker assisted selection in sunflower for

resistance to downy mildew. Crop Sci 40:676–682

Buss GR, Ma G, Chen P, Tolin SA (1997) Registration of V94–

5152 soybean germplasm resistant to soybean mosaic

potyvirus. Crop Sci 37:1987–1988

Buss GR, Ma G, Kristipati S, Chen P, Tolin SA (1999) A new

allele at the Rsv3 locus for resistance to soybean mosaic

virus. In: World Soybean Res Conf VI, Chicago, IL, 4–7

Aug, Proceedings 490

Buzzell RI, Tu JC (1989) Inheritance of a soybean stem-tip N

reaction to soybean mosaic virus. J Hered 80:400–401

Castro AJ, Chen X, Hayes PM, Johnston M (2003) Pyramiding

quantitative trait locus (QTL) alleles determining resis-

tance to barley stripe rust: effects on resistance at the

seedling stage. Crop Sci 43:651–659

Chen P, Choi CW (2007) Characterization of genetic interac-

tion between soybean and soybean mosaic virus. In: Rao

GP (ed) Molecular diagnosis of plant viruses. Studium

Press, Houston, TX, pp 389–422

Chen P, Buss GR, Roane CW, Tolin SA (1991) Allelism

among genes for resistance to soybean mosaic virus in

strain differential soybean cultivars. Crop Sci 31:305–309

Chen P, Buss GR, Tolin SA (1993) Resistance to soybean

mosaic virus conferred by two independent dominant

genes in PI 486355. J Hered 84:25–28

Chen P, Buss GR, Roane CW, Tolin SA (1994) Inheritance in

soybean of resistant and necrotic reactions to soybean

mosaic virus strains. Crop Sci 34:414–422

Cho EK, Goodman RM (1979) Strains of soybean mosaic

virus: classification based on virulence in resistant soy-

bean cultivars. Phytopathology 69:467–470

Cho EK, Goodman RM (1982) Evaluation of resistance in

soybeans to soybean mosaic virus strains. Crop Sci 22:

1133–1136

Choi BK, Koo JM, Ahn HJ, Yum HJ, Choi CW, Ryu RH et al

(2005) Emergence of Rsv-resistance breaking soybean

mosaic virus isolates from Korean soybean cultivars.

Virus Res 112:42–51. doi:10.1016/j.virusres.2005.03.020

Concibido VC, Diers BW, Arelli P (2004) A decade of QTL

mapping for cyst nematode resistance in soybean. Crop

Sci 44:1121–1131

Collier CC, Ohm HW, Jones MJ, Williams CE (1997)

Molecular markers to track the pyramiding of several

Hessian fly resistance genes into a single line of common

wheat. In: Plant and animal genome V, 18–22 Jan, San

Diego, CA 92

Fjellstrom R, Conaway-Bormans CA, McClung AM, Marchetti

MA, Shank AR, Park WD (2004) Development of DNA

markers suitable for marker assisted selection of three Pi

genes conferring resistance to multiple Pyricularia griseapathotypes. Crop Sci 44:1790–1798

Gore MA, Hayes AJ, Jeong SC, Yu YG, Buss GR, Saghai

Maroof MA (2002) Mapping tightly linked genes con-

trolling potyvirus infection at the Rsv1 and Rpv1 region in

soybean. Genome 45:592–599. doi:10.1139/g02-009

Gunduz I, Buss GR, Chen P, Tolin SA (2001) Genetic analysis

of soybean mosaic virus in OX 670 and Harosoy soybean.

Crop Sci 41:1785–1791

Gunduz I, Buss GR, Chen P, Tolin SA (2002) Characterization

of genes in Tousan 140 and Hourei soybean for resistance

to soybean mosaic virus. Crop Sci 42:90–95

Gunduz I, Buss GR, Chen P, Tolin SA (2004) Genetic and

phenotypic analysis of soybean mosaic virus resistance in

PI 88788 soybean. Phytopathology 94:678–692. doi:

10.1094/PHYTO.2004.94.7.687

Guo DQ, Zhi HJ, Wang YW, Gai JY, Zhou XA, Yang CL

(2005) Identification and distribution of strains of soybean

mosaic virus in middle and northern of Huang Huai

Region of China. Soybean Sci 27:64–68

Hayes AJ, Ma G, Buss GR, Saghai Maroof MA (2000)

Molecular marker mapping of Rsv4, a gene conferring

resistance to all known strains of soybean mosaic virus.

Crop Sci 40:1434–1437

Hayes AJ, Jeong SC, Gore MA, Yu YG, Buss GR, Tolin SA,

Saghai Maroof MA (2004) Recombination within a

nucleotide-binding-site/leucine-rich-repeat gene cluster

produces new variants conditioning resistance to soybean

mosaic virus in soybeans. Genetics 166:493–503

Hill JH (1999) Soybean mosaic. In: Hartman GL, Sinclair JB,

Rupe JC (eds) Compendium of soybean diseases, 4th edn.

American Phytopathology Society Press, St. Paul, pp 70–71

Hittalmani S, Parco A, Mew TV, Zeigler RS, Huang N (2000)

Fine mapping and DNA marker-assisted pyramiding of

the three major genes for blast resistance in rice. Theor

Appl Genet 100:1121–1128. doi:10.1007/s001220051395

Huang N, Angeles ER, Domingo J, Magpantay G, Singh S,

Zhang G et al (2004) Pyramiding of bacterial blight

resistance genes in rice: marker-assisted selection using

RFLP and PCR. Theor Appl Genet 95:313–320. doi:

10.1007/s001220050565

Hwang TY, Yu S, Yang K, Kim HM, Jeong SC (2006)

Application of comparative genomics in developing

molecular markers tightly linked to the virus resistance

gene Rsv4 in soybean. Genome 49:380–388. doi:10.1139/

G05-111

Jeong SC, Kristipati S, Hayes AJ, Maughan PJ, Noffsinger SL,

Gunduz I et al (2002) Genetic and sequence analysis of

markers tightly linked to the soybean mosaic virus resis-

tance gene, Rsv3. Crop Sci 42:265–270

Kisha T, Sneller CH, Diers BW (1997) Relationship between

genetic distance among parents and genetic variance in

populations of soybean. Crop Sci 37:1317–1325

Kiihl RAS, Hartwig EE (1979) Inheritance of reaction to

soybean mosaic virus in soybeans. Crop Sci 19:372–375

Liu J, Liu D, Tao W, Li W, Wang S, Chen P et al (2000)

Molecular marker-facilitated pyramiding of different

genes for powdery mildew resistance in wheat. Plant

Breed 119:21–24. doi:10.1046/j.1439-0523.2000.00431.x

Ma G, Chen P, Buss GR, Tolin SA (1995) Genetic character-

istics of two genes for resistance to soybean mosaic virus

Mol Breeding

123

in PI 486355 soybean. Theor Appl Genet 91:907–914.

doi:10.1007/BF00223899

Ma G, Chen P, Buss GR, Tolin SA (2002) Complementary

action of two independent dominant genes in Columbia

soybean for resistance to soybean mosaic virus. J Hered

93:179–184. doi:10.1093/jhered/93.3.179

Ma G, Chen P, Buss GR, Tolin SA (2003) Genetic study of a

lethal necrosis to soybean mosaic virus in PI 507389

soybean. J Hered 94:205–211. doi:10.1093/jhered/esg059

Mohler V, Singrun C (2004) General considerations: marker-

assisted selection. In: L}orz H, Wenzel G (eds) Molecular

marker systems in plant breeding and crop improvement

(biotechnology in agriculture and forestry), vol 55.

Springer, Berlin, pp 305–317

Mundt CC (1990) Probability of mutation to multiple virulence

and durability of resistance gene pyramids. Phytopathol-

ogy 80:221–223

Roane CW, Tolin SA, Buss GR (1983) Inheritance of reaction

to two viruses in the soybean cross ‘York’ 9 ‘Lee68’.

J Hered 74:289–291

Schaffer JF, Roelfs AP (1985) Estimated relationship between

numbers of urediniospores of Puccinia graminis f. sp.

tritici and rates of occurrence of virulence. Phytopathol-

ogy 75:749–750

Shi A, Chen P, Zheng C, Hou A, Zhang B (2008) A PCR-based

marker for the Rsv1 locus conferring resistance to soybean

mosaic virus. Crop Sci 48:262–268. doi:10.2135/cropsci

2007.02.0076

Shigemori I (1991) Studies on the breeding on soybeans for the

resistance to soybean mosaic virus (SMV). Bull Nagano

Chusin Agric Exp Stn 10:1–60

Singh S, Sidhu JS, Huang N, Vikal NY, Li Z, Brar DS et al

(2001) Pyramiding three bacterial blight resistance genes

(xa5, xa13, Xa21) using marker-assisted selection into

indica rice cultivar PR106. Theor Appl Genet 102:1011–

1015. doi:10.1007/s001220000495

Song QJ, Marek LF, Shoemaker RC, Lark KG, Concibido VC,

Delannay X et al (2004) A new integrated genetic linkage

map of the soybean. Theor Appl Genet 109:122–128. doi:

10.1007/s00122-004-1602-3

Takahashi KT, Tanaka T, Iida W, Tsuda Y (1980) Studies on

virus diseases and causal viruses of soybean in Japan. Bull

Tohoku Natl Agric Exp Stn 62:1–130

Wang J, Chapmanb SC, Bonnettc DG, Rebetzkec GJ, Croucha

J (2007) Application of population genetic theory and

simulation models to efficiently pyramid multiple genes

via marker-assisted selection. Crop Sci 47:582–588

Yu YG, Saghai Maroof MA, Buss GR, Maughan PJ, Tolin SA

(1994) RFLP and microsatellite mapping of a gene for

soybean mosaic virus resistance. Phytopathology 84:60–

64. doi:10.1094/Phyto-84-60

Yu YG, Saghai Maroof MA, Buss GR (1996) Divergence and

allelomorphic relationship of a soybean virus resistance

gene based on tightly linked DNA microsatellite and

RFLP markers. Theor Appl Genet 92:64–69. doi:10.1007/

BF00222952

Zheng C, Chang R, Qiu L, Chen P, Wu X, Chen S (2003)

Identification and characterization of a RAPD/SCAR

marker linked to a resistance gene for soybean mosaic

virus in soybean. Euphytica 132:199–210. doi:10.1023/

A:1024682415598

Zheng C, Chen P, Gergerich R (2005) Characterization of

resistance to soybean mosaic virus in diverse soybean

germplasm. Crop Sci 45:2503–2509. doi:10.2135/cropsci

2005.0114

Zheng C, Chen P, Gergerich R (2006) Genetic analysis of

resistance to soybean mosaic virus in J05 soybean.

J Hered 97:429–437. doi:10.1093/jhered/esl024

Mol Breeding

123