SEED PHYSIOLOGY, PRODUCTION & TECHNOLOGY

Published January, 2002

SEED PHYSIOLOGY, PRODUCTION & TECHNOLOGY Soybean mosaic virus (SMV) and the SMV Resistance Gene (Rsv1 ): Influence on Phomopsis spp. Seed Infection in an Aphid Free Environment Gwen Koning, Dennis M. TeKrony,* Said A. Ghabrial, and Todd W. Pfeiffer ABSTRACT

caused up to 91% of the seed to show seedcoat mottling (Koning et al., 2001). Higher levels of SMV were reported in mottled than in nonmottled seedcoats (Koning, 1999). As the virus is seedborne, seeds provide an important inoculum source between production seasons (Gardner and Kendrick, 1921; Ross, 1969; Hill et al., 1980). The rate of seed transmission varies from 0 to 68%, but is most often close to 10%, depending on the host genotype, virus strain and time of infection (Shepherd, 1972). Although seedcoat infection in the absence of embryonic infection does not lead to seed transmission, SMV antigen can be detected in ⬎99% of seeds collected from SMV infected plants (Bossenec and Maury, 1978; S.A. Ghabrial, unpublished data, 1997). Therefore, detection of seedcoat infection, which is independent of host genotype and virus strain, serves as a good indicator of SMV infection of the individual plants and provides reliable estimates of SMV incidence in the field (S.A. Ghabrial, unpublished data, 1997). Additionally, aphids are a vector for infection of healthy plants (Abney et al., 1976; Halbert et al., 1981), thus plant infection percentages can be much higher than seed levels. Aphid transmission depends on aphid activity, including the timing, numbers and species composition of transient alate aphids (Halbert and Irwin, 1981; Irwin and Goodman, 1981; Schultz et al., 1985). In North Carolina, the incidence of Phomopsis sojae seed infection in SMV susceptible plants was higher than in SMV resistant plants (Ross, 1977). Infection of susceptible plants with SMV in these studies was solely dependent upon the transmission of SMV from source inocula provided by spreader rows. In Illinois, SMVinoculated susceptible plants had a higher incidence of Phomopsis spp. seed infection than noninoculated plants (Hepperly et al., 1979). In Kentucky, Stuckey et al. (1982), using a mild isolate of SMV, did not consistently identify significant increases in Phomopsis spp. seed infection as a result of SMV infection. In contrast, Koning et al. (2001) reported that SMV susceptible plants, mechani-

Infection of soybean [Glycine max (L.) Merr.] plants with Soybean mosaic virus (SMV) has been reported to enhance Phomopsis spp. infection, which reduces seed quality. The timing and incidence of SMV infection depends largely upon the level of primary inoculum and aphid-activity. Two field experiments were conducted in aphidfree environments, to examine the influence of (i) SMV-infection, and (ii) SMV-resistance alleles of the Rsv1 gene, on the incidence of Phomopsis spp. seed infection. In the first experiment, mock inoculated (potassium phosphate buffer) SMV-susceptible cultivars (Clark and Williams), and their SMV-resistant isolines (L78-434 and L78379, with dominant Rsv1 allele conferring resistance to SMV strains G1-G6), showed low levels (⬍10%) of Phomopsis spp. seed infection. In contrast, susceptible cultivars mechanically inoculated with SMV (G2 strain, V8 stage) exhibited a 3- to 8-fold increase in the incidence Phomopsis spp. seed infection. In the second experiment, mock inoculation of the susceptible cultivar, Clark, and two SMV-resistant lines (10-rsv1y and 18-rsv1y, with recessive rsv1y allele conferring resistance to SMV strains G1-G3), resulted in ⬍20% Phomopsis spp. seed infection. In contrast, those plants mechanically inoculated with SMV (strain G6, V8 stage) had significantly higher levels of Phomopsis spp. seed infection (52 to 78%). It is concluded that the lower incidence of Phomopsis spp. seed infection in SMV-resistant plants was not due to the SMV resistance alleles of the Rsv1 gene per se, but rather due to the absence of SMV infection. Thus, the use of SMV-resistant varieties prevented/reduced SMV and Phomopsis spp. seed infection.

P

homopsis spp. infection is a major cause of poor seed quality (Wallen and Cuddy, 1960; Kmetz et al., 1978; Koning et al., 2001). Phomopsis spp. seed infections occur primarily during or after physiological maturity under conditions of high temperature and relative humidity (Miller and Roy, 1982; Balducchi and McGee, 1987; Thomison et al., 1988). Previous workers have reported that infection of soybean [Glycine max (L.) Merr.] with SMV enhanced Phomopsis spp. seed infection (Ross, 1977; Hepperly et al., 1979; Koning et al., 2001) and led to reduced seed quality and vigor (Koning et al., 2001). Soybean mosaic, caused by SMV, is a common soybean viral disease. Infections by SMV at or before soybean floral development have resulted in up to 40% yield loss (Ross, 1969; Quiniones et al., 1971; Irwin and Goodman, 1981; Ren et al., 1997a). Such infections also

Abbreviations: SMV, Soybean mosaic virus; BS, beginning seedfill; FS, full seed; YP, yellow pod; HM, harvest maturity; Rsv1-MI, SMVresistant isolines (L78-434, L78-379, with dominant Rsv1 allele conferring resistance to SMV strains G1-G6) mechanically-inoculated with 0.05 M potassium phosphate buffer (mock inoculation); S-MI, SMVsusceptible cultivars (Clark, Williams) mechanically inoculated with 0.05 M potassium phosphate buffer (mock inoculation); S-SMV, SMVsusceptible cultivars mechanically inoculated with the G2 strain of SMV; 10-rsv1y and 18-rsv1y, SMV-resistant lines (with recessive rsv1y allele conferring resistance to SMV strains G1-G3); MI, mock inoculation; SMV-G6, mechanically inoculated with G6 strain of SMV; aPDA, potato dextrose agar.

G. Koning, D.M. TeKrony, T.W. Pfeiffer, Dep. of Agronomy, and S.A. Ghabrial, Dep. of Plant Pathology, Univ. of Kentucky, Lexington, KY 40546. Part of a dissertation submitted by the senior author in partial fulfillment of the Ph.D. degree. Received 5 Sept. 2000. *Corresponding author ([email protected]). Published in Crop Sci. 42:178–185 (2002).

178

KONING ET AL.: INFLUENCE OF SMV RESISTANCE GENE ON PHOMOPSIS SPP. SEED INFECTION

cally inoculated with a more severe Kentucky isolate of SMV than used by Stuckey et al. (1982), had four to seven times more Phomopsis spp. seed infection than noninoculated susceptible or resistant plants. Although these previous workers supplied SMV inocula to soybean plants, they paid little attention to the significant role of aphids in the transmission of SMV. They did not include any apparent form of vector control, or monitor aphid activity, which could influence the timing and incidence of SMV infection significantly. Furthermore, not all of these studies included SMV resistant isolines, which, in the absence of vector control, could provide a valuable experimental control. In our previous study (Koning et al., 2001), noninoculated SMV susceptible cultivars became infected by SMV via aphid transmission, and had a higher level of Phomopsis spp. seed infection than their SMV resistant isolines. Though unlikely, the question remained whether the Rsv1 resistance allele was providing some direct resistance to Phomopsis spp. seed infection, as opposed to eliminating predisposition to Phomopsis spp. by preventing SMV infection. In this study, we established aphid free field environments and included both SMV susceptible and resistant genotypes, to investigate the influence of (i) SMV infection, and (ii) SMV resistance alleles of the Rsv1 gene, on the incidence of Phomopsis spp. seed infection. MATERIALS AND METHODS Two field trials were conducted in 1996 and 1997 at the Spindletop Experimental farm at Lexington, KY (38⬚N lat). All field grown soybeans were enclosed in cages (6 ⫻ 3 ⫻ 1.5 m) made of nylon netting (12.8 mesh cm⫺1 ), to prevent the transmission of SMV by aphid vectors.

Experiment I, 1996 and 1997 Two SMV susceptible cultivars, Clark (Maturity Group (MG) early IV) and Williams (MG III), and their respective SMV resistant isolines (L78-434 and L78-379) were used to investigate the effect of SMV infection on the incidence of Phomopsis spp. seed infection. The susceptible cultivars have the recessive rsv1 allele at the Rsv1 locus and are susceptible to all strains of SMV. Isolines resistant to SMV strains G1 to G6 were produced by backcrossing the most dominant Rsv1 resistance allele from PI 96983 into either Clark or Williams (Chen et al., 1991). Two hill plots from each susceptible cultivar, and one hill plot of seeds from each resistant isoline, were planted (24 May 1996, 5 June 1997) in a randomized complete block design with three replications. Hill plots were arranged 0.76 m apart in an equidistant pattern and consisted of 15 seeds, planted at a depth of 2.5 cm within a linear distance of 36 cm. As the seedlings reached the V5 growth stage (Fehr and Caviness, 1977), plots were thinned to ten seedlings per hill and the plants were enclosed in a preconstructed cage. The cage interior and plants within were sprayed with malathion (3 ml L⫺1, [(Dimethoxyphosphinothioyl)thio]butanedioic acid diethyl ester) to ensure an insect free environment, and monitored biweekly for aphids for the remainder of the study. Plants were irrigated with an overhead sprinkler system, as needed to minimize moisture stress and maintain a favorable environment for disease infection. The daily rainfall, relative humidity and air temperatures during the growing seasons were obtained from the Agricultural Weather Station, located

179

苲0.5 km from the experiment. Hourly air temperatures (above and below plant canopy) were acquired via thermocouples, shielded from direct radiation, at randomly selected periods during seed development and maturation, inside and outside (in an adjacent soybean row plot) the cages. The percentage of photosynthetically active radiation intercepted by the nylon material was estimated with a Licor line quantum sensor (LICOR, Lincoln, NE). At the V8 growth stage, plants in one plot of each SMV susceptible cultivar were either mechanically inoculated with a moderately severe Kentucky isolate of SMV (belonging to G2 strain) (S-SMV) or mock inoculated with 0.05 M potassium phosphate buffer (S-MI), as described by Koning et al. (2001). The control consisted of SMV resistant isolines mock inoculated with 0.05 M potassium phosphate buffer at the V8 growth stage (Rsv1-MI). In order to quantify the incidence level of premature SMV infection, experimental plants in each plot were visually examined and no symptoms of SMV were found before inoculations were made. To provide a source of Phomopsis spp. inocula, Phomopsis infested stems and residue collected from the previous seasons’ soybean fields were scattered between plants at the R2 growth stage, and a spore suspension, prepared from cultures grown on acidified (pH 4.5) Potato Dextrose Agar (aPDA) for 14 to 21 d, was atomized onto plants at beginning seedfill (growth stage R5). To provide a pool of pods at the same stage of development, a minimum of 20 3-seeded pods per hill plot (at two pods per plant) were marked with acrylic paint (Egli, 1999), at the beginning seedfill stage (BS, pods were dark green, seeds were green and just starting to swell with a 3–4 mm diam.). Marked pods were hand harvested at (i) yellow pod (YP, pods were 苲90% yellow, and the seeds were yellow and at a moisture content of 苲550 g kg⫺1 ) and (ii) harvest maturity (HM, pods were brown, and the seeds were yellow and at a moisture content of ⱕ140 g kg⫺1 ). After sampling, pods were stored at 10⬚C until use. Apical and basal seeds were removed from the pod, and bisected (each half consisting of a cotyledon and its seedcoat half). One seed half was evaluated for Phomopsis spp. infection, after 14 d incubation on aPDA at 25⬚C under continuous light (TeKrony et al., 1984), and the other seed half was evaluated for the accumulation of SMV antigen in the seedcoat, with the direct form of ELISA (Ghabrial and Schultz, 1983). A composite hand harvest was made at HM, and 50 seeds from each treatment were evaluated for Phomopsis spp. and the accumulation of SMV as described above. In 1997, seeds from mock inoculated SMV resistant and susceptible plants were evaluated for the effect of the cage environment on seed quality. Two replicates of 100 seeds from each treatment were tested for germination and vigor by the standard germination (Association of Official Seed Analysts, 1998), accelerated aging (International Seed Testing Association, 1995), and bulk conductivity (Loeffler et al., 1988) tests.

Experiment II, 1997 The effect of the SMV resistance alleles of the Rsv1 gene on Phomopsis spp. infection was investigated in 1997 by using two SMV resistant Kentucky breeding lines (10-rsv1y and 18rsv1y) and the susceptible ‘Clark’ as the control. The two resistant lines have a recessive rsv1y allele which confers resistance to SMV strains G1 to G3 (Cho and Goodman, 1979; Buss et al., 1989). Resistance in these lines was derived from ‘Hutcheson’, whose resistance originated from ‘York’ (Smith, 1968). While 10-rsv1y and 18-rsv1y are phenotypically similar, they have different background genotypes originating from ‘A4393’ ⫻ ‘Hutcheson’ and ‘A3935’ ⫻ ‘Hutcheson’, respectively. The three genotypes have similar maturity dates. Experiment II was conducted as Experiment I, except that

180

CROP SCIENCE, VOL. 42, JANUARY–FEBRUARY 2002

different genotypes were used and inoculum was the G6 strain of SMV (provided by Dr. S.A. Ghabrial, Department of Plant Pathology, University of Kentucky, Lexington, KY). Two hill plots of seeds from each genotype were planted (5 June 1997). At the V8 growth stage, plants in one plot of each genotype were either mechanically inoculated with the G6 strain of SMV (SMV-G6) or mock inoculated with 0.05 M potassium phosphate buffer (MI). At YP and HM, ten marked pods were hand harvested. Seeds were stored at 10⬚C until laboratory evaluations were made for the accumulation of SMV in seedcoats (by ELISA) and Phomopsis spp. seed infection (on aPDA).

Statistical Analyses For Experiments I and II, data collected at different growth stages were analyzed as a factorial treatment structure (genotype ⫻ treatment ⫻ seed growth stage) in a randomized complete block design with repeated measures with PROC MIXED of SAS (SAS Institute, 1997). Variance components were used as the covariance structure for all response variables. Differences were determined by the Least Significant Difference (LSD) procedure, and PROC CORR and PROC REG of SAS were used for correlation and simple linear regression analysis, respectively. Data collected from the composite harvests at harvest maturity (Experiment I) were analyzed as a factorial treatment structure (genotype ⫻ treatment) in a randomized complete block design with PROC GLM of SAS. Differences were determined by the LSD procedure, and PROC CORR of SAS姞 was used for correlation analysis.

RESULTS Experiment I, 1996 and 1997 There were no significant differences in the responses between the two SMV susceptible cultivars (Clark, Williams) or the two SMV resistant isolines (L78-434, L78379), for any of the variables evaluated. The data are therefore presented as averages across the susceptible cultivars or the resistant isolines. Mock inoculated resistant and susceptible (Rsv1-MI and S-MI, respectively) plants reached the YP and HM stages at approximately the same time (Table 1). On the other hand, SMV susceptible plants inoculated with the G2 strain of SMV (S-SMV) extended the interval between full seed (FS, pods were dark green, seeds were green and immature but completely filled the locular cavity) and YP by two to four days, compared with the length of the same period in mock inoculated plants. Also, in 1997, SMV-infection extended the time from YP to HM by 12 d, compared with S-MI plants.

Fig. 1. Effects of SMV (strain G2) infection at the V8 growth stage in caged hill plots (Expt. I), on the accumulation of SMV in seedcoats in (a) 1996 and (b) 1997, and the incidence of Phomopsis spp. seedcoat infection in (c) 1996 and (d) 1997, at yellow pod and harvest maturity. Data averaged across apical and basal seeds of two genotypes. Rsv1-MI, resistant isolines (L78-434, L78-379) mechanically inoculated with 0.05 M potassium phosphate buffer (mock inoculation); S-MI, SMV susceptible cultivars (Clark, Williams) mechanically inoculated with 0.05 M potassium phosphate buffer (mock inoculation); S-SMV, SMV susceptible cultivars mechanically inoculated with the G2; ND, not detected by ELISA. Bars with different letters, within each year and response variable, are significantly different at the 0.05 probability level.

Accumulation of SMV Mock inoculated resistant and susceptible plants did not show symptoms of SMV and their seedcoats at YP, HM (Fig. 1a, b) and in the composite sample at HM (Table 2) were consistently SMV free. In contrast, SMV infected susceptible plants showed symptoms typical of SMV infection (McGee, 1992; Sinclair, 1992), and SMV antigen was detected in seedcoats of their seeds at a concentration ranging from 7 to 13 ␮g g⫺1 (fresh weight basis) in 1996 (Fig. 1a; Table 2), and 27 to 48 ␮g g⫺1 in 1997 (Fig. 1b; Table 2). The concentration of SMV antigen in seedcoats at HM was 31 and 44% lower than at YP in 1996 and 1997, respectively. Phomopsis spp. Infection All pods harvested from SMV infected plants exhibited visual signs (McGee, 1992) of Phomopsis spp. infection (Koning, 1999). The incidence of Phomopsis spp.

Table 1. Effect of inoculation with Soybean mosaic virus (SMV) on the duration of plant reproductive development in Experiment I. Genotype-treatment, 1996 Growth period

Genotype-treatment, 1997

Rsv1-MI†

S-MI

S-SMV

19a§ 24b 11a

19a 24b 11a

19a 28a 11a

Rsv1-MI

S-MI

S-SMV

15a 22b 10b

15a 24ab 8b

15a 26a 20a

Duration (days) BS–FS‡ FS–YP YP–HM

† Rsv1-MI, resistant isolines (L78-434, L78-379) mechanically inoculated with 0.05 M potassium phosphate buffer (mock inoculation); S-MI, SMV susceptible cultivars (Clark, Williams) mechanically inoculated with 0.05 M potassium phosphate buffer (mock inoculation); S-SMV, SMV-susceptible cultivars mechanically inoculated with the G2, at V8 growth stage. ‡ BS, beginning seedfill; FS, full seed; YP, yellow pod; HM, harvest maturity. § Means followed by different letters in each row within years are significantly different at the 0.05 probability level.


181

Table 2. Soybean mosaic virus (SMV) and Phomopsis spp. seed infection, and soybean seed quality, of seeds from the composite harvest at harvest maturity in 1996 and 1997, from plants grown in caged hill plots in Experiment I. Genotype-treatment Rsv1-MI† SMV, ␮g g⫺1 fw seedcoat Phomopsis spp., % Seedcoat Cotyledon SMV, ␮g g⫺1 fw seedcoat Phomopsis spp., % Seedcoat Cotyledon Standard germination, % Accelerated aging germination, % Bulk conductivity, mS m⫺1 g⫺1

0b‡ 5b 1b 0b 2b 1b 98a 98a 5.7a

S-MI 1996 0b 6b 2b 1997 0b 2b 1b 98a 98a 5.7a

S-SMV 13a 79a 59a 30a 54a 34a nd§ nd nd

† Rsv1-MI, resistant isolines (L78-434, L78-379) mechanically inoculated with 0.05 M potassium phosphate buffer (mock inoculation); S-MI, SMV susceptible cultivars (Clark, Williams) mechanically inoculated with 0.05 M potassium phosphate buffer (mock inoculation); S-SMV, SMV susceptible cultivars mechanically inoculated with the G2, at V8 growth stage. ‡ Means followed by different letters in each row are significantly different at the 0.05 probability level. § nd, not determined due to seed number limitations.

infection in seeds from mock inoculated plants, at YP, HM (Fig. 1c, d), and the composite sample at HM (Table 2) was consistently low (⬍10%). Infection of susceptible plants with SMV resulted in a significant three- to eightfold increase in the incidence of Phomopsis spp. seedcoat infection compared with mock inoculated plants (Fig. 1c, d), with levels ranging from 43 to 64% in 1996, and 34 to 83% in 1997. In 1996, the incidence of Phomopsis spp. seed infection was 33% higher at HM than at YP, whereas in 1997, infection was 59% higher at YP than at HM. At the composite harvest, levels of cotyledonary and seedcoat Phomopsis spp. infection of seeds from SMV infected plants were, respectively, 33 to 57% and 52 to 73% higher than of seeds from mock inoculated plants (Table 2).

Fig. 2. Relationship between the incidence of Phomopsis spp. seed infection and the concentration of SMV (G2 strain) antigen (Expt. I) in seedcoats, at yellow pod and harvest maturity in (a) 1996 and (b) 1997. Averaged across apical and basal seeds (n ⫽ 20), and regressed across individual plots of all SMV inoculation treatments of two SMV susceptible cultivars (Clark, Williams) and their SMV resistant isolines (L78-434, L78-379). **, *** Significant at the 0.01 and 0.001 probability levels, respectively.

The incidence of Phomopsis spp. seedcoat infection was consistently low (⬍20%) in seeds from mock inoculated plants (Fig. 3c, d). Plants infected with the G6 strain of SMV, however, showed a significantly higher incidence of Phomopsis spp. seed infection, ranging from 65 to 73% at YP and 52 to 78% at HM (Fig. 3c, d).

Relationship between SMV and Phomopsis spp. Infection There was a positive and highly significant linear relationship between the incidence of Phomopsis spp. seed infection and the concentration of SMV antigen in seedcoats at YP and HM (Fig. 2). Thus, as the accumulation of SMV in seedcoats increased, so did the incidence of Phomopsis spp. infection in maturing seeds.

Experiment II, 1997 There were no significant differences in the responses between the three genotypes. No SMV symptoms were observed in mock inoculated plants, and SMV antigen was not detected in their seedcoats at either YP or HM (Fig. 3a, b). In contrast, plants infected with the G6 strain of SMV showed symptoms of SMV infection, and SMV antigen was detected in their seedcoats. The concentration of SMV detected in seedcoats of seeds harvested at YP was between 41 and 46 ␮g g⫺1, and between 27 and 30 ␮g g⫺1 at HM (Fig. 3a, b).

Fig. 3. Effects of SMV (strain G6) infection at the V8 growth stage, in caged-hill plots (Expt. II) in 1997, on the amount of SMV in seedcoats at (a) yellow pod and (b) harvest maturity; and the incidence of Phomopsis spp. seed infection at (c) yellow pod and (d) harvest maturity. Data averaged across apical and basal seeds. MI, mock inoculation (0.05 M potassium phosphate buffer); SMVG6, mechanically inoculated with G6 strain of SMV, at the V8 growth stage; Clark, SMV susceptible cultivar; 10-rsv1y and 18-rsv1y, SMV resistant lines (with recessive rsv1y allele conferring resistance to SMV strains G1-G3); ND, not detected by ELISA. Bars with different letters are significantly different at the 0.05 probability level.

182


Table 3. Weather conditions during seed development and maturation in Experiment I, 1996 and 1997.† Air temperature Seed maturation period

Min.

Max.

Relative humidity Mean

ⴗC BS–FS§ FS–YP YP–HM

18 14 11

28 25 22

23 20 17

BS–FS FS–YP YP–HM

14 13 11

27 25 24

21 19 18

Precipitation‡ mm 1996 30 75 46 1997 42 30 4

Min.

Max.

Mean

% 60 62 60

100 100 100

80 81 80

58 59 49

100 100 98

79 80 74

† Recorded at the Agricultural Weather Station, University of Kentucky. Averaged across all inoculation-treatments and genotypes. ‡ Includes irrigation. § BS, beginning seedfill; FS, full seed; YP, yellow pod; HM, harvest maturity.

Environmental Conditions The air temperature and relative humidity during seed development and maturation in 1996 and 1997 were similar (Table 3). Precipitation during early seed development from BS to FS, which included the time at which the Phomopsis spp. spore suspension was atomized onto the plants, was 12 mm higher in 1997 than in 1996. Precipitation during the FS-YP and YP-HM seed maturation phases was, however, between 42 and 45 mm greater in 1996 than in 1997 (Table 3). The nylon cage material reduced photosynthetically active radiation by approximately 50% (data not shown). As a consequence, plants became elongated and required staking to provide vertical support. The nylon material did not influence the passage of water, and the total precipitation and relative humidity above the plant canopy, both in and outside the cage (in an adjacent soybean row plot of another experiment; Koning, 1999), were similar. The material did, however, appear to restrict air movement and trap heat inside the cage. While the air temperatures above the plant canopy, in and outside the cage, did not differ significantly, maximum temperatures below the plant canopy were 4 to 16⬚C higher inside the cage than outside the cage (Koning, 1999). The higher temperatures were maintained for approximately one hour, after which they gradually declined to those temperatures outside the cage. Regardless of the insolation levels (ranging between 7 and 24 MJ m⫺2 d⫺1 ), this phenomenon was apparent both before and after leaf abscission (i.e., at R5 and R8, respectively). The higher temperatures inside the cage (experienced by all plants in this experiment) did not, however, appear to directly influence seed quality, as reflected by the high standard germination and accelerated aging germination, and low bulk conductivity, of seeds harvested from mock inoculated plants in 1997 (Table 2).

DISCUSSION During two consecutive years, in environments which were considered to be aphid free and favorable for Phomopsis spp. infection, the SMV Kentucky isolate (G2 strain) predisposed soybean plants to Phomopsis spp. seed infection. In the absence of aphids, mock inoculated SMV susceptible and resistant plants remained

SMV free, and yielded low levels of Phomopsis spp. seed infection. In contrast, those plants infected with SMV showed a three to eight-fold increase in Phomopsis spp. seed infection. In agreement with earlier field studies (Koning et al., 2001), a positive and highly significant linear relationship was observed between the accumulation of SMV antigen in seedcoat tissues and the incidence of Phomopsis spp. seed infection. Clearly, SMV infection increased the susceptibility of plants to Phomopsis spp. seed infection. The SMV susceptible and resistant plants responded to Phomopsis spp. infection in a similar manner since, in the absence of SMV infection (mock inoculation), the incidence of Phomopsis spp. seed infection was negligible in both types. This suggested that additional Phomopsis susceptibility/resistance characteristics were not incorporated into the SMV resistant isolines during backcrossing, and that the Rsv1 gene per se did not appear to directly influence Phomopsis spp. seed infection. The question of whether SMV resistance alleles of the Rsv1 gene influenced Phomopsis spp. infection, however, still remained and was addressed in Experiment II. Resistance to SMV is provided by a single resistance gene (Chen et al., 1991), which has multiple alleles at the common Rsv1 locus, whereby resistance is conferred to different strains of SMV on the basis of the allele. Allele Rsv1 provides resistance to SMV strains G1-G6, while allele rsv1y provides resistance to strains G1-G3 but not to strains G4-G7. Inoculation of soybean containing the rsv1y resistance allele with an SMV strain that overcomes the resistance (e.g., G6) will show whether an SMV resistance allele at the Rsv1 gene provides a direct reduction of Phomopsis spp. seed infection in the presence of soybean mosaic, or only provides an indirect reduction of Phomopsis spp. seed infection through the prevention of soybean mosaic. Experiment II clearly demonstrated that genotypes susceptible to the G6 strain of SMV (Clark, 10-rsv1y and 18-rsv1y) responded to Phomopsis spp. infection in a similar manner, regardless of the specific alleles (i.e., rsv1y or rsv1) present at the Rsv1 locus. In the absence of SMV infection, all genotypes had ⬍20% Phomopsis spp. seed infection. In the presence of SMV infection (G6 strain) however, the levels of Phomopsis spp. seed infection in


all genotypes increased up to 78%, despite the presence of different alleles at the Rsv1 locus. It was therefore concluded that the lower incidence of Phomopsis spp. seed infection in SMV resistant plants was not due to SMV resistance alleles of the Rsv1 gene per se, but rather due to the absence of SMV infection. Soybean plants were successfully field grown in caged hill plots, with both the timing and incidence of SMV infection under control. The SMV infection of planting seeds used in this study and the aphid populations prior to caging were not determined. However, the rate of seed transmission is most often low (10%) (Shepherd, 1972), and in Kentucky few plants are infected by SMV via natural aphid infection before growth stage R3 in early planting or growth stage R1 in late planting (Ren et al., 1997a). Infection of experimental plants by SMV via seed or aphids was therefore assumed to be negligible. The inability to detect SMV infection in the susceptible and resistant plants before they were enclosed in cages, and the absence of any visible aphid-vectors in the caged experiments, allowed mock inoculated SMV susceptible plants to be regarded as an experimental control, in addition to the SMV resistant isolines. Thus, the singular difference between mock inoculated and SMV inoculated SMV susceptible plants was the presence or absence of SMV infection. These data confirmed the reports by Ross (1977), Hepperly et al. (1979), and Koning et al. (2001) that SMV infection increases Phomopsis spp. seed infection. They were not in complete agreement with Stuckey et al. (1982) however, who, using a mild strain of SMV, found a consistent significant increase in Phomopsis spp. seed infection only in plants that were doubly infected with SMV and bean pod mottle virus (BPMV). These authors explained the differences between their results and those of Hepperly et al. (1979) who used a severe strain of SMV, on the basis of the virus strain used. The SMV isolate used in both field (Koning et al., 2001) and these cage studies was a more severe isolate than the mild isolate used in earlier experiments by Stuckey et al. (1982) in Kentucky. The levels of BPMV were not measured in our field studies (Koning et al., 2001). A significant increase in Phomopsis spp. seed infection occurred following SMV-inoculation (G2 strain, V8 stage) in this caged experiment however, where natural infection by either SMV and BPMV was prevented by controlling insect levels. Thus, there was little doubt that an increase in the accumulation of SMV in seedcoats of seeds harvested from SMV susceptible plants resulted in an increase in the incidence of Phomopsis spp. seed infection. An important aspect relating to the establishment of a disease, is the prevailing environmental conditions. The optimal temperature for virus multiplication and accumulation is 25⬚C (Tu, 1992), and washing or spraying leaves with water after inoculation is a fairly widespread practice to increase the number of local lesions and susceptibility to viral infection (Matthews, 1991). Favorable warm and wet conditions at and/or following SMV inoculations in 1997, may have created an environment more conducive for SMV infection, contributing

183

to the greater accumulation of SMV in 1997 compared to 1996. Similarly, warm, wet conditions occurring before the yellow pod stage play an important role in the dissemination of Phomopsis spp. inoculum to pods, and favor the movement of Phomopsis spp. from pods to seeds after the yellow pod stage (Spilker et al., 1981; McGee, 1983; Wilcox et al., 1985; TeKrony et al., 1987). The prevailing environmental conditions in our study failed to completely explain the variation in levels of Phomopsis spp. seed infection. We atomized plants with a Phomopsis spp. spore suspension, and were not dependent upon the environment for the dissemination of inocula. High precipitation after spore atomization and during seed filling in 1997 may, however, have favored the establishment of Phomopsis spp. infection in pods, and led to higher levels of seed infection at yellow pod compared to 1996. Drier conditions after yellow pod in 1997 may explain the loss of Phomopsis spp. infection as the seeds reached harvest maturity. Infection may not have been well established at yellow pod, and under less favorable conditions, infection may have been reduced. As previously reported (Tu, 1989; Koning et al., 2001), SMV infected plants matured later than noninfected plants, with the delay in maturation being primarily due to an extension of seed development (FS-YP). In 1997, however, there was also a longer interval between the yellow pod stage and harvest maturity. As several workers have previously reported, the length of the FS-YP seed development period (Tu, 1989; Vaughan et al., 1989; Koning et al., 2001), and the YP-HM seed maturation period (Abney and Ploper, 1991; Ploper et al., 1992; Abney and Ploper, 1994) were positively and significantly correlated to Phomopsis spp. seed infection at harvest maturity. Due to the critical influence of the environment on Phomopsis spp. seed infection, extended exposure of pods and seeds to warm, wet conditions may provide a greater opportunity for Phomopsis spp. infection. Our study therefore supports the hypothesis that infection by SMV extended seed maturation, and in part, prolonged the exposure of pods and seeds to infection by Phomopsis spp. The higher levels of Phomopsis spp. seed infection of SMV infected plants, however, was not directly related to more rainfall events during the extended period of seed maturation (Koning, 1999). In conclusion, these caged experiments (absence of aphids) clearly illustrated that two soybean cultivars had significantly different levels of Phomopsis spp. seed infection depending upon the presence/absence of SMV infection. When SMV infection was not detected, the incidence of Phomopsis spp. seed infection was negligible, even under conditions conducive to Phomopsis spp. infection. In the presence of SMV infection, however, the incidence of Phomopsis spp. seed infection increased significantly. The incidence of Phomopsis spp. seed infection was undoubtedly associated with the accumulation of SMV in seedcoats. The mechanism whereby soybean plants are predisposed to Phomopsis spp. infection is as yet unknown. Furthermore, it may be concluded that the SMV-resistance alleles of the Rsv1 gene do not

184


prevent Phomopsis spp. seed infection per se, but rather prevent SMV infection. Soybean genotypes resistant to SMV, thereby provide a practical means of managing two soybean pathogens. Although planting late maturing cultivars or delaying planting of early and mid season cultivars may reduce Phomopsis spp. seed infection (TeKrony et al., 1984; Thomison et al., 1990), late planted soybean are more susceptible to yield reductions due to SMV, than are early planted soybean (Ren et al., 1997b). Thus, the use of late-planted SMV resistant genotypes where there is predisposition to infection by Phomopsis spp. by SMV infection, would prevent SMV and reduce Phomopsis spp. seed infection. ACKNOWLEDGMENTS We acknowledge the South African Foundation of Research and Development, and the Research Challenge Trust of the University of Kentucky, for financial support, and thank Dr. R.L. Bernard, USDA-ARS and Department of Agronomy, University of Illinois, Urbana, IL, the originator of the resistant isolines, for providing the seeds used in these experiments.

REFERENCES Abney, T.S., J.O. Silling, T.L. Richards, and D.B. Boersma. 1976. Aphids and other insects as vectors of soybean mosaic virus. J. Econ. Entomol. 69:254–256. Abney, T.S., and L.D. Ploper. 1991. Growth regulator effects on soybean seed maturation and seedborne fungi. Plant Dis. 75:585– 589. Abney, T.S., and L.D. Ploper. 1994. Effects of bean pod mottle virus on soybean seed maturation and seedborne Phomopsis spp. Plant Dis. 78:33–37. Association of Official Seed Analysts. 1998. Rules for testing seeds. Assoc. Off. Seed Anal., Lincoln, NE. Balducchi, A.J., and D.C. McGee. 1987. Environmental factors influencing infection of soybean seeds by Phomopsis and Diaporthe species during seed maturation. Plant Dis. 71:209–212. Bossenec, J.M., and Y. Maury. 1978. Utilization of an ELISA technique for the detection of soybean mosaic virus in the grain of soybean. Ann. Phytopathol. 10:263–268. Buss, G.R., P. Chen, S.A. Tolin, and C.W. Roane. 1989. Breeding soybeans for resistance to soybean mosaic virus. p. 1144–1154. In A.J. Pascale (ed.) World Soybean Research Conference IV Proceedings. Orientacion Grafica Editora, Buenos Aires, Argentina. Chen, P., G.R. Buss, C.W. Roane, and S.A Tolin. 1991. Allelism among genes for resistance to soybean mosaic virus in strain-differential soybean cultivars. Crop Sci. 31:305–309. Cho, E.K., and R.M. Goodman. 1979. Strains of soybean mosaic virus: classification based on virulence in resistant soybean cultivars. Phytopathology 69:467–470. Egli, D.B. 1999. Variation in leaf starch and sink limitations during seed filling in soybean. Crop Sci. 39:1361–1368. Fehr, W.R., and C.E. Caviness. 1977. Stages of soybean development. Spec. Rep. 80. Iowa Agric. Home Econ. Exp. Stn., Iowa State Univ., Ames, IA. Gardner, M.W., and J.B. Kendrick. 1921. Soybean mosaic. J. Agric. Res. 22:111–114. Ghabrial, S.A., and F.J. Schultz. 1983. Serological detection of bean pod mottle virus in bean leaf beetles. Phytopathology 73:480–483. Halbert, S.E., and M.E. Irwin. 1981. Effect of soybean canopy closure on landing rates of aphids with implications for restricting spread of soybean mosaic virus. Ann. Appl. Biol. 98:15–19. Halbert, S.E., M.E. Irwin, and R.M. Goodman. 1981. Alate aphid (Homoptera: Aphididae) species and their relative importance as field vectors of soybean mosaic virus. Ann. Appl. Biol. 97:1–9. Hepperly, P.R., G.R. Bowers, J.B. Sinclair, and R.M. Goodman. 1979.

Predisposition to seed infection by Phomopsis sojae in soybean plants infected by soybean mosaic virus. Phytopathology 69:846– 848. Hill, H.J., B.S. Lucas, H.I. Benner, H. Tachibana, R.B. Hammond, and L.P. Pedigo. 1980. Factors associated with the epidemiology of soybean mosaic virus in Iowa. Phytopathology 70:536–540. International Seed Testing Association. 1995. Handbook of vigour test methods. 3rd ed., Int. Seed Test. Assoc., Zurich. Irwin, M.E., and R.M. Goodman. 1981. Ecology and control of soybean mosaic virus. In K. Maramorosch and K.F. Harris (ed.) Plant diseases and vectors: Ecology and epidemiology. Academic Press, New York. Kmetz, K.T., A.F. Schmitthenner, and C.W. Ellett. 1978. Soybean seed decay: Prevalence of infection and symptom expression caused by Phomopsis sp., Diaporthe phaseolorum var. sojae and Diaporthe phaseolorum var. caulivora. Phytopathology 68:836–840. Koning, G. 1999. Relationship between soybean mosaic virus and Phomopsis spp. infection of soybean seeds. Ph.D. diss. Univ. of Kentucky, Lexington, KY (Diss. Abstr. 99-48882). Koning, G., D.M. TeKrony, T.W. Pfeiffer, and S.A. Ghabrial. 2001. Infection of soybean with soybean mosaic virus increases susceptibility to Phomopsis spp. seed infection. Crop Sci. 41:1850–1856. Loeffler, T.M., D.M. TeKrony, and D.B. Egli. 1988. The bulk conductivity test as an indicator of soybean seed quality. J. Seed Technol. 12:37–53. Matthews, R.E.F. 1991. Plant virology. 3rd ed. Academic Press, New York. McGee, D.C. 1983. Epidemiology of soybean seed decay by Phomopsis and Diaporthe spp. Seed Sci. Technol. 11:719–729. McGee, D.C. 1992. Soybean diseases. American Phytopathological Society, St. Paul, MN. Miller, W.A., and K.W. Roy. 1982. Mycoflora of soybean leaves, pods, and seeds. Can. J. Bot. 60:2716–2723. Ploper, L.D., T.S. Abney, and K.W. Roy. 1992. Influence of soybean genotype on rate of seed maturation and its impact on seedborne fungi. Plant Dis. 76:287–292. Quiniones, S.S., J.M. Dunleavy, and J.W. Fischer. 1971. Performance of three soybean varieties inoculated with soybean mosaic virus and bean pod mottle virus. Crop Sci. 11:662–664. Ren, Q., T.W. Pfeiffer, and S.A. Ghabrial. 1997a. Soybean mosaic virus incidence level and infection time: interaction effects on soybean. Crop Sci. 37:1706–1711. Ren, Q., T.W. Pfeiffer, and S.A. Ghabrial. 1997b. Soybean mosaic resistance improves productivity of double-cropped soybeans. Crop Sci. 37:1712–1718. Ross, J.P. 1969. Effect of time and sequence of inoculation of soybean mosaic and bean pod mottle virus on yield and seed characters. Phytopathology 59:1404–1408. Ross, J.P. 1977. Effect of aphid-transmitted soybean mosaic virus on yields of closely related resistant and susceptible soybean lines. Crop Sci. 17:869–872. SAS Institute. 1997. SAS/STAT Software: Changes and Enhancements through Release 6.12. SAS Inst. Inc., Cary, NC. Schultz, G.A., M.E. Irwin, and R.M. Goodman. 1985. Relationship of aphid (Homoptera: Aphididae ) landing rates to the field spread of soybean mosaic virus. J. Econ. Entomol. 78:143–147. Shepherd, R.J. 1972. Transmission of viruses through seed and pollen. p. 267–294. In C.I. Kado and H.O. Agrawal (ed.) Principles and Techniques in Plant Virology. Von Nostrand Rienhold, New York. Sinclair, J.B. 1992. Discoloration of soybean seeds – an indicator of quality. Plant Dis. 76:1087–1091. Smith, T.J. 1968. Registration of York soybeans. Crop Sci. 8:776. Spilker, D.A., A.F. Schmitthenner, and C.W. Ellett. 1981. Effects of humidity, temperature, fertility, and cultivar on the reduction of soybean seed quality by Phomopsis sp. Phytopathology 71:1027–1029. Stuckey, R.E., S.A. Ghabrial, and D.A. Reicosky. 1982. Increased incidence of Phomopsis sp. in seeds from soybeans infected with bean pod mottle virus. Plant Dis. 66:826–829. TeKrony, D.M., D.B. Egli, J. Balles, L. Tomes, and R.E. Stuckey. 1984. Effect of date of harvest maturity on soybean quality and Phomopsis sp. seed infection. Crop Sci. 24:189–193. TeKrony, D.M., D.B. Egli, and G.M. White. 1987. Seed production

TIAN ET AL.: GAMAGRASS SEED DORMANCY—CUPULE AND SCARIFICATION EFFECTS

and technology. p. 295–353. In J.R. Wilcox (ed.) Soybeans: Improvement, production, and uses. 2nd ed Agron. Monogr. 16. ASA, CSSA, and SSSA, Madison, WI. Thomison, P.R., D.L. Jeffers, and A.F. Schmitthenner. 1988. Phomopsis seed infection and nutrient accumulation in pods of soybean with reduced fruit loads. Agron. J. 80:55–59. Thomison, P.R., W.J. Kenworthy, and M.S. McIntosh. 1990. Phomopsis seed decay in soybean isolines differing in stem termination, time of flowering, and maturity. Crop Sci. 30:193–188. Tu, J.C. 1989. Effect of different strains of soybean mosaic virus on growth, maturity, yield, seed mottling and seed transmission in several soybean cultivars. J. Phytopathol. 126:231–236.

185

Tu, J.C. 1992. Symptom severity, yield, seed mottling and seed transmission in susceptible and resistant soybean: The influence of infection stage and growth temperature. J. Phytopathol. 135:28–36. Vaughan, D.A., R.L. Bernard, and J.B. Sinclair. 1989. Soybean seed quality in relation to days between development stages. Agron. J. 81:215–219. Wallen, V.R., and T.F. Cuddy. 1960. Relation of seed-borne Diaporthe phaseolorum to the germination of soybeans. Proc. Ass. Off. Seed Anal. 50:137–140. Wilcox, J.R., T.S. Abney, and E.M. Frankenberger. 1985. Relationships between seedborne soybean fungi and altered photoperiod. Phytopathology 75:797–800.

Cupule Removal and Caryopsis Scarification Improves Germination of Eastern Gamagrass Seed X. Tian, A. D. Knapp,* K. J. Moore, E. C. Brummer, and T. B. Bailey ABSTRACT

X. Tian, A.D. Knapp, K.J. Moore, and E.C. Brummer, Dep. of Agronomy, Iowa State Univ., Ames, IA 50011; T.B. Bailey, Dep. of Statistics, Iowa State Univ., Ames, IA 50011. Journal Paper No. J-18415 of the Iowa Agric. and Home Economics Exp. Stn., Ames, IA 50011. Project No. 3244, supported by Hatch Act and State of Iowa, and a grant from the Leopold Center for Sustainable Agriculture. Received 3 June 1999. *Corresponding author ([email protected]).

factors often cause seed dormancy. Toole et al. (1956) stated that seed dormancy in many species is caused by the inhibitory influence of structures covering the embryo rather than by factors within the embryo itself. In wild oat (Avena fatua L.), the lemma and palea contributed to reduced germination (Hay, 1960; Hsiao and Quick, 1985) and pricking the outer layers of the caryopses promoted germination of dormant seeds (Crocker, 1906). Woods and Gutek (1974) reported that freshly harvested wild rice (Zizania palustris L.) germinated following removal of the lemma and palea and scraping the pericarp covering the embryo. Simpson (1990) summarized the effect of removing the lemma and palea on germination in 25 species that normally exhibit a high degree of dormancy at the time of seed maturation. He observed that in one of the 54 comparisons, removing the lemma and palea gave 100% germination and in the others removal of hulls improved germination from 25 to 54%, on average. Simpson (1990) also summarized the influence of puncturing the pericarp and testa on overcoming dormancy in 31 grass species. He reported that scarifying the coat produced a significant increase in germination and 30% of the puncturing treatments achieved germination greater than 90%. The influence of encompassing structures on germination and dormancy in several warm-season grasses has also been studied. Coukos (1944) indicated that hammer mill processing tended to induce germination of dormant caryopses of little bluestem (Andropogon scoparius Michx.), big bluestem (Andropogon gerardii Vitman.), side-oats grama (Bouteloua curtipendula Michx.), indiangrass (Sorghastrum nutans L.), and smooth bromegrass (Bromus inermis Leyss.). Sautter (1962) scarified switchgrass (Panicum virgatum L.) with emery cloth to remove the lemma and palea and total germination was 84% with 74% of the seed germinating in 3 d. Ahring and Todd (1977) demonstrated that buffalograss [Buchle dactyloides (Nutt.) Engelm.] caryopses extracted from the seed-burs germinated readily. Seed dormancy in eastern gamagrass has also been studied. Ahring and Frank (1968) found that prechill

Published in Crop Sci. 42:185–189 (2002).

Abbreviations: 2, 3, 5-triphenyl tetrazolium chloride, TZ.

Eastern gamagrass (Tripsacum dactyloides L.) is a warm-season, perennial grass with high palatability and productivity. However, poor stand establishment, often due to seed dormancy, limits its widespread use. Seed dormancy is often caused by structures surrounding the embryo, the physiological state of the embryo itself, or a combination of these factors. The eastern gamagrass dispersal unit is a floret within a thick, hard cupule. The objective of this study was to evaluate effects of cupule (including lemma and palea) removal and caryopsis scarification on germination of eastern gamagrass by means of different commercial seed lots produced in different locations and years. Germination tests were conducted at 20/30ⴗC alternating temperature with light during 30ⴗC for 8 h daily. Germination counts were made every 7 d. After 28 d, the germination of decupulated caryopses from different seed lots germinated from 16 to 49% across seed lots, compared with 5 to 18% germination for caryopses with cupule intact. Scarifying the pericarp over the embryo, however, resulted in germination of all dormant seeds. We conclude that while the cupule (including the lemma and palea) contributes to the dormancy of eastern gamagrass, the pericarp and/or testa are the main factors restricting germination of this species. In addition, caryopsis scarification increased the germination rate and the germination test could be shortened to 21 or even 14 d depending on the seed lot.

E

astern gamagrass is a warm-season, perennial grass with high palatability and productivity (Polk and Adcock, 1964). Recently, interest in eastern gamagrass has increased because of its potential use for high quality forage, soil conservation, and wildlife habitat (Burns et al., 1992; Hardin, 1994). Poor stand establishment, often due to seed dormancy, restricts its widespread use, affects planting practices, and complicates seed inventory management. Structures encompassing the embryo, the physiological state of the embryo, itself or a combination of these