|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
a Dep. of Plant Sciences, North Dakota State Univ., Fargo, ND 58105-5051
b Dep. of Plant Pathology, North Dakota State Univ., Fargo, ND 58105
* Corresponding author (ted.helms{at}ndsu.edu).
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: G x E, genotype-by-environment
a Dep. of Plant Sciences, North Dakota State Univ., Fargo, ND 58105-5051
b Dep. of Plant Pathology, North Dakota State Univ., Fargo, ND 58105
* Corresponding author (ted.helms{at}ndsu.edu).
There are differences among soybean [Glycine max (L.) Merr.] genotypes for tolerance to water-saturated soil conditions. Phytophthora root rot caused by Phytophthora sojae (M.J. Kaufmann and J.W. Gerdemann) is problematic for soybean when the soil is excessively wet. Our objectives were (i) to determine if genotype-by-environment (G x E) interactions existed for yield between water-saturated (irrigated) vs. control (nonirrigated) environments using 37 soybean genotypes; and (ii) to determine whether resistance and/or partial resistance to P. sojae was associated with tolerance to water-saturated soil conditions. The G x E was significant between the water-saturated treatment and control. Some genotypes that yielded well under control conditions did not yield well under saturated conditions. Based on yield, we identified eight genotypes that were tolerant and eight genotypes that lacked tolerance to water-saturated soil conditions. These 16 genotypes were evaluated for resistance and partial resistance to Phytophthora using three virulence phenotypes. Those genotypes that were tolerant to water-saturated soil conditions possessed either a major gene or good partial resistance to P. sojae. We concluded that resistance to P. sojae was an important factor that increased tolerance to water-saturated soil, but that unknown other genetic factors were also important.
Abbreviations: G x E, genotype-by-environment
| INTRODUCTION |
|---|
|
|
|---|
Plants are adversely affected in several ways when a soil becomes water saturated. Soil pores ordinarily comprised of oxygen are replaced with water, restricting the flow of oxygen through the soil (Pezeshki, 1994). The remaining oxygen left in the soil is readily depleted through the respiration of plant roots and soil microorganisms. Once a soil becomes anaerobic, adverse effects on plants occur such as chlorosis, reduced growth rate, reduction in mineral uptake, altered growth regulator relationships, stomatal closure, leaf wilting and epinasty, reduced photosynthesis and respiration, altered carbohydrate partitioning, and potentially death.
Yield losses due to saturated soils during July, have been reported to be approximately 25% after a 4-wk flooding period (VanToai et al., 1994). However, a 20% yield loss has been reported when soybean [Glycine max (L.) Merr.] plants were subject to only 3 d of saturated soil conditions during V2 and V3 stages (Fehr and Caviness, 1977) of soybean development (Sullivan et al., 2001). Scott et al. (1989) concluded that soybean yield is more affected by saturated soil conditions that occur during the reproductive stages than at vegetative stages. Regardless of the stage of development, stress from water-saturated soils produces adverse affects on soybean yield.
Phytophthora root and stem rot is the second leading biotic cause of yield loss in soybean (Wrather et al., 2001). This disease is associated with heavy-clay, poorly drained soils, which are common in the Red River Valley of eastern North Dakota and western Minnesota. Once the pathogen is present in the soil, it can never be eradicated. There are no economical and effective chemical controls for this disease on adult plants. Therefore, genetic resistance to the disease has become an important management strategy.
Single dominant Rps genes (Bernard et al., 1957; Hartwig et al., 1968) have been adequate for genetic control of P. sojae in North Dakota. However, changes in the virulence of the pathogen population in this area have resulted in Rps1c and other alleles becoming less effective as sources of resistance (Nelson et al., 2005). As a result, the current focus on future management of P. sojae has been to identify genotypes that display tolerance or partial resistance to the disease.
Tolerance and/or partial resistance to P. sojae has been described as the ability of plants to survive root infection without displaying severe disease symptoms such as death, stunting, or yield loss (Schmitthenner, 1985). The terms tolerance and partial resistance have been used interchangeably throughout the literature on P. sojae since the 1980s. Partial resistance is the term that has been generally accepted in recent literature and will be used throughout this manuscript.
Walker and Schmitthenner (1984) developed the inoculum-layer method to screen soybean seedlings for partial resistance to P. sojae. McBlain et al. (1991) evaluated several methods to determine the partial resistance of genotypes. They reported that both the inoculum-layer and the slant-board methods are useful for screening for partial resistance, but these tests only measure partial resistance at the seedling stage. Field evaluation was preferred to the greenhouse methods because it was less expensive and would screen for partial resistance mechanisms that might be present only in older plants.
Our objectives were (i) to determine if G x E interactions exist for soybean between water-saturated vs. rainfed environmental conditions and (ii) to determine whether resistance and/or partial resistance to P. sojae was associated with tolerance to water-saturated soil conditions among genotypes.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Overhead irrigation was used to create the saturated soil conditions for 14 d at the reproductive stages of about R1 to R3. Adequate and consistent soil saturation conditions were achieved and maintained. Water-saturated soil conditions were monitored by the use of piezometers. We developed our own piezometers with materials purchased locally. Redox potential was monitored with platinum-tipped electrodes placed at 15 and 60 cm depths to describe the aerobic/anaerobic environment. The piezometers and electrodes were placed at two equidistant locations in each main plot. While soil redox potentials of the nonsaturated soil were consistently in a range between 400 and 450 mV, the values in flooded soils averaged about 100 mV.
Soybean Genotypes
Sixteen soybean genotypes were selected, based on the results of the saturated soil experiments conducted in Fargo, ND, during 1999, 2000, and 2002. Selections were based on yield in water-saturated soil conditions. We will use the term "tolerance to water-saturated soils" to mean high yield under those conditions. The eight genotypes that were the highest yielding under water-saturated field conditions were considered tolerant. The eight genotypes that were lowest yielding under those conditions were considered intolerant to water-saturated field conditions.
Genotypes tolerant to water-saturated soil conditions included: Northstar Genetics 0923RR, Mycogen 5121, Hyland Seeds Sentry, Croplan Genetics L1187, Wensman Seed W3139N, Peterson Farm Seed Korada, Monsanto CX105, and Asgrow 0801. Genotypes intolerant to water-saturated soil conditions included: Stine Seed 1101-6, Pioneer Hi-Bred 9071, South Dakota State University Surge, Mustang Seed 0700, North Dakota State University Barnes, Hyland Seeds RR Rattler, University of Minnesota MN0301, and University of Minnesota Lambert.
To aid in the assessment of partial resistance to P. sojae, the check genotypes Conrad, Sloan, and OX20-8 were obtained from Ohio State University. In standard tests for partial resistance, the average disease ratings for Conrad, Sloan, and OX20-8 were 3.5, 6, and 9, respectively (Dorrance and Schmitthenner, 2000; McBlain et al., 1991). The scoring for partial resistance is on a 1 to 10 scale with a low score indicating less disease.
We determined if there were major Rps genes in these genotypes since Rps genes can influence an evaluation of partial resistance. Determination of the Rps genes in each genotype was based on either the pedigree information of that genotype, the information provided by the company, or on hypocotyl-inoculation tests (Haas and Buzzell, 1976) that we conducted with select races. When there was a discrepancy between the information provided by the company marketing the genotype and the results of our test, we used the results of our hypocotyl-inoculation test.
Virulence of Phytophthora sojae Isolates
Three isolates of P. sojae with different virulence phenotypes (the Rps genes defeated by the pathogen) were used in the majority of these experiments. These isolates were originally baited from North Dakota soils and previously tested for virulence on soybean (Nelson et al., 2005). The isolates were grown on V-8 agar (0.5 g sucrose, 0.1 g yeast extract, 20 mL V-8 juice, 0.3 g calcium carbonate, 10 g agar, and 480 mL distilled water) at 23°C and stored on V-8 agar in sterile water banks.
Eight standard differential genotypes with known single genes (Rps1a, Rps1b, Rps1c, Rps1d, Rps1k, Rps3a, Rps6, and Rps7) (Wagner and Bernard, 1991) were used to verify the virulence phenotypes of isolates of P. sojae used in this study (Hartman et al., 1999). A potting soil mixture consisting of 50% dry soil, 25% premium coarse grade vermiculite, and 25% sunshine mix was moistened and pasteurized. Sixteen soybean seeds of each differential genotype were planted into potting mix in 470-mL plastic cups (eight seeds per cup and two cups) with drainage holes and placed in a greenhouse. When plants were 8- to 9-d old with the unifoliolate leaves unfolded, plants were inoculated using the hypocotyl-injection test developed by Haas and Buzzell (1976). Plants were maintained in a growth room at 24 ± 2°C for 10 d and then evaluated for resistance/susceptibility.
The virulence phenotypes of the isolates were as follows: Phenotype I = Rps1a, Rps1c, Rps1d, Rps7; Phenotype II = Rps1a, Rps1b, Rps1k, Rps7; and Phenotype III = Rps1a, Rps1c, Rps7. When we initiated these studies, we did not have an isolate with a virulence phenotype that would defeat Rps6. However, following the evaluation of the genotypes with the three isolates, we obtained an isolate with a virulence phenotype of Rps1a, Rps1c, Rps1d, Rps6, Rps7. This isolate, Virulence Phenotype IV, was used specifically to test the partial resistance of Barnes and Sentry, two genotypes with Rps6. Both Barnes and Sentry are susceptible to Virulence Phenotype IV when inoculated using the hypocotyl-injection test.
Determination of Partial Resistance
The inoculum-layer method developed by Schmitthenner and Bhat (1994) was used to screen for partial resistance to P. sojae in the greenhouse. The pathogen was grown on V-8 agar (20 mL inoculum in 15 by 100 mm Petri dishes) at 23°C in the dark until the mycelium covered the agar surface. One liter Styrofoam containers with three 5-mm diameter drainage holes in the bottom were filled with 11 cm of premium coarse-grade vermiculite and watered thoroughly. The disk-shaped agar cultures were removed from the Petri dishes and one culture was placed in each container with the mycelium side up. The culture was slightly greater in diameter than the container, thus creating a seal along the inner wall of the container. The noninoculated controls received a disk of V-8 agar with no pathogen. The cultures were then covered with 2 to 5 cm of vermiculite and lightly dampened. Fifteen seeds of each genotype were placed on the vermiculite surface within each container and covered with 2 cm of vermiculite and watered again.
Plants were grown for 3 wk in the greenhouse. Average daily temperatures ranged from 18 to 25°C. Natural daylight was supplemented to 16 h d–1 with high pressure sodium lamps, providing an irradiance of 200 W m–2. An automatic watering system supplied water three times daily to maintain a high moisture level in the vermiculite.
We used three criteria to measure partial resistance: emergence, survival, and a disease rating. Plant emergence was recorded 1 wk after planting and was based on the total number of plants that showed cotyledons and/or the first trifoliolate leaves above the vermiculite surface in each container. Plant survival was recorded 3 wk after planting as the total number of living plants per container.
Following the 3 wk growth period, the entire plant mass within the container was removed from the vermiculite and roots were rinsed with warm tap water. Disease was rated on the entire plant mass using a 1 to 10 scale developed by Schmitthenner and Bhat (1994): 1 = no root rot; 2 = trace of root rot; 3 = bottom third of root mass rotted; 4 = bottom two-thirds of root mass rotted; 5 = all roots rotted, 10% seedling kill, and slight stunting of tops of plants; 6 = 50% seedling kill and moderate stunting of tops of plants; 7 = 75% seedling kill and severe stunting of tops of plants; 8 = 90% seedling kill; 9 = all seedlings dead; and 10 = all seedlings killed before emergence. Development of disease to levels previously reported for the check genotypes indicated that the inoculations were successful.
Preliminary experiments were conducted in a growth chamber to determine emergence and survival rates of noninoculated seedlings and to determine that noninoculated plants grew normally for each genotype. The emergence rates among noninoculated genotypes were significantly different; therefore, the emergence and survival from inoculated treatments was adjusted to percentages by dividing by the emergence and survival of the noninoculated control of the corresponding genotype. The residuals were evaluated to see whether they fit a normal distribution using the Proc Univariate procedure of the SAS Institute (1985). The residuals of the percentage data fit a normal distribution without the need for transformation.
Greenhouse Experimental Design
Each greenhouse experiment was a randomized complete block with four replicates. Three separate experiments were conducted with Virulence Phenotypes I, II, and III of the pathogen. Each experiment consisted of one virulence phenotype and 19 genotypes and was run three times. The three check genotypes were included in each experiment to verify disease pressure. Genotypes and virulence phenotypes were considered fixed effects, while experiments and blocks were random effects.
The eight genotypes tolerant to water-saturated soil conditions and the eight genotypes intolerant to water-saturated soil conditions were considered to be nested within tolerance levels. The three check genotypes, Conrad, Sloan, and OX20-8, were not included in the analysis of variance that included the 16 genotypes. Conrad, Sloan, and OX20-8 were evaluated in a separate analysis of variance. Tolerance to water-saturated soil conditions in the field was considered a fixed effect. Residual mean squares for each repetition of each experiment (Run) were homogeneous; thus, data were combined across runs and experiments.
A separate experiment was conducted to test partial resistance of Barnes and Sentry to Virulence Phenotype IV. The same methodology and design as previously described was used except the experiment had three replicates and was repeated. The three check genotypes were included to verify a successful test.
Statistical Procedures
For field experiments, analyses of variance were computed for each experiment and combined across years. Both the water treatment and genotypes were considered fixed. The year and block sources were treated as random. The method of Carmer et al. (1989) was used to calculate the LSD for comparing the same genotype across water treatments. Expected mean squares were equated to mean squares to determine the proper denominator of the F-tests.
In the greenhouse experiments, we determined whether partial resistance was greater for tolerant genotypes vs. intolerant genotypes by eliminating the confounding effects of major Rps genes. t tests were performed to compare the mean of tolerant genotypes that lacked a major Rps gene that was effective on that virulence phenotype to the mean of intolerant genotypes that also lacked a major Rps gene effective on that virulence phenotype. The t tests for Virulence Phenotype I excluded genotypes with Rps genes 1k and 6. The t tests for Virulence Phenotype II excluded genotypes with Rps genes 1c and 6 and the t tests for Virulence Phenotype III excluded genotypes with Rps genes 1k and 6. These t tests used the genotype x tolerance x run x virulence phenotype as the mean square in the denominator with unequal numbers in each mean.
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
|
|
Role of Resistance to Phytophthora sojae
The mean emergence (P = 0.05), survival (P = 0.01), and disease rating (P = 0.01) of the eight genotypes that were tolerant to water-saturated conditions was significantly different from the mean of genotypes that were intolerant to water-saturated conditions (Table 3
). The mean emergence, survival, and disease rating of the tolerant genotypes were 88%, 79%, and a score of 4.5, respectively. The mean emergence, survival and disease resistance of the intolerant genotypes were 82%, 63%, and a score of 5.4, respectively. This is evidence that those genotypes tolerant to water-saturated soil conditions also have either major gene or effective partial resistance to P. sojae. The virulence phenotype x genotype within tolerance level interaction was significant for disease rating (P = 0.01), emergence (P = 0.01), and survival (P = 0.01). This is evidence that the effect of the virulence phenotype depended on both the major gene and level of partial resistance in the genotypes.
|
|
|
|
Sentry had a high level of partial resistance compared with Barnes when both genotypes were tested with Virulence Phenotype IV. The mean emergence and survival (percent of control) and disease ratings for Sentry averaged over two experiments were 90%, 70%, and 4, respectively, while for Barnes they were 73%, 7%, and 8. Sentry therefore showed a high level of partial resistance to an isolate of P. sojae that can defeat Rps6 while Barnes showed no partial resistance.
Temperature is known to affect the expression of resistance in soybean to P. sojae. Gijzen et al. (1996) demonstrated that during a high temperature of 33°C, certain Rps genes in soybean were found to be ineffective in controlling Race 1, while at 25°C they were effective. In our experiment, each run of each experiment (virulence phenotype) was conducted at different times in the greenhouse and temperatures varied. However, there was no evidence from the results that the differences in temperature between runs of the same experiment, or among different experiments affected the results (data not shown). There was no trend across time, indicating no increase or decrease in the means of the dependent variables for the runs of the experiments. Also, the fact that the three check genotypes performed as expected across all three experiments and runs indicated that temperature differences had no major effect on the results.
Association of Partial Resistance with Tolerance to Water-Saturated Soil Conditions
There were differences in survival, emergence, and disease rating between the mean of the eight genotypes that were tolerant and the mean of the eight genotypes that were intolerant to water-saturated soil conditions (Table 3). However, some genotypes in each of these two groups have major Rps gene resistance to P. sojae. Therefore, the difference in disease rating, emergence, and survival between the tolerant and intolerant groups is mostly likely due to both major gene resistance as well as partial resistance to P. sojae.
To accurately assess the association of partial resistance with tolerance to water-saturated soil conditions, t tests were performed to eliminate the confounding effects of those major Rps genes (Table 7 ). When genotypes containing effective Rps genes were not included in the mean comparison of tolerant vs. intolerant, plant emergence of the tolerant genotypes was equal to that of the intolerant genotypes for Virulence Phenotypes I, II, and III (Table 7). When the same test was conducted for survival and disease rating, tolerant genotypes had greater survival and lower disease ratings than intolerant genotypes for Virulence Phenotypes I and II, but not III. We cannot explain why the average disease ratings were the same between the tolerant and intolerant groups for Virulence Phenotype III. We expected that partial resistance would more likely be detected with the disease rating data, than with emergence and survival, especially since four of the tolerant genotypes had the Rps1k gene that would confer resistance to Virulence Phenotype III. These results show that the level of partial resistance to P. sojae depended on which virulence phenotype was used in the evaluation. This study provides evidence that partial resistance is associated with increased tolerance to water-saturated soil conditions at the Fargo, ND, site. This field site has a history of P. sojae with race 3 the predominant race, but other races have been identified from infected plants.
|
Races 3 and 4 are the most common races of P. sojae in North Dakota (Nelson et al., 2005). Genotypes that were tolerant to water-saturated soil conditions either had an Rps gene that would defeat Race 3 or 4 of P. sojae, or else they had good partial resistance to Virulence Phenotypes I and II. Genotypes that were intolerant to water-saturated soil conditions generally lacked a Rps gene that conferred resistance to Races 3 and 4 or had less partial resistance than tolerant genotypes (Table 7).
Barnes contains the Rps6 gene and showed a moderate level of resistance to Virulence Phenotypes I to III in the partial resistance tests. This resistance was expected since Rps6 would confer resistance to those virulence phenotypes. When Barnes was challenged with Virulence Phenotype IV which defeats Rps6, there was no evidence of true partial resistance. Water-saturated soil conditions reduced the yield of Barnes to only 55% of the yield of that genotype compared to the control conditions (Table 2). Barnes was one of the genotypes that were least tolerant to water-saturated conditions. Yet Barnes had major gene resistance to races 3 and 4, the predominant races of P. sojae in that field. This exception showed that major gene and/or partial resistance is not the only factor required to provide tolerance to water-saturated soil conditions.
VanToai et al. (1994) stated that "flooding tolerance does not appear to be related to nonflooded yield, Phytophthora tolerance, Phytophthora resistance, or relative maturity in soybean". Our study, however, showed that at a site where P. sojae was present, resistance and/or partial resistance to P. sojae was one factor that influenced tolerance to water-saturated soil, but it was not the only factor.
| NOTES |
|---|
|
|
|---|
Received for publication March 28, 2007.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |