Journal of Heredity 1993;84:109 111;0022-1503/93/S4.00

Soybean Genes Rj2, Rmd, and Rps2 in Linkage Group 19

D. G. Lohnes, R. E. Wagner, and R. L. Bernard

From the Department of Agronomy, University of Illinois 1102 S. Goodwin Ave., Urbana, IL 61801. The authors would like to acknowledge K. Kollipara and M. Cho for their assistance in the trypsin classification and bradyrhizobium screening, respectively; we also thank J. E Harper and T. Hymowitz for access to their laboratories and T. E. Devine for identifying Rj2 in the parent lines and supplying culture USDA 7 of Bradyrhizobium japonicum. This research was supported in part by the Agricultural Research Service, USDA.

Abstract

Four genes backcrossed into soybean [Glycine max (L.) Merr.] cv. Williams were evaluated for linkage. The genes investigated were as follows: Rj2, conferring non- nodulation with certain strains of Bradyrhizobium japonicum (Kirchner) Buchanan; Rmd, conferring resistance to Microsphaera diffusa Cooke & Peck (causal agent of powdery mildew); Rps2, conferring resistance to Phytophthora sojae Kaufmann & Gerdemann (causal agent of phytophthora root and stem rot); and Ti controlling a trypsin inhibitor variant in seeds. A cross was made that would segregate for all four gene pairs, and F3 families were classified for the four traits. Chi-square analysis of the data indicated that Ti segregated independently, but that Rj2, Rmd, and Rps2 were linked. Maximum likelihood estimates of gene linkage gave a recombination value of 1.9% for Rj2 and Rmd, 3.6% for Rj2 and Rps2, and 2.3% for Rmd and Rps2. The order of the genes is therefore Rj2 Rmd Rps2.

Introduction

Four soybean [Glycine max (L.) Merr.] loci Rj2, Rmd, Rps2, and Ti have potential economic importance. The dominant allele Rj2 prevents normal nodulation response to three strains (USDA 7, 14, and 122) of Bradyrhizobium japonicum (Kirchner) Buchanan (Caldwell 1966). Rmd-c provides resistance to Microsphaera diffusa Cooke & Peck (causal agent of powdery mildew) (Lohnes and Bernard 1992), as opposed to the more common allele Rmd, which only gives adult plant resistance (Buzzell and Haas 1978), and the third allele Rmd, which conditions susceptibility. Rps2 provides resistance to at least 18 (1-5, 9-20, 27) of the 27 reported races of Phytophthora sojae Kaufmann & Gerdemann (causal agent of phytophthora root and stem rot) (Kilen et al. 1974). Both powdery mildew and phytophthora rot can cause major crop loss in soybeans (Anderson 1986; Phillips 1984). The fourth locus has an allele Ti that eliminates Kunitz trypsin inhibitor in the seed, thus reducing total trypsin inhibitor activity by 30% to 60% (Orf and Hymowitz 1979). Most soybeans have Kunitz trypsin inhibitor present and in either of three variant forms controlled by the codominant alleles Ti-a, Ti-b, and Ti-c at the same locus as ti (Hymowitz and Hadley 1972; Orf and Hymowitz 1977). Devine et al. (1991a) reported the linkage of Rj2 and Rps2 with a recombination value of 2.7% + 1.0% and assigned them to linkage group 19; Devine et al. (1991b) also reported the linkage of Rj2 with Aco2 with a recombination value of 44.7% + 2.1%. Genetic linkage has been reported for Ti with Ap, Lap1, and Pgd2 in linkage group 9 (Palmer and Kiang 1990). Ap, Lap1, and Pgd2 are genes controlling the electrophoretic forms of acid phosphatase, leucine aminopeptidase, and phosphogluconate dehydrogenase, respectively. No other linkages have been reported with these genes.

Based on its phenotype, the widely grown maturity group III cultivar Williams has the alleles Rj2 Rmd Rps2 Ti-a. The gene Rps2 conferring resistance to P. sojae was transferred from a cv. Harosoy isoline to Williams through five backcrosses (Bernard et al. 1991). Evaluation of a selected homozygous Rps2 Rps2 BC line L76-1988 indicated that it had ineffective nodulation with strain USDA 7 of B. japonicum, establishing that the gene Rj2 also was present (Devine TC, personal communication). Subsequent evaluation of L76- 1988 showed that the allele Rmd for adult plant resistance to powdery mildew in Williams had been replaced by the gene Rmd-c for complete resistance. Previously, after backcrossing to transfer the same Rps2 gene to Harosoy, it was noted that the resultant line was resistant to powdery mildew, whereas Harosoy is susceptible (Bernard et al. 1991). The source of the alleles is the cultivar CNS. The gene Ti-b for a trypsin inhibitor variant was transferred from cv. Jefferson to Williams by backcrossing, and a homozygous Ti-b BC line L82-2024 was released (Bernard and Hymowitz 1986). This line was later found susceptible to powdery mildew, which led to our suspicion of a linkage between Ti-b and Rmd Based on this evidence, it appeared likely that all four genes were linked. The objectives of this study were to (a) verify the existence of linkages among the four loci, (b) estimate the linkage intensity and, (c) ascertain the gene sequence.

Materials and Methods

Two isolines of Williams with contrasting alleles at all four loci were cross-pollinated in summer 1988 to produce populations for the linkage study. The isolines used in the cross were the BC5 F3 lines L76-1988 (Rj2 Rmd Rps2 Ti-a) and L82-2024 (Rj2 Rmd Rps2 Ti-b). In summer 1989, we grew F2 plants in the field at Urbana and harvested 246 individual F2 plants. In 1990 we tested the F3 progenies to determine the genotypes of the F2 plants.

Testing for bradyrhizobium response was done with bradyrhizobium strain USDA 7. Strain USDA 7 was maintained on modified yeast extract-mannitol (YEM) agar slants (Bhuvaneswari et al. 1980), transferred to agar plates, and then transferred to a modified YEM broth to produce inoculant. We surface-sterilized seeds with a 10% Clorox solution, rinsed them in distilled water, and planted them in a sand bench that had been steam heated. Plants were inoculated by placing a drop of the broth on each seed at the time of planting. The effectiveness of nodulation was observed approximately 2 weeks after planting. Greenhouse temperature ranged from 24°C to 30°C. Lighting was supplemented with incandescent and fluorescent lamps set on a 14-h photoperiod.

In 1989, after natural infection of powdery mildew occurred in the field, we classified the 245 adult F2 plants, and they fit a 3:1 resistant to susceptible ratio. We tested most F3 progenies for powdery mildew response in the field during summer 1990. Due to short seed supply and poor germination in the field, 88 F3 families were classified in the greenhouse in combination with the classification of nodulation response to bradyrhizobium. We field planted approximately 25 seeds from each F2 plant. We inoculated plants with M. diffusa when the second trifolioliates appeared by brushing them with infected Harosoy leaf tissue. The adult plants were classified in early September. In the greenhouse, we inoculated seedlings just after the unifoliates opened. The plants were classified approximately 2 weeks after inoculation. Greenhouse conditions were the same as for the bradyrhizobium classification.

The polyacrylamide gel electrophoresis procedure for testing seed for the presence of Kunitz trypsin inhibitor variants A (Ti-a Ti-a), B (Ti-b Ti-b), or both (Ti-a Ti-b) has been published (Hymowitz and Hadley 1972). we analyzed a single sample from each F2 plant, obtained by combining chips from five seeds, which gave only a 0.2% chance of misclassifying a heterozygous F2 plant.

Evaluation for Rps2 segregation was done with race 3 of P. sojae on taproots of soybeans grown in aeroponic culture (Wagner and Wilkinson 1992). we inoculated 5-day-old plants by submerging the taproot tip in 500 ml of a solution containing 1,000 zoospores for 20 min. Plants were classified for disease reaction 4 days after inoculation.

In classifying phytophthora and bradyrhizobium response, we screened eight F3 plants to determine which F2 plants were segregating. For F2 plants that appeared to be homozygous dominant, we screened another eight F3 plants to give a probability of 1% for misclassifying a heterozygous F2 as a homozygous F2. Loci were tested for independence using the linkage chi-square (Devine et al. 1991b; Mather 1951). Linkages were analyzed by using the maximum likelihood procedure (Allard 1956).

Results and Discussion

F3 families derived from the cross of L76-1988 x L82-2024 were classified for response to strain USDA 7 of B. japonicum, M. diffusa, and race 3 of P. sojae and for the trypsin inhibitor variant.

The F2:3 segregation ratios for all four traits were not significantly different from the expected 1:2:1 ratio (Table 1). All locus pairs were tested for independent segregation, and a recombination percentage was calculated for locus pairs not segregating independently (Table 2). The Rj2, Rmd, and Rps2 loci were linked closely together, whereas Ti segregated independently. Rj2 and Rps2 were found to have a recombination value of 3.6%, which is similar to the value of 2.7% reported by Devine et al. (1991a). Rmd has a 1.9% recombination value with Rj2 and a 2.3% re- combination value with Rps2. Therefore the order of the genes is Rj2 Rmd Rps2.

Because of the economic significance of phytophthora stem and root rot with its many races, One important use of this linkage would be in the selection of cultivars containing the Rps2 allele. Due to the difficulty in screening for the Rps2 allele, little work has been done using Rps2 resistance. This linkage will allow easy selection for Rps2 by selection for a resistant powdery mildew reaction. Most US soybean cultivars are susceptible (Rmd) to powdery mildew or have only adult plant resistance (Rmd), which are both readily distinguishable in the greenhouse from the complete resistance imparted by Rmd-c. In addition, the powdery mildew resistance itself has economic value in some environments. Single Crossovers are being identified, and isolines are being selected and will be made available for breeding use and genetic studies.

References