III. 1. Utilization of sexl msg6 stocks for hybridization in barley.
D. E. Falk, E. Reinbergs and K. J. Kasha, Department of Crop Science, University of Guelph, Guelph, Ontario, Canada. "R"
The use of presowing selection of genetic male sterile plants for hybridization in barley has been proposed by Falk and Kasha (1980) and further elaborated on by Falk, Kasha and Reinbergs (1981). This system is built around the tight linkage of the shrunken endosperm gene which expresses xenia (sexl) and the male sterile gene (msg6) located near the centromere of barley chromosome 6. Upon selfing, plants heterozygous for the shrunken endosperm gene (Sexl/sexl) produce a 3:1 ratio of plump seeds containing at least one dominant allele and shrunken endosperm seeds which are homozygous for the recessive allele (sexl/sexl). Since the msg6 allele has been coupled to the sexl allele, the shrunken endosperm seeds will all produce male sterile plants and the plump seeds will all produce fertile plants, except for the rare crossovers between the two loci. The shrunken endosperm seeds can be separated before sowing to give populations of all male sterile plants which can be used as the female parent in hybrid seed production. The results of using this system to produce outcrossed seed to initiate recurrent selection populations are reported in this paper.
Hand sorting of shrunken endosperm seeds has consistently given populations containing 95-97% male sterile plants. An average of 1.2% of the selected seeds have been crossovers (sexl Msg6/sexl msg6) and another 1.6% have been misclassified seeds (Sexl Msg6/sexl msg6), both giving fertile plants in the selected population. Seed set on the male sterile plants has been poor in the Guelph environment where the summers are quite hot and humid.
In an effort to produce more outcrossed seeds in a favorable environment, shrunken endosperm seeds were selected from the germplasm GBC 695 (USDA GP 59), (see Falk and Kash, 1982) to be included in the winter nursery grown at Brawley, California in 1980-81. The source population (GBC 695) is homozygous for the orange lemma gene (o) and segregating for shrunken endosperm (sexl) and male sterile (msg6) in coupling and uniculm (uc2) in repulsion (o msg6 sexl Uc2/o Msg6 Sexl uc2). Eight hundred shrunken endosperm seeds were space planted in two row plots alternately with non-male sterile F2 breeding populations. Thus, the pollen parents were homozygous dominant at the 0, Sexl, Msg6 and Uc2 loci.
Approximately 23,000 F1 seeds were harvested from the plots planted to shrunken endosperm seeds using conventional plot equipment. No effort was made to increase pollen dispersal. This gave a return of 29 outcrossed seeds for each shrunken endosperm seed planted. Fertile plants were not classified in these populations as they couldn't be detected at harvesting. No uniculms were observed.
The F1 populations from the seed harvested in California were grown at Guelph in 1981. Nearly 3000 plants were classified for origin (based on phenotype) at maturity. All outcrossed seeds should have produced plants carrying the dominant O allele for non-orange lemma from the normal pollen parents. Those seeds originating through selfing of a plant carrying a crossover or included due to a misclassification, would have been homozygous for the orange lemma (oo) alleles. Plants were also classified for male sterility, shrunken endosperm and the uniculm trails. Over 96% of the F1 plants were a result of outcrosses to the F2 breeding populations (Table 1).
If misclassified seeds (o sexl msg6 Uc2/o Sexl Msg6 uc2) had been grown in California, they should have given selfed progeny in a ratio of 1 oo sexl sexl msg6 msg6 Uc2 ___: 2 oo Sexl sexl Msg6 ___ Uc2 ___ : 1 oo Sexl Sexl Msg6 ___ uc2 uc2. The 23 oo sexl sexl Msg6 ___ Uc2 ___ plants observed in Guelph should have also had 12 oo sexl sexl msg6 msg6 Uc2 ___ and 12 oo Sexl Sexl Msg6 ___ uc2 uc2 siblings. Crossover plants (o sexl msg6 Uc2/o sexl msg6 Uc2) in the California population should have produced a ratio of 1 oo sexl sexl msg6 msg6 Uc2 : 3 oo sexl sexl Msg6 ___ Uc2 ___ progeny. Thus, the eight oo sexl sexl Msg6 ___ Uc2 ___ plants observed at Guelph should have had 3 oo sexl sexl msg6 msg6 Uc2 ___ siblings. Therefore, a total of 15 oo sexl sexl msg6 msg6 Uc2 ___ plants and 12 oo Sexl Sexl Msg6 ___ uc2 uc2 plants would have been expected in the classified population. The observed frequencies were 66 and one, respectively.
Competition from the hybrid plants may have crowded out the uniculm plants so that they were well below the expected frequency at harvesting.
On the other hand, the 66 orange lemma, male sterile plants observed is more than four times the number which can be accounted for, based on observed crossover and misclassified phenotypes.
There may be two explanations. The first is from "leakage" of the msg6 msg6 homozygotes. It is possible that some of the pollen grains of the male sterile plants may occasionally function to give seeds. This phenomenon has been reported for Heines Hanna msg6 (the original source of our msg6) by Hockett and Eslick (1968). A population related to GBC 695 has also given oo sexl sexl msg6 msg6 plants which show some selfing ability indoors (138 seeds on 137 covered spikes). The 51 "extra" orange lemma, male sterile plants in the classified population suggest that the msg6 msg6 genotype may be expressing some fertility (leakage) in the California winter environment. Secondly, pollen from misclassified or crossover plants could sib pollinate and account for such plants. However, given the low number of crossovers and misclassified plants in relation to the F2 breeding populations, the chances of such sib pollination are quite low.
Leakage may be a serious drawback to the use of sexl msg6 stocks for hybrid cultivar production where the male sterile plants would show up in the farmer's fields. However, the use of this system of presowing selection of male sterile plants in a breeding program may not be seriously affected. Selfed seeds from "leaky" male sterile plants will be homozygous for the orange lemma gene (oo) and so can be easily rogued in the field or at harvesting. It may also be possible to separate the selfed, shrunken endosperm seeds from the crossed, plump seeds before planting the F(1) populations.
Shrunken endosperm F(2) seeds were hand selected from the F1 outcrossed plants grown in Guelph and sent back to California for a second cycle of outcrossing with normal breeding materials. Remnant F(2) seed from the 1980-81 crosses is available for distribution, as is the source germplasm GBC 695 (GP 59) used to develop these populations.
Preliminary tests indicate that it may also be possible to separate shrunken endosperm seeds from plump seeds based on specific gravity differences. This would make the use of presowing selection techniques much more efficient than hand sorting. Mass selection for or against shrunken endosperm seed would give male sterile facilitated recurrent selection and open-pollinated barley populations which are easily maintained and manipulated for outcrossing or elimination of the male sterility, as desired.
References:
Falk, D. E. and K. J. Kasha. 1980. A shrunken endosperm, male sterile stock to facilitate crossing in barley. BGN 10:81-83.
Falk, D. E. and K. J. Kasha. 1982. Registration of a shrunken endosperm, male sterile germplasm to facilitate hybridization in barley. Crop Science 22: In Press.
Falk, D. E., K. J. Kasha and E. Reinbergs. 1981. Presowing selection of genetic male sterile plants to facilitate hybridization in barley. Barley Genetics IV: In Press.
Hockett, E. A. and R. F. Eslick. 1968. Genetic male sterility in barley. I. Non-allelic genes. Crop Sci. 8:218-220.
