Central cell incongruity as a barrier to introgression in (Allium cepa L. x A. fistulosum L.) x A. cepa backcross F1BC3 generations
Polyacrylamide gel electrophoresis (PAGE) was used to study the inheritance of esterase isozyme (EST) phenotypes in Allium fistulosum L. Three A. fistulosum specific loci were identified from the most anodal zones of acitivity, EST-1, EST-4, EST-6. Each locus was found to have two alleles. Both EST-1 and EST-4 encode a null allele. Plants homozygous for the nuli atieies lacked expression of the respective band on the PAGE gel. Starch gel electrophoresis was used to study the inheritance of the Phosphoglucoisomerase (PGI) phenotype. Pgi-1 in A. fistulosum has two alleles, Pgi-1^, and Pgi-1'*. Goodnessof-fit of these alleles to Mendelian ratios was tested in an F2 population with the chi-square statistic. Linkage analysis was also conducted befr.vscn the new alleles presented in this report and with previously identified alleles of ADH, 6-PGDH, and PGM. An incongruity model is presented as a possible expiaiiaiiun for an anomaly in the germination and establishment percentages, as well as a distorted isozyme segregation pattern, observed in F2BG3 seed of an A. cepa x A. fistulosum interspecific hybrid. The F1BG3 parent expressed Pgi-1^, PGI alleles of both species; Pgi-1^ coming from A. cepa (Ac) and Pgi-1^ allele from A. fistulosum (Af). Data collected on the F2BG3 progeny population were percent germination and establishment, segregation of Ac and Af Pgi-1 alleles, and viability and genotype of non-germinated seed, seed that was not planted, and isolated embryo and endosperm tissue. Seed, embryo, and endosperm tissues exhibited 100% viability, despite lower germination and establishment peicentayes.. A pooled goodness-of-fit test of the segregation of Pgi-1 alleles in the populations to the expected Mendelian 1:2:1 ratio using the chi-square statistic gave a x^ = 185.9, well beyond the accepted limits at 2 degrees of freedom, and the 1:2:1 ratio was rejected. Another pooled chi-square goodness-of-fit test of the segregation of Pgi-1 alleles in the populations to a 1:1 ratio based on the incongruity model gave a x^ = 0.203, within the accepted limits at 2 degrees of freedom and the 1:1 ratio was not rejected. Chi-square tests of data from other populations reporting similar to PGI isozyme distortion suppoit the hypothesis.