Genes interaction in maize endosperm development

In this report we present the results of a complementation test involving nine emp (empty pericarp) mutants of maize representing single gene mutants, isolated as independent events. These mutants are embryo lethal and drastically reduced in their endosperm size. As to the endosperm, they can be subdivided into two major subgroups: one including those with a flat appearance of the kernel and smooth pericarp and another one with a wrinkled pericarp. In all mutants, the analysis of longitudinal sections of mature seeds reveals absence of morphogenesis in the embryo proper, an observation the correlates with the failure of these mutants to germinate. By crossing inter-se plants heterozygous for the nine emp mutants we identified those non complementing (i.e. allelic) and those complementing (i.e. not allelic) in the F1 generation. Most results in the F1 were concordant to those obtained in the F2 generation with the exception of three cases where the F1 results suggest allelism (i.e. one gene) whereas those in the F2 suggest segregation of two genes. These intriguing results seem to suggest an interaction between different emp mutants, attributable to a phenomenon that is often referred to as second site non-complementation (SSNC). SSNC can be explained by assuming interaction between two different mutant proteins leading to a toxic product or that the mutant form of one protein sequesters the wild-type form of the other protein into an inactive complex. In some cases of 9 to 7 segregation in the F2 generation we recovered the two phenotypes (emp flat and emp wrinkled) with values fitting the 9:4:3 ratio suggesting the presence of an epistatic interactions between the two mutants. For example, in the case of the cross between emp4 and emp*-9475 the scoring of F2 ears revealed a segregation ratio of 9:4:3 (wt: emp wrinkled: emp flat) in which we can assume that emp*-9475, that has a wrinkled phenotype, is epistatic to emp4, that has a flat phenotype. However the data need to be increased to demonstrate that the hypothesis postulated is correct. Another important point was the finding that in some ears segregating for a single emp mutant in different genetic backgrounds (A636, W23 and Mo17) some mutant seeds were identified exhibiting a more abundant endosperm tissue and occasionally an embryonic axis. A low percentage of these seeds germinate. This observation could be explained in two ways. The first implies that emp mutants uncover a cryptic variability and the second possibility is that the improved endosperm of the mutant is the result of an interaction between the mutant and a second factor originally present in the inbred line. The results of crosses made to test for a segregation of a modifier have allowed us to identify a dominant and a recessive modifier. If these results are confirmed, emp mutants could be used as a tool for the detection of genetic factors enhancing the amount of endosperm in the maize kernel to be exploited in breeding programs