Reversion of cms-S to fertility in the presence of a nuclear restorer gene

Studies have been carried out to identify inbred line backgrounds in which the S1 and S2 mitochondrial episomes characteristic of S-type male-sterile cytoplasm (cms-S) occur in different relative amounts. Only a few nuclear backgrounds are known in which the molar amounts of the S1 and S2 molecules differ (J.R. Laughnan et al., Stadler Symp. 13:93, 1981). In all other inbred line backgrounds examined, S1 and S2 occur in equimolar amounts. A number of the more modern inbred lines not previously available with cms-S have been converted to S-type cytoplasm to determine if they condition S1:S2 differences. None was identified. In the course of these studies a fertile strain of the inbred line H95 (restores cms-S) which pedigree records indicated should carry cms-S was identified as lacking the linear S1 and S2 molecules entirely. This result could be explained by a seed mix with the normal (N) or C cytoplasm versions of H95 (T cytoplasm H95 is sterile) or by reversion to fertility of cms-S. In 2 of the 3 nuclear backgrounds in which cytoplasmic reversion to fertility has so far been characterized, the reversion event results in the loss of free S1 and S2 molecules. No cms-S strains have been observed to lose S1 and S2 and remain sterile. Because the H95 inbred line restores fertility to cms-S, molecular analysis of the mitochondrial DNA (mtDNA) was required to differentiate between seed mix and cytoplasmic reversion.

DNA was isolated from mitochondria of N and C cytoplasm sources of H95 as well as from cms-S W23 and from two H95 fertile S-cytoplasm strains derived by conversion of cms-S W23 to H95, one which maintains S1 and S2 as linear episomes (designated H95S+) and the strain in question which lacks S1 and S2 (designated H95S-). Ethidium bromide stained restriction endonuclease profiles were obtained after digestion of the mtDNA with enzymes HindIII, PstI and SalI. The restriction fragment patterns clearly demonstrated that the H95S- mtDNA was indeed of the S type and no seed mix had occurred. These gels were blotted onto Biodyne nylon membranes and probed with the following DNA fragments: (1) an S1 fragment which contains some homology to S2, (2) an S2 specific fragment, (3) an S1 specific fragment and (4) the cytochrome oxidase I (COXI) gene plus flanking regions. The S1 and S2 probes substantiate the fact that cms-S W23 and H95S + maintain free copies of S1 and S2. They also indicate that integrated copies of S2 and a lesser number of integrated copies of S1 are present. In contrast the mitochondrial genome of the H95S- strain lacks S2 integrated copies as well as both the S1 and S2 free episomes; S1 integrated copies are retained. The COXI probe detects differences between H95S- and its progenitor strain H95S+. In the WF9, 38-11 and M825 nuclear backgrounds, cytoplasmic reversion to fertility also involves reorganization in the COXI region as well as in the S1 and/or S2 regions of the mitochondrial genome.

These data tell us that there has been reorganization of the cms-S mtDNA associated with loss of the S1 and S2 episomes. Loss of this type has been observed previously only in cytoplasmic reversion to fertility. However, it can not be determined by mtDNA studies alone whether reversion occurred in this H95 case. Genetic studies have been conducted on the H95S - strain in order to make this determination. Since the H95 inbred line carries a restorer-of-fertility (Rf) gene the H95S + and H95S- strains are both fertile. Cytoplasmic reversion to fertility must be diagnosed by backcrossing the H95S- strain to maintainers of cms-S. This has been done and the H95S- strain retains complete fertility after 2 backcrosses by maintainer strains. This indicates that the mtDNA changes observed in H95S- were associated with reversion to fertility. This is the first reported case of cytoplasmic reversion to fertility occurring in the presence of restorer genes.

Heidi Feiler, Gracia Zabala, Susan Gabay-Laughnan and J.R. Laughnan

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors.

Return to the MNL 62 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page