"Nonchromosomal stripe" (NCS) refers to unstable, maternally inherited mutants that have drastic and variable phenotypic effects including short stature, reduced viability and vigor, abnormal growth patterns and leaf striping (Shumway and Bauman, Genetics 55:33; Coe, Maydica 28:151). The specific materials studied by Shumway and Bauman, termed NCS, have been lost; however, two others, NCS2 and NCS3, were available for our study. Each of them arose in WF9, or the closely related line H49, with T-type male sterile cytoplasm. The generation of these mutants is directly dependent on the nuclear genotype (NCS-like plants can be generated with any cytoplasmic type in the WF9 background--see Duvick, Adv. Genet. 13:1). However, once present, the inheritance of the defective phenotype appears to be strictly maternal. We decided to examine the organellar genomes of NCS2 and NCS3, to compare them with the progenitor and with normal derivatives of NCS lineages.

Differences in restriction endonuclease fragment patterns between the progenitor mitochondrial DNA (cms-T) and both NCS2 and NCS3 have been seen with two different enzymes, XhoI and PstI. NCS2 has lost an XhoI band at approximately 8 kb and gained a 20.5 kb novel fragment. A new 20 kb XhoI band and a correlated reduction of a 16 kb fragment are observed with NCS3. When chloroplast DNA from NCS3 was examined with the same enzyme no changes in restriction patterns were seen. With PstI, a novel band appears at 7.6 kb in NCS2 mitochondrial DNA, and there is a corresponding loss of a 3.7 kb band. A new PstI fragment of 23 kb has been observed with mitochondrial DNA from NCS3. While an explanation of these results is not yet clear, they do seem to rule out the possibility that NCS2 and NCS3 result from simple losses in specific XhoI restriction sites.

The severity of the NCS3 phenotype is correlated with the relative amount of DNA in the new restriction enzyme band. NCS3 gives a continuum of phenotypes which appear to be quantitatively different expressions of the same basic defect. Affected plants range from nearly normal (tall with few striations, moderately affected) to shorter and heavily striated, with asymmetries of leaf and cob, to severely affected (extremely short, highly distorted, with few stripes but large losses of leaf tissue). The most severely affected plants are so morphologically abnormal that no tassels or ear shoots are formed. Mitochondrial DNA was isolated from the ear shoots of individual NCS3 plants, which were scored for their phenotypic severity. Following XhoI digestion and gel analysis, we found that there is a positive correlation between the severity of the defect and the amount of DNA in the novel 20 kb band. As the amount of DNA in the 20 kb band increases, the amount of DNA in a 16 kb band is correspondingly decreased.

In any NCS3 family, the sib plants are highly variable in their expression of the mutant phenotype. A number of plants in a single, variably-expressing NCS3 family were pollinated by the same male inbred. Subsequently F1 kernels from ears of a normal looking plant and kernels from an ear of a strongly expressing sib plant were planted. The progeny of the normal looking NCS3 plant were all non-mutant in appearance; hence the NCS3 determinants had been lost from this lineage. The progeny of the affected plant showed a range of phenotypes, but most were visibly mutant. MtDNA was isolated from individuals of both the stable, normal plants and from the NCS3-affected plants. XhoI digests showed that the normal plants derived from an NCS3 lineage lack the 20 kb band, whereas affected plants carry this band. The plants used had exactly the same grandmother, and mtDNA in maize is strictly maternally inherited. However, only NCS3-affected plants have the 20 kb XhoI restriction enzyme fragment. We propose that NCS3-affected plants carry a mixture of mutant and normal mitochondrial genotypes, that these genomes sort out during development of the plant and that cells carrying only NCS3 mtDNA die, resulting in stripes and tissue loss. Lineages that contain purely normal mtDNA can no longer sort out defectives.

There is also a correlation of NCS2 phenotypes with the XhoI band differences. An NCS2 derivative plant that had lost the NCS2 phenotype (striping, shortened stature etc.) had lost the 20.5 kb XhoI mtDNA band as well. We therefore propose that the cells of NCS2-affected plants, like those of NCS3, carry a mixture of normal and mutant that sort out during development. If no mutant mitochondrial genomes are represented in the cells giving rise to an ear on an individual plant, the mutant phenotype will no longer be transmitted.

Kathleen J. Newton and Ed Coe
 
 

Return to the MNL 58 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage