Early sectoring and germinal reversion in a Mutator line with the bz2-mu2 reporter allele

--Virginia Walbot

In MNL 65:96-97 I reported the isolation and initial characterization of a Mutator line carrying the bz2-mu2 reporter allele that had exceptionally early sectoring, many large null sectors on kernels, and a 4% frequency of germinal reversion to Bz2. One observation in 1990-91 was that the stock appeared to be losing Mutator activity based on an excess of bz2 kernels and the presence of weakly active (very fine spotting pattern) kernels. Further crosses with large spotted bz2-mu2 and bz2-mu2/bz2 individuals indicated, however, that at least some plants were segregating a single factor required for Mutator activity. Southern blot hybridization indicates that these individuals contain only a single copy of Mu9, our candidate for the regulator of Mutator (Hershberger, RJ et al., 1991, and unpublished data of C. A. Warren, CA). In addition these lines contain only one or a few copies of Mu1, the element in the reporter gene. The segregating bz2-mu2 lines are identical at the genetic level to the a1-Mum2 lines showing single copy inheritance of Mutator activity described by Robertson, DS and Stinard, P (Develop. Genet. 10:282, 1989) and by Chomet et al. (Genetics 129:261, 1991; note that their line had only one Mu1 element in it as well). Our Mu9 probes identify a single copy of a Mu9 element in Robertson's a1-Mum2 line.

The correlation between a single copy of Mu9 and the big spot phenotype is less clear, however. The bz1-mu1 big spot stock has many copies of Mu9 and Mu1. Neither Robertson, DS and Stinard, P nor Chomet et al. comment on early somatic sectoring in the a1-Mum2 lines. This feature in the bz2-mu2 big spot line is completely atypical of standard Mutator lines and would be readily noticed with any anthocyanin reporter allele. It is possible that an additional factor--the background of our lines, the "phase" of Mu9, or another gene--is responsible for the early sectoring phenotype. In crosses with other Mutator lines containing diverse reporter alleles, the big spot phenotype is "dominant" if the big spot line is used as the female. In crosses in which the F1 progeny were purple and could not be scored for excision behavior, i.e. Bz2/bz2-mu2 A2/a2-mu1, F2 crosses to the respective testers showed retention of the big spot phenotype on many bz2 ears in the W23 background and sporadic occurrence in the a2 crosses in a different inbred line. Examples of the big spot phenotype transferred to lines with other reporter alleles are being grown now to test for transmission to the next generation.

As for the timing of somatic and germinal excision events, bz2-mu2 contrasts sharply with the bz1-mu1 big spot line (this volume). First, early excision of bz2-mu2 is visible in the plant body with B pl as large stripes, tassel branches, clusters of anthers, individual anthers and in one instance a half-tassel on a tiller. The half-tassel produced just over 50% revertant progeny in a cross to bz2. Plants with several large sectors in the tassel also produced many germinal revertants. These observations suggest that in the bz2-mu2 line some excision events occur in the apical meristem cells or their immediate derivatives to produce large sectors. Many small sectors are also present in the soma and in the aleurone indicating that events occur throughout development. Second, if there is a bias between the ear and tassel in producing germinal revertants it is more subtle than in the bz1-mu1 line and much more complicated to follow as revertants do not follow a Poisson distribution.

Third, in selections of independent Bz2' revertants, picked as individual kernels from different bz2 X bz2-mu2/bz2 crosses, I found non-concordance between the endosperm phenotype (purple) and embryo genotype (bz2 or bz2-mu2) 19% of the time (15 cases/79 trials). This means that about one-fifth of the germinal revertants result from events in pollen that produce sperm of two genotypes. There are several possible scenarios. Excision could occur in one sperm nucleus but not the other. If true, then because there is 4% reversion to purple by scoring the aleurone we should expect that among the kernels with a spotted aleurone, there should be a similar percentage with a Bz2' embryo. Such an embryo should yield a revertant plant, i.e. purple body and 50% purple progeny kernels (parent was Bz2'/bz2 by virtue of the original cross). To date, I have recovered only two such cases, too few given the populations examined to conclude that excision in individual sperm nuclei is the only mechanism underlying non-concordance.

But consider another surprising fact: 9/15 of the non-concordant cases transmit bz2 not bz2-mu2. As the pollen is already haploid, the Bz2' revertant allele must arise from the mutable allele. This means that in the majority of non-concordant cases and in 11% of the total population (9/79) bz2 alleles are created in the pollen by excision or by loss of Mu9. The excision hypothesis could encompass cases [1] that involve "mistakes" in DNA repair occurring from a single excision event in the germinal nucleus that are repaired to Bz2' or bz2 as a consequence of DNA replication prior to sperm formation and [2] a contribution from excision in individual sperm nuclei that leads to either Bz2 or bz2 alleles. Both scenarios predict that among the kernels with a bronze phenotype, there will be some with a Bz2' embryo. The number of Bz2' embryos among progeny with spotted and bronze-colored aleurones will be determined in a large population in 1992. The excision scenario also predicts that all of the bz2 cases will represent Mu1 excision events from the reporter allele; this can be tested by Southern blotting. The Mu9 segregation, or change of phase, scenario will be tested on these same Southern blots; in this scenario the bz2 kernels are actually bz2-mu2 but Mu9 is not present or is non-functional. It may also be instructive to test what fraction of the Bz2' derivatives also lack Mu9.

The 6/15 non-concordant cases that transmit bz2-mu2 most likely represent the true, independent cases of excision in individual sperm nuclei (6/79 = 8%). This is a high percentage of events in a single cell type. Concordant cases (81%, 64/79) represent the majority of cases. These most likely represent excision events that precede pollen formation or events that occur early in gametophytic development and hence result in a single allele in the sperm. As the big spot bz2-mu2 stock produces purple anthers and tassel sectors of all sizes, it is likely that excision events occur throughout tassel development.

Return to the MNL 66 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage