Mutator system-controlled instability of the Cg2 locus

--N. V. Krivov

[ed. note: this note, received after the 1991 issue was printed, overlaps with the next article but contains supplemental information and photographs.]

Morphogenetic mutants are a most suitable object for gaining insight into regulatory principles of plant development (Sawhney, VK and Greyson, RI, Am. J. Bot. 60:514-522, 1973; Koornneef, M et al., J. Hered. 74:265-272, 1983). In maize it is expedient to use semidominant mutations (Tp1, Tp2, Tp3 and corngrass) to study mechanisms underlying organogenesis of a plant as a whole as well as that of inflorescences (Lindstrom, EW, J. Hered. 16:135-140, 1925; Peterson, H, MNL 33:41, 1959; Singleton, WR, MNL 23:7, 1949; Sheridan, WF, Annu. Rev. Genet. 22:353-385, 1988). These mutations cause profound atavistic changes in the morphology of vegetative organs as well as transform reproductive structures into leaves (Singleton, WR, Amer. Nat. 85:88-96, 1951; Weatherwax, PJ, J. Hered. 20:325-330, 1929). A phenotype like this has been shown to be due to a defective transition from juvenile to mature phase in shoot development (Poethig, RS, Nature 336:82-83, 1988).

Reported here is an unusual behaviour of an unstable Cg2 macromutation phenotypically similar, but not allelic, to Cg1 (Fig. 1a and b), the former being located on the short arm of chromosome 3 (Lysikov, VN et al., Sov. Genet. 20:90-99, 1984; Krivov, NV and Lysikov, VN, Bulletin AN SSRM 2:20-24, 1988).

Genetic analysis of the Cg2 macromutation has shown it to be a monogenic dominant mutation which is highly sensitive to maize genetic background and gene dose. Sensitivity to genetic background is reflected in that in crosses of Cg2 carriers with different maize lines, plants indistinguishable from normal maize plants as well as those with weak (Cg2-w), moderate (Cg2-m), and strong (Cg2-s) expression of the mutant phene (Fig. 1c) occur.

A study of lineages established in 1978 has revealed that the Cg2 macromutation exhibiting 100% penetrance on the genetic background of specially selected lines mutates at a high rate in somatic and generative cells. In somatic cells, the level of mutability was assessed from the frequency of the Cg2-mosaic plants whose lateral shoots usually had mutant phenotypes, with the leading one being non-mutant (Fig. 1d). The Cg2-mosaic plants were considered normal if the majority of shoots in their bushes had normal phenotypes (shoots ending in a tassel), and mutant if most of the shoots had the corngrass phenotype. In generative cells, mutability was assessed from the occurrence of normal maize plants in homozygote progeny (Cg2/Cg2), with the rate of Cg2 to Cg2+ transition amounting to 55%. Among Cg2+/Cg2+ homozygotes, mutation of the Cg2+ to Cg2 type occurred at the rates of 2-2.5% and 5.5% in generative and somatic cells respectively.

However, high mutability is not confined to the Cg2 locus since mutant forms phenotypically resembling ys1, virescent seedling, striate, pericarp and cob colour, ramosa, and brachytic, have been observed among the Cg2 progeny. Later, these mutations have been shown to be unlinked to the Cg2 locus. To determine the chromosomal locations of newly arising mutations, carriers of these mutations were crossed with phenotypically similar mutants from the All-Union Institute of Plant Growing (St. Petersburg) collection. Tests of allelism have shown the new mutation br to be non-allelic with br1, br2 and br3. The new variant sr is not allelic with sr1 and sr2, and the new mutation ra is non-allelic with ra1. Thus, recessive mutations occurring among the Cg2 progeny reflect the properties of mutator activity of the unstable Cg2 locus or those of the genetic background of the lines carrying Cg2.

Tests of allelism have also revealed that mutations ys*-253 and ys*-143 are not allelic to each other, and crossing lines of non-corngrass phenotype carrying these mutations results in up to 51.5% of the F1 plants having the Cg2 phenotype (Table 1). Tests for the ability to induce instability at the Cg2 locus on hybridization of Cg2+ homozygous lines carrying recessive mutations sr*-220, br*-220, ys*-253 P-rr, ys*-253 and ys*-143 have shown that Cg2 and Cg2-mosaic plants occur at a rate of 44.1 to 51.5% in combinations ys*-253 x ys*-143, ys*-143 x ys*-253 P-rr, and ys*-253 P-rr x br*-220. In the latter case, the appearance of Cg2 plants entirely depended on the direction of the cross. This was indicated by the fact that not a single plant was found in the progeny of the cross with the br*-220 line used as a female parent.

Figure 1. Maize plants carrying mutation Cg1, (a) usual mutant phenotype, and (b) m(Cg)-mosaic phenotype. Maize plants carrying Cg2 3S mutation, (c) usual mutant phenotype, and (d) m(Cg2)-mosaic phenotype.

Table 1. Mutation rates at the Cg2 locus in crosses between stable Cg2+/Cg2+ derivatives.
Combination No. earcorns Cg2+ Cg2-w Cg2-m Cg2-s m(Cg2) Cg2 carriers (%) Total
Cg2+ 220 x ys*-143 3 141           141
ys*-143 x Cg2+ 220 2 10           10
Cg2+ 220 x ys*-253 2 49           49
ys*-253 x Cg2+ 220 1 8           8
br*-220 x ys*-253 3 110       1 0.9 111
ys*-253 x br*-220 1 96           96
br*-220 x ys*-253 P-rr 4 101           101
ys*-253 P-rr x br*-220 1 18   14 3   48.6 35
br*-220 x sr*-220 2 45           45
sr*-220 x br*-220 3 95           95
ys*-143 x sr*-220 2 6           6
sr*-220 x ys*-143 3 164           164
sr*-220 x ys*-253 2 164           164
ys*-253 x sr*-220 1 8           8
sr*-220 x Cg2+ 220 1 146           146
ys*-143 x ys*-253 P-rr 2 47 7 37   6 51.5 97
ys*-253 x ys*-143 2 76 3 43   14 44.1 136
ys*-143 x br*-220 2 90           90
br*-220 x ys*-143 3 77           77

The same situation was observed in crossing the marker line gs bm2 x sr*-220 upon which the proportion of mutant plants was 51.7%. However, no mutant plants have been identified in crosses using sr*-220 as a female parent. Mutant Cg2 maize plants also occurred in bm bv bt x ys*-253 and j v16 ms8 x ys*-253 crosses. A total of 5 marker lines were subjected to test. Three of these appeared to be capable of inducing, in coaction with sr*-220, br*-220 and ys*-253, the formation of Cg2 plants in F1. Genotypes of the lines (F2, F7, MK01, VIR-44, Co125) occurring among the recognized hybrids failed to induce mutations of the Cg2+ to Cg2 type. It is only in the O92 x ys*-253 P-rr combination that a single plant among 135 plants tested has been identified as a Cg2 mutant.

The spontaneous appearance of mutant Cg2 plants as well as Cg2-mosaic plants in crosses of phenotypically normal plants seems, like previously observed mutability (Cg2 to Cg2+), to be due to a genetic factor exhibiting unique regulatory functions. According to Poethig, RS (Nature 336:82-83, 1988), the Tp1 mutation can have similar functions. Genes interacting with Tp1 and Tp2 can also be classified into this group of mutants (Poethig, RS, MNL 62:99). Cases like these are also known to occur among other higher plants. For example, a tissue-specific regulatory protein DEF A affecting flower organogenesis has been isolated from a homeotic mutant deficiens (defA+) of lion's mouth. It has been suggested that regulation here is performed due to the response of this protein to general cellular signals (Sommer, H et al., EMBO J. 9:605-613, 1990). However, equally likely is the suggestion that the Cg2 macromutation instability is caused by mobile elements (ME) similar to Ac, En, Uq, Mu and other elements (McClintock, B, Cold Spring Harbor Symp. Quant. Biol. 16:13-47, 1951; Peterson, PA, Mol. Gen. Genet. 183:440-448, 1981; Friedemann, P and Peterson, PA, Mol. Gen. Genet. 187:19-29, 1982; Robertson, DS, Mutat. Res. 51:21-28, 1978; Saedler, H and Nevers, P, EMBO J. 4:585-590, 1985). MEs have been found in nearly all of the mutant genes studied. Thus, "pure" mutant genes seem to be non-existent (Peterson, PA, Proc. Int. Symp. Plant Transposable Elem., 43-68, 1988). MEs are present both in the nucleus and in cytoplasm. The Robertson, DS Mu-strains, for example, carry extrachromosomal Mu1 and Mu1.7 elements whose origin is associated with Mu activity. These have been suggested to be generated during Mu transpositions as intermediates resulting from these transpositions or as products of Mu excision (Sundaresan, V and Freeling, M, Proc. Natl. Acad. Sci. USA 84:4924-4928, 1987). Three groups of evidence point to the presence of MEs in the Cg2 progeny: 1) high rate of mutation at the Cg2 locus in generative and somatic cells; 2) induction of instability at the Cg2 locus in crossing specially selected lines, and specific role of the cytoplasm genotypes of crossed lines; 3) high mutator activity of the Cg2 unstable macromutation evidenced by mutants recovered from the Cg2 progeny which, despite their phenotypic similarity, are non-allelic to known mutations from the collection and to one another.

As the Cg corngrass phenotype is due to a defective juvenile-to-mature transition in shoot development, the Cg2 being phenotypically similar to the Cg1 heterochronous mutation, we have named the ME responsible for the Cg2 locus instability Fpj (factor prolonged juvenile). The presence of an Fpj mobile element is evidenced by consistent irregularities in the time of transition from juvenile to mature (Cg2 to Cg2+) and in restitution time (Cg2+ to Cg2). The disruption of reciprocality in the induction of such transitions seems to be due to the presence, in some lines, of extrachromosomal Fpj elements similar to extrachromosomal Mu elements.

If, as noted above, a delay in the juvenile-to-mature transition is accompanied by a mutator effect, then introduction of an Fpj element into marker lines can induce mutations in marker genes as well. This hypothesis has been tested experimentally using wx sh x Cg2+ and Cg2+ x wx sh crosses. A total of 64,168 Wx/wx kernels have been examined; 4,099 out of these (i.e. 6.4%) proved to be mosaic for the wx locus. In some families the proportion of mosaic kernels was as high as 20-30% and higher. The high rate of mutation at the wx locus is an indirect evidence for the existence of an Fpj mobile element which appears to serve the function of switching the genetic program in Zea mays.

Finally, lines sr*-220, br*-220, ys*-143, ys*-253 and MK01 all of which, except for MK01, carry Fpj have been tested for the presence of Dt, Uq and Mrh regulatory elements using lines having receptors at a-m Dt, a-ruq, a-m(r)h. It has been found that all of these lines contain Dt, Uq and Mrh (half or more of the kernels from a hybrid ear are mosaic), while the tester lines themselves carry no Fpj elements (not a single Cg2 plant has been found). An attempt to induce the Cg2+ to Cg2 mutation in reciprocal cross of MK01 carrying Dt, Uq and Mrh regulatory elements with sr*-220, br*-220, ys*-143 and ys*-253 lines has been unsuccessful. The lack of interaction between Dt, Uq, Mrh regulatory elements and the Fpj system suggests that Fpj is not genetically identical to these families of mobile elements.

Acknowledgements: The author is very grateful to Prof. A. B. Korol, AB for his useful criticisms in discussing the manuscript. Many thanks are also due to Prof. P. A. Peterson, PA for kindly providing seeds of the tester lines used in this study and to Mr. G. K. Lakhman, GK for translating the manuscript into English.

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

Return to the MNL 66 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page