University of Minnesota

Spontaneous activation of transposable elements following an interracial cross in maize
--Vladutu, CI, Phillips, RL

In order to initiate a QTL analysis for several traits, we developed a Gaspe Flint (GF) x N28 F2 population. The F2 plants, all wild type, were selfed and the resulting F3 families were planted in 1996 in two replications in the field. Among the 99 F3 families, 6 segregated for obvious plant mutations: branched-silkless-like, chronic wilting, yellow-green, pigmy-like, male-sterility, white-sheath-like and stunted plants. Another F3 family gave rise to an andromonoecious dwarf. Two other putative mutations (earless and disease mimic) need further confirmation. The branched-silkless-like and the chronic wilting mutants segregated within the same F3 family. Two of the mutations showed somatic instability; some of the white-sheath-like mutants had dark green stripes in the internodes, leaf sheath and blade and one of the plants of the F3 family segregating for male-sterility had a chimeric tassel with fertile and sterile branches. In another F3 family, one plant had a chimeric tassel for anther pigmentation. Within each of the families segregating for plant mutations, several plants were selfed and occasionally outcrossed. Within the families that did not segregate for any obvious plant mutation, only one or two plants were selfed. Upon selfing, in four other families, F3 plants were found to segregate for pale yellow (y), floury (fl), brittle (bt ) and shrunken (sh)-like kernels. Some of the fl-like kernels showed patches of wild type endosperm. In the family with the andromonoecious dwarf, one F3 plant was found to segregate for viviparous (vp) kernels. F3 plants within several other families segregated for different types of defective kernels. Eighteen GF x N28 F3:4 families were grown in the field in 1997. Five additional mutations arose among the 18 families. Four mutations (albino, yellow-stripe, narrow-leaf and knotted) appeared within four different F3:4 families of previously identified mutants; a zebra-crossband mutation occurred in one of the six F3:4 families derived from F3 families that in 1996 did not segregate for visible mutations. In 1997, in addition to white-sheath-like, male-sterile and fl-like, two other mutations, (yellow-green and narrow-leaf) showed somatic instability.

Considering that all F2 plants and F3 kernels were wild type, and that somatic instability occurred with some of the mutations, the high rate of forward mutations (resembling the behavior of Robertson's Mu stock) is likely the result of the activation of transposable elements (TEs) in the F1 plant. Since the retroelements are not expected to excise and thus induce reversion to the wild phenotype, and no excision events have been so far reported for MITEs, the unstable mutations, at least, are probably caused by DNA-elements. It seems that the putative TE activity that occurred in the F1 plant was maintained in some of the F2 plants. The fact that no two F3 families segregated for the same mutation suggests that the TE insertional activity occurred late in the development of the GF x N28 F1 plant. Accurate estimates of mutation rates can not be computed due to the small size of the F3 population. However, among 300 N28 x N28E (N28E is an early backcross derivative of N28, having two chromosomal segments retained from GF) F3 families grown in 1993 in the same location, no mutation had been identified. If the N28 x N28E F3 population would have had one F3 family segregating for a mutation, the proportion (~7%) of GF x N28 F3 families segregating for a different plant mutation would be at least 20 times higher.

In the summer of 1997, several of the mutants were outcrossed with different inbred lines and complementation tests were performed for some of the kernel mutations and the branched-silkless-like mutation (the testers were kindly provided by the Maize Stock Center). The complementation tests revealed that the sh-like, bt-like and vp mutations had occurred in the Sh1, Bt2 and Vp1 genes respectively, all of which have been cloned (Werr et al., EMBO J. 4:1373-1380, 1985; Bae et al., Maydica 35:317-322, 1990; and McCarty et al., Plant Cell 1:523-532, 1989, respectively). The result of the complementation test involving branched-silkless-like mutation will be apparent next summer (1998).

In the summer of 1997, somatic reversion assays were initiated, using testers (generously provided by the Maize Stock Center and P.A. Peterson) for Ac, Spm, Mu, Mrh and Dt activity, having the corresponding defective elements in anthocyanin genes (A1, A2, Bz1, Bz2 and C1) expressed in the aleurone layer. Mutability assayed by somatic reversion and mutagenicity do not always correlate. However, since the events leading to mutagenicity appeared to occur in the F1, we are testing for the existence of autonomous element activity in the GF x N28 F1 compared to the parents (GF and N28). In order to avoid potential confusion between the variegation caused by TE excisions with the mottling effect caused by the imprinting of R alleles transmitted through the male, the transposon testers were used as the female parent. Since both GF and N28 are homozygous wild-type for A1, Bz1 and Bz2, the testcrosses for Dt, Mrh and Mu activity have to be carried on for one more generation (i.e. the hybrids between the testers and the tested material have to be backcrossed, as males, onto the testers as females). Negative results (no reversions) for the three genotypes have been obtained with the Ac testers. Spm activity in half of the kernels was detected in a testcross with GF but not with GF x N28 F1 and N28. This result indicates segregation for an active Spm element within GF. GF is not an inbred line but a cultivar obtained through mass selection. Thus, it is expected to be heterogeneous/heterozygous, and signs of genetic instability (segregation for su1, y1 and lg) within GF had been noticed in the past (Vladutu MSC 1996). However, since no Spm activity occurred in testcrosses with GF x N28 F1, it is unlikely that Spm activity had induced the burst of mutations in the GF x N28 progenies.

The RFLP and the phenotypic data provide compelling evidence against the possibility of potential contamination with transposon stocks as the source of genetic instability in the GF x N28 progenies. Also, limited cytological analysis of the microsporocytes from the GF x N28 F1 plants has not detected structural heterozygosity (such as inversions or translocations) that could have triggered chromosome breakage and subsequently activated TEs. Thus, the "genomic shock" that activated TEs in GF x N28 progenies was probably due to heterozygosity per se at yet unknown host loci.

GF is a Northern Flint open-pollinated population. N28 is a Corn Belt Dent line. Since the parental cultivars belong to historically and economically important maize races and have been used in recent breeding programs, this case has relevance in terms of the induction of de novo genetic variability in maize improvement/evolution. The high rate of forward mutations that have occurred in progenies of the above cross shows that the high level of mutagenic activity is not an exclusive attribute of Robertson's Mu stock, and suggests that heterozygosity per se at yet unidentified host loci can induce the activation of silent TEs. Since complementation tests have shown that putative TE insertions have occurred in Sh1, Bt2 andVp1 genes, all of which have been cloned, the identification and characterization of the activated TEs may be straightforward.

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

Return to the MNL 72 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page