DAVIS, CALIFORNIA

University of California

STANFORD, CALIFORNIA

Stanford University

Products of Mu insertion and excision at the bronze-1 gene

--Anne Bagg Britt and Virginia Walbot

Last year we reported the position of the bz-mu1 and bz-mu2 insertion sites, as well as the sequence of several somatic excision products of Mu1 from bz-mu1. Here we report the sequence of two germinal revertants of Mu1 from bz-mu1, and revise our report of the bz-mu2 insertion site.

We recovered two independent germinal revertants of bz-mu1. Both were isolated from bz-mu1 homozygous plants which produced ear sectors segregating for revertant (purple) kernels. Bz1-R1 was isolated from a self-pollinated ear, while the ear which produced Bz1-R2 was crossed by bz tester. DNA isolated from each of these germinal revertants was employed as a template for PCR amplification of the bz-mu1 target site. The resulting products were cloned and sequenced. Because the sequence of the cloned product for Bz1-R1 was identical to that of a previously cloned somatic excision product, the amplification, cloning, and sequencing procedures were performed a second time, and yielded identical results.

The sequences of these two germinal revertants are as follows:

bz-mu1...GCCGCAAACAGGG(MU1)CAAACAGGGTGGACG...

Bz1-R1 ...GCCGCAAACCCAAACAGGGTGGACG...

Bz2-R2 ...GCCGACG...

Like the two germinal revertants of bz-rcy sequenced by Schnable, PetersonPA, and Saedler (MGG 217:459, 1989), these revertant alleles are the result of imprecise excision of Mu1. Bz1-R1 carries a deletion of the four bases 5' to the Mu target site, as well as a single base insertion 3' of the target site. Bz1-R2 includes a deletion of the entire 9 bp repeat, as well as an additional 3 bp deletion 3' of the original Mu insertion. Unlike the two revertant alleles sequenced by Schnable et al., both of these excision products restore the original Bz-W22 reading frame.

In order to determine the points of insertion of the Mu2 element in bz-mu2, PCR was used to amplify the Mu ends and their contiguous Bz sequences at both the 5' and 3' ends of each insertion. We found that the amplification protocol recommended by Perkin-Elmer Cetus produced a very nonspecific product in this region of the Bz1 gene. We therefore modified our regular three step cycling procedure (1 min. at 94�, 1 min. at 55�, 1 min. at 74� 30 reps) to a two step procedure recommended to us by Tom Sullivan (1 min. at 94�, 1 min. at 74� 30 reps.). This modified reaction produced a very specific product. The amplified fragments were then cloned and sequenced. The sequence of the 9 bp repeat of bz-mu2 is GCCCAACTG.

This sequence is located just 3' of the XhoII site in Bz1. Previously published Southern blot analysis had indicated that the Mu2 element had inserted 5' of the XhoII site. Last year we published an estimation of the position of the insertion site based on Southern blot data which indicated that the element had inserted, paradoxically, 5' of the XhoII (AGATCC) site, but 3' of the Sau3a (GATC) site located within that XhoII site. The sequence reported above, however, indicates that the insertion is located 3' of both the XhoII and the Sau3a sites.

Why then did the digests reported by Taylor and Walbot (Genetics 117:297, 1987) indicate that the element was positioned 3' to the XhoII site? XhoII is sensitive to methylation, and will not digest sites at which C is methylated in the 5' position. The XhoII site in Bz1 (GAGATCCG) includes two C's which are susceptible to methylation. If this site was methylated in the DNA preparation analyzed by Taylor and Walbot, their results (but not their conclusions) would be consistent with the sequence presented here.

Return to the MNL 65 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage