Asymmetric ends are required for excision of En/Spm

--Monika Frey, Julio Reinecke, Heinz Saedler and Alfons Gierl

The correlation of genetic, molecular and biochemical data identified two kinds of cis-acting elements: the terminal inverted repeats (TIRs) and the subterminal TNPA binding motifs. It has been proposed that binding of TNPA to its motifs induces complex formation between both ends by a "zipper"-like mechanism leading in consequence to alignment of the TIRs and subsequently, after binding of TNPD to the TIRs, to the excision of the element (Frey et al., EMBO J. 9:4037-4044). One prediction of this model is that the pattern of the TNPA binding motifs at the subtermini plays a critical role in excision because altered distributions might interfere with "zipper" formation. On the other hand, alterations of the inner part of the element should only gradually influence the excision rates.

A mutational analysis of the cis-determinants was initiated in order to test the predictions mentioned above. So far two dSpm derivatives have been constructed and analyzed in the transgenic tobacco test system. The first is 716bp long, has intact cis-determinants and consists of 270bp of the 5' end and 446bp of the 3' end of En-1. This element is excised in tobacco in the presence of TNPA and TNPD as revealed by restoration of the reporter gene (Gus gene), Southern analysis and PCR amplification of the excision site. Excision generates the expected footprints, however, the excision frequency is reduced to about 20% compared to a native 2.2kb dSpm element (Spm-I8).

The second dSpm construct has two identical ends, i.e., the 446bp of the 3' end of the 2.2kb dSpm element are substituted by the 270bp of the 5' end. Neither histological staining for Gus-enzyme activity, nor Southern analysis revealed excision of this element, only with PCR excision events could be amplified. Sequence analysis of two of these PCR products displayed a striking feature. In contrast to all excision events analyzed in the tobacco system before, employing native maize dSpm elements or En-1, only "symmetrical" footprints are generated: in the one case both target site duplications (TSDs) were retained and in the other case one base pair was deleted from each TSD. Therefore, the excision frequency for the 5' end/5' end element is not only extremely low, but also the mechanism might be altered, however, excision is still dependent on TNPA and TNPD.

These results seem to confirm the prediction of the "zipper"-model that asymmetric ends are required for excision. In this respect En/Spm resembles Ac and the P element of Drosophila, which are also defective in excision when they have identical ends.

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