DO THE INTEGRASES OF LTR-RETROTRANSPOSONS AND CLASS-II ELEMENT TRANSPOSASES HAVE A COMMON ANCESTOR

The integrases of retrotransposons (class I) and retroviruses and the transposases of bacterial type elements (class II) were compared. The DDE signature that is crucial for the integration of these elements is present in most of them, except for the non-LTR retrotransposons and members of the hAT and P super-families. Alignment of this region was used to infer the relationships between class II elements, retrotransp osons, and retroviruses. The mariner-Tcl and the Pogo-Fot1 super-famil ies were found to be closely related and probably monophyletic, as wer e LTR retrotransposons and retroviruses. The IS elements of bacteria w ere clustered in several families, some of them being closely related to the transposase of the mariner-Tcl super-family or to the LTR retro transposon and retrovirus integrases. These results plus that of Xiong and Eickbush (1990) were used to develop an evolutionary history sugg esting a common ancestral origin(s) for the integrases and transposase s containing the DDE signature. The position of the telomeric elements (Het-A and TART) was assessed by comparing their gag and reverse tran scriptase domains (when present) to those of group II introns and non- LTR retrotransposons. This preliminary analysis suggests that telomeri c elements may be derived from non-LTR retrotransposons.