Ty3 reverse transcriptase complexed with an RNA-DNA hybrid shows structural and functional asymmetry

Retrotransposons are a class of mobile genetic elements that replicate by converting their single-stranded RNA intermediate to double-stranded DNA through the combined DNA polymerase and ribonuclease H (RNase H) activities of the element-encoded reverse transcriptase (RT). However, while a wealth of structural information is available for lentiviral and gammaretroviral RTs, equivalent studies on counterpart enzymes of long terminal repeat (LTR)-containing retrotransposons, from which they are evolutionarily derived, is lacking. In this study, we report the first crystal structure of a complex of RT from the Saccharomyces cerevisiae LTR-retrotransposon Ty3 in the presence of its polypurine tract-containing RNA-DNA hybrid. In contrast to its retroviral counterparts, Ty3 RT adopts an asymmetric homodimeric architecture, whose assembly is substrate-dependent. More strikingly, our structure and biochemical data suggest that the RNase H and DNA polymerase activities are contributed by individual subunits of the homodimer.

Numbers on the top of the alignment correspond to Ty3 RT residues.
1RTD 1 ). Identical residues are marked with '*' and similar ones with ':'. Secondary structure elements are shown as tubes (helices) and arrows (strands) and labeled and colored as in Figure 1.

SUPPLEMENTARY NOTE Predicted substrate binding by Ty3 RNase H
In bacterial and human RNase H many substrate interactions are mediated by the backbone of the protein and binding of the DNA at the phosphate binding pocket is enhanced by an interaction with the positive dipole an -helix 2,3 . Therefore, for RNases H1 the overall fold of the domain is important for the substrate recognition. Also the position of the active site of Ty3 RNase H relative to the rest of the structure is nearly the same as in 'cellular' RNases H1. We therefore expect that the functional elements would be located in similar positions of the domain.
Critical active site residues of the Ty3 RNase H domain are Asp358, Glu401, Asp426, and Asp469, which are superimposable with their counterparts of cellular enzymes ( Supplementary Fig   5a), although their match is not as good as between human and bacterial RNase H1, indicating that the geometry of metal ion and scissile phosphate binding is also altered. D358N, E401Q and D426N substitutions eliminated RNase H activity and an D469N mutation led to its reduction, while all mutations prevented transposition 4 . One feature of Ty3 RT noted previously 4,5 is the lack of a loop located proximal to the active site harboring a histidine residue (His264 in human RNase H1 and His539 in HIV-1 RNase H), whose conservation has led to the postulation that it assists product dissociation following hydrolysis 3 . The corresponding loop is shorter in Ty3 RT and no histidine equivalent was identified. These differences notwithstanding, conservation of the active site and the constraints of the geometry of metal ion and scissile phosphate binding imply that interactions with the RNA strand at and around the RNase H active site of Ty3 RT are similar to those described for cellular enzymes 2,3 . Based on this assumption, Arg473 and Tyr459, located close to the active site, are positioned to interact with the backbone of the RNA strand. The first is well conserved among Gypsy retroelements (Supplementary Fig 2) and mutating the second to Ala greatly reduced RNase H activity 4 . Another residue studied biochemically was His427, whose substitutions reduced RNase H activity 4 . In our structure, this residue is located close to the active site and plays a structural role. in cellular enzymes is the phosphate-binding pocket, formed by two to four residues and tightly binding a phosphate group of the DNA, leading to its deformation (Fig 2c, d, e). These deformations are only possible for DNA, which serves for its recognition. The most conserved residue of the phosphate pocket is a Thr or Ser located at the N-terminus of the first helix of the RNase H fold 3 .
This residue is critical to activity of E. coli RNase H 6 and HIV-1 RT 7 . No phosphate-binding pocket or equivalents of this Thr were identified in Ty3 RNase H. Taken together, these observations suggest that, although the active site of Ty3 RNase H likely functions through a very similar mechanism to cellular enzymes, the mode of RNA-DNA binding may involve fewer contacts with nucleic acid, and in particular with the DNA strand.
Ty3 RNase H domain is structurally similar to the HIV-1 connection subdomain (Fig 2a, b).
However, no functional residues can be identified in the latter, including amino acids forming the active site and substrate binding residues, in agreement with the fact that the connection subdomain lost its catalytic function.