Sumoylation of thymine DNA glycosylase occurs efficiently and weakens DNA binding but does not regulate enzymatic turnover

Abstract

Thymine DNA glycosylase (TDG) performs essential functions in maintaining genetic integrity and epigenetic regulation. Initiating base excision repair (BER), TDG removes thymine from mutagenic G-T mispairs caused by 5-methylcytosine (5mC) deamination, and other lesions including uracil. In DNA demethylation, TDG excises 5-formylcytosine (5fC) and 5-carboxylcytosine (5caC), generated from mC by Tet (ten-eleven translocation) enzymes. TDG is also modified at a single lysine residue (K330) by small ubiquitin-like modifier (SUMO) proteins, which modify target proteins to perform and regulate a diverse array of functions. Sumoylation does not regulate enzymatic turnover for any other target proteins, but it has been proposed to perform this function for TDG, allowing efficient handoff during BER. Our initial studies examined the specificity of the modification reaction itself, determining that TDG sumoylation occurs efficiently but demonstrates no selectivity for TDG bound in an enzyme-product complex. This specificity would be central to the idea that sumoylation regulates TDG turnover, and yet our in vitro studies do not indicate that TDG bound to abasic DNA is preferentially modified. Moreover, our studies demonstrated that TDG~SUMO-1 still possesses high affinity for abasic DNA, indicating that modified TDG can bind efficiently to product. Further studies indicated that, while sumoylation weakens DNA binding, both TDG~SUMO-1 and TDG~SUMO-2 can bind its known substrates, including DNA duplexes containing thymine (T), uracil (U), 5fC, or 5caC paired with guanine (G). In contrast to prior findings with TDG~SUMO, our studies also demonstrate that modified TDG excises its target base from each of these substrates. Excision of T, however, is weak and appears to be inhibited by the presence of magnesium, while excision of other bases is unaffected by the divalent cation. That TDG~SUMO is capable of not only binding to both substrate and product but also possesses activity on its canonical substrates strongly suggests that the function of TDG sumoylation in vivo is not to regulate efficient release of TDG from product. Additional studies need to be performed to uncover the function of TDG sumoylation.