Most, if not all, DNA glycosylases show some degree of product inhibition, because their dissociation from the product, AP-site, is rate-limiting for enzymatic turnover [19
]. This property is most pronounced in the mismatch-specific uracil glycosylase (MUG) family proteins. The eukaryotic thymine-DNA-glycosylase bind AP-sites with exceptional strength and are thus fully product inhibited [23
], whereas the bacterial MUG protein turns over, albeit at a very low rate [25
]. The bifunctional, DNA glycosylases, such as NTH1 and OGG1, also remain tightly bound to their AP lyase product [26
For MPG, it is also thought that after the base is removed, the AP-site is the limiting factor for product dissociation. Product inhibition by excised free base, the second reaction product, to reduce enzymatic activity of MPG or any DNA glycosylase has been tried without much success in the past [8
]. In the current study as the reaction proceeds towards the pseudo-single-turnover condition, the difference in product formation is diminished (). Interestingly, the KD
s for at least two different substrates (Hx- and εA-DNA) differ very little (). Also there is no difference in the reaction rate under single-turnover condition for Hx-and εA-DNA, but the difference is apparent under multiple-turnover condition ( and ).
Being a broad substrate enzyme, MPG must be flexible for DNA binding in order to recognize DNA lesions of varied structures. In fact, for similar reasons T. Ellenberger and his colleagues proposed that a “nonspecific catalytic mechanism” must be met for an enzyme to succeed as a generalist as one of the major criteria, which comes “at the expense of catalytic power” [29
]. That minimal difference in binding does not have any effect on overall catalysis. In our previous study, we showed that although the KD
was apparently ~160-fold less for NΔ100 mouse MPG (mMPG) compared to full-length protein, the overall product formation was only ~3-fold different. One possible explanation was that the binding of full-length mMPG to Hx was extremely fast, as evident from the KD
value of 0.15 nM compared to its chemistry and turnover step [14
]. The same logic possibly holds true for this study. That study and a paper by Saparbaev et al also indicated the importance of N-terminal tail in the turnover [14
]. Moreover, in a recent study Dr. L.D. Samson and colleagues showed the requirement of N-terminal for a variety of new substrates of MPG [31
]. As N-terminal tail seems to be important for MPG’s reaction, it is highly recommended to do the detailed characterization of full-length, rather than the truncated MPG. To date, most studies reported using the truncated form of MPG. To our knowledge, this is the first detailed pre-steady state kinetic characterization of stable human full-length MPG.
After the base is excised from DNA, MPG generates two products: AP-site and the free damaged base. AP-site is a common product of MPG’s excision reaction towards εA- and Hx-DNA. The product dissociation rate, however, is different, ~2-fold, for these two lesions (). Although this difference amounts to less than an order of magnitude, it is possible that such differences could be amplified in vivo, given that MPG activity could be affected by various repair modulatory proteins in the cell. This difference can only be explained if the modified base dissociates from the enzyme at a slower rate compared to the AP-site. In fact, we observed much faster dissociation of AP-site at 37°C in SPR experiments than the overall product dissociation of MPG under MTO condition. Furthermore, there is a notable difference in the dissociation rates (kpd) for the modified bases; εA dissociates faster than Hx (). Thus, it is tempting to speculate that dissociation rates of MPG from its reaction products would be in the order of AP-DNA>εA>Hx. This, to our knowledge, is the first evidence to show the importance of excised base for any DNA glycosylase in its enzymatic turnover.
Note that, previously, in an elegant study on the role of product inhibition in glycosylase function, S. David and her colleagues demonstrated that the product dissociation rate of MutY (Adenine DNA-glycosylase) from the cleaved DNA product was different for G:A and OG:A mismatch substrates. The authors noted that "the slower dissociation rate of MutY from an OG:(AP site) compared to the corresponding G:(AP-site) is suggestive of specific contacts between MutY and OG that persist after the adenine has been removed.” This result showed that the orphan base on DNA was responsible for product inhibition [32
]. Our study indicates the role of the removed base in product inhibition.
The contribution of excised base in the overall product dissociation can also be explained from the existing crystal structures of MPG. From the tertiary structures of MPG in complexes with εA-containing DNA, it was suggested that the enzyme selects the substrate bases through a combination of hydrogen bonds and the steric constraints of the active site [29
]. Several other principles, such as shape and electrostatic properties of the binding pocket which accepts the flipped-out nucleotide, also dictate MPG in selecting its substrates. As revealed from crystal structures, the backbone amide of His-136 donates a hydrogen bond to the etheno N-6 nitrogen of εA [29
]. This hydrogen bond can discriminate against adenine, which has a 6-amino group that cannot accept a hydrogen bond [29
]. Thus, this hydrogen bond may also be important in discrimination and excision of Hx and εA, and perhaps in the dissociation of the excised bases and overall turnover. Particularly, the dissociation of excised Hx would be slower compared to that of εA, if we assume that the free bases are transiently remaining in the catalytic pocket through the same hydrogen bonding and consider that the oxygen atom at the 6-oxo group of Hx in the Hx-MPG complex forms a stronger hydrogen bond than the nitrogen atom in εA-MPG complex [29
]. Similar studies will be required to extend to other DNA glycosylases to examine the participation of excised damaged bases in enzyme turnover and also to test whether this phenomenon is general for reaction mechanisms of DNA glycosylases.