br Materials and methods br Results br Discussion The redox
Materials and methods
Discussion The redox domain that spans the region between amino hydroxychloroquine sulfate positions 35 and 127 of the major human AP endonuclease, APE1, regulates the sequence-specific DNA binding of various transcription factors [, ]. The existent crystal structures of APE1 in complex with DNA containing an AP site do not show any additional interactions of the N-terminal amino acid residues with a DNA duplex . Indeed, the truncated APE1-NΔ61 protein has a proficient AP site cleavage activity similar to that of the full-length APE1 protein, indicating that N-terminal residues are not required for BER functions . Nevertheless, the first 42 N-terminal disordered amino acid residues of APE1, not visible in crystal structures, are highly enriched in positively charged basic lysine and arginine residues, which can be engaged in electrostatic interactions with DNA phosphates. Several lines of evidence show that N-terminal amino acid residues of APE1 participate in DNA binding during NIR activity  and in interaction of the protein with RNA and negative calcium responsive sequence elements (nCaRE) of certain gene promoters [, , ]. The deletion, mutation, and/or acetylation of lysine residues among the first 33 amino acid residues of APE1 lead to increased AP site cleavage activity [, ]. Taken together, these observations point to the existence of two N- and C-terminal DNA-binding sites in the APE1 protein and that the additional interactions of the N-terminal amino acid residues of APE1 with DNA may contribute to both transcription regulation and DNA repair. The function of the first 61 N-terminal residues of the redox domain of APE1 in the DNA glycosylase-independent NIR pathway and specific binding to DNA and RNA suggest that the redox domain was acquired during vertebrate evolution to perform additional biological functions absent in bacterial homologs such as E. coli exonuclease III. In the present work, we are interested in the participation of the redox domain of APE1 in the stimulation of DNA glycosylases. It has been shown that APE1 stimulates multiple turnover of human DNA glycosylases by actively displacing them from the AP site, the end product of the reaction [, , ]. Moreover, the APE1-catalyzed stimulation does not require the AP site cleavage activity suggesting that APE1 can displace DNA glycosylases without cutting AP site in DNA [, ]. In addition, several studies suggest that APE1 does not interact with DNA glycosylases via specific protein–protein interactions [, ]. For example, the ability of APE1-WT to stimulate truncated DNA glycosylase ANPGcat [present study and  and even the catalytic domain of MBD4 and TDG [present study and  indicates the absence of specific protein–protein contacts. This observation implies that the effect of APE1-WT is mediated by the conformational changes of DNA induced by APE1-WT. Our results here showed that the truncated APE1-NΔ61 protein deficient in redox and NIR functions cannot stimulate the repair activities of human DNA glycosylases OGG1, MBD4cat, and ANPGcat. All these findings reveal that the APE1-catalyzed redox function, NIR activity, and DNA glycosylase stimulation share a common underlying molecular mechanism. Here, we hypothesized that interactions of the N-terminal part of the redox domain of APE1 with DNA and conformational changes of DNA induced by APE1 binding may be involved in the disruption of the complex of a DNA glycosylase with an AP site. To elucidate the molecular mechanism of the APE1-catalyzed stimulation, we employed the stopped-flow fluorescence analyses of the interaction of APE1 with OGG1–FRET-8oxoG·C and MBD4cat–FRET-U·G complexes under the experimental conditions used (Fig. 2, Fig. 4). It is known that the DNA damage recognition by a DNA glycosylase involves several key steps: (i) nonspecific DNA binding; (ii) DNA bending at the lesion site, (iii) eversion of the damaged base from the double helix into the enzyme’s active site, and (iv) insertion of some amino acids of the enzyme into the resulting void in DNA. It is clear that certain steps may influence stability of the DNA duplex, which in turn would affect the distance between the fluorophores on DNA and produce a change in the FRET signal. Most likely, stabilization of the ends of the DNA duplex, which affects the distance between the fluorophores, occurs already at the first step of a nonspecific enzyme–substrate complex formation. Based on this assumption, the FRET data revealed no difference in the observed rate constant (k1) of the formation of nonspecific DNA glycosylase–DNA substrate complex in the presence or absence of APE1. The dramatic change in FRET signal amplitude in the presence of APE-WT after initial DNA glycosylase substrate binding suggests that APE1 influences subsequent steps of catalytic enzyme–substrate complex formation such as DNA bending, damaged base flipping out, and insertion of amino acid residues (Fig. 2, Fig. 4). The action of APE1 stabilizes these intermediate transition states of the DNA damage-specific binding by DNA glycosylases. This APE1-mediated stabilization of the DNA glycosylase-specific interactions with DNA damage results in more effective accumulation of the catalytically proficient enzyme–substrate complexes. The same DNA distortions caused by APE1 may lead to expulsion of the DNA glycosylases from abasic site DNA owing to the loss of “anchor” interactions between an everted damaged base and the enzyme active site, which in turn may increase turnover rates of the enzymes. Overall, these data support the hypothesis that APE1 forms a transient complex with DNA and that this APE1–DNA complex does not accelerate nonspecific DNA binding by DNA glycosylases but stabilizes intermediate stages of catalysis between a DNA glycosylase and its DNA substrate. The nature of the stabilization effect is probably related to DNA helix distortion by APE1, which facilitates bypassing of certain steps of catalytic complex formation by a DNA glycosylase, for example, DNA bending and flipping out of the damaged base. Due to acceleration of catalytic complex formation steps, we observed an increase in the rate of catalytic reaction of DNA glycosylases. The loss of the N-terminal domain of APE1-WT leads to a loss of both stabilization of the enzyme–substrate complex and active disruption of the enzyme–product complex and thereby of the ability to modulate DNA-N-glycosylase activity.