Vanderbilt Institute of Chemical Biology



Discovery at the VICB







New Way to Unhook DNA


By: Carol A. Rouzer, VICB Communications
Published:  April 11, 2017


Structural studies of the DNA glycosylase AlkZ explain how it repairs DNA inter-strand cross-links.


Chemical and physical insults, both natural and manmade, are a constant source of damage to DNA. If not repaired, this damage can lead to disease-causing DNA mutations. One form of DNA damage, alkylation, involves the adduction of molecules or fragments of molecules to nucleophilic sites on DNA bases. There are two major pathways for repair of this kind of damage, the base-excision repair (BER) pathway that generally targets small adducts, and the nucleotide excision repair (NER) pathway for removal of bulky adducts. A particularly toxic form of damage can occur through reaction of DNA with a bifunctional alkylating agent that cross-links the two DNA strands. In general, repair of these inter-strand cross-links (ICLs) requires the work of multiple repair pathways and is more likely to lead to mutation than repair of simpler lesions. Recent work, however, suggests that the BER pathway may be involved in repairing complex ICLs, at least in some cases. To learn more about this unexpected role for BER, Vanderbilt Institute of Chemical Biology member Brandt Eichman and his laboratory report on the structure and mechanism of a bacterial enzyme they call AlkZ that catalyzes the first step in BER pathway-mediated repair of a complex cross-link (E. A. Mullins, et al. Proc. Natl. Acad. Sci. U.S.A., published online April 11, 2017, DOI 10.1073/pnas.1703066114).


The first step of BER is removal of the damaged nucleobase through the action of a DNA glycosylase that cleaves the glycosidic bond between the nucleobase and the sugar (Figure 1). The result is an apurinic/apyrimidinic (AP) site, which is quickly removed by an AP endonuclease, forming a single-strand break in the DNA duplex. The break is repaired by the consecutive action of a DNA polymerase and a DNA ligase that add the correct complementary base and reunite the sugar-phosphate backbone, respectively. Although ICL repair by BER is not well characterized, prior work had shown that the eukaryotic NEIL3 DNA glycosylase is capable of unhooking psoralen and abasic site-adenine ICLs and that human NEIL1 can both unhook psoralen cross-links and remove psoralen monoadducts.




FIGURE 1. The base excision repair pathway. (A) A lesion in DNA is removed by a DNA glycosylase, which cleaves the glycosidic bond between the nucleobase and the deoxyribose forming an AP site. (B) An AP endonuclease removes the AP site by cleaving bonds in the sugar-phosphate backbone. The result is a single-strand break. (C) A DNA polymerase adds the appropriate base to fill the void left by removal of the AP site. (E) DNA ligase reunites the sugar-phosphate backbone. Figure reproduced with permission from S. S. David and S. D. Williams (1998) Chem. Rev., 98, 1221. Copyright 1998 American Chemical Society.



Azinomycin B (AZB) (Figure 2A) is a bifunctional alkylating agent produced by the bacterium Streptomyces sahachiroi. Through its reactive aziridine and epoxide rings, AZB can adduct both guanine and adenine bases at their N7 positions and form ICLs between two of these bases when they are positioned opposite one another in a GNC or GNT sequence (Figure 2B). In the ICL, AZB lies in the major groove of the DNA. To protect itself from DNA damage by this toxic metabolite, S. sahachiroi produces the enzyme Orf1, which acts as a DNA glycosylase, hydrolyzing the glycosidic bond between the adducted base and the deoxyribose on one or both sides of the ICL (Figure 2C). After confirming that Orf1 acts more generally as a DNA glycoylase, as indicated by its ability to cleave the glycosidic bond linking an N7-methyguanine adduct to a synthetic duplex oligonucleotide substrate, the Eichman lab suggested that Orf1 be renamed AlkZ.


FIGURE 2. (A) Structure of azinomycin B (AZB), which reacts at the N7 position of dG or dA bases (B) to form an inter-strand cross-link. (C) AlkZ hydrolyzes the glycosidic bond at one or both of the adduction sites, resulting in the formation of an abasic site at that location. Figure reproduced with permission from E. A. Mullins, et al. Proc. Natl. Acad. Sci. U.S.A., published online April 11, 2017, DOI 10.1073/pnas.1703066114. Copyright 2017, E. A. Mullins, et al.


AlkZ belongs to the HTH_42 protein superfamily that is notable for the presence of a winged helix (WH) structure comprising an N-terminal bundle of three α-helices and a C-terminal three-stranded β-sheet. The lack of structural information for this group of proteins, however, made it almost impossible to understand how AlkZ might execute its DNA repair function. To fill this void, the Eichman lab obtained a crystal structure for the AlkZ protein, making it the first known member of an eighth class of DNA glycosylase structural architectures. Their data confirmed that AlkZ contains not one, but three N-terminal tandem WH motifs that pack against a C-terminal β-barrel domain (Figure 3). The three WH domains form a C-shaped structure, the inner surface of which is lined with positively charged residues and is correctly sized to bind a double-stranded DNA substrate, suggesting that this cavity contains the active site of the enzyme.



FIGURE 3. (A) Two views of the crystal structure of AlkZ. The three WH domains are shown in red, green, and blue, and the C-terminal β-barrel is in yellow. (B) Topology diagram of the protein's structure, color coded as in (A).  Figure reproduced with permission from E. A. Mullins, et al. Proc. Natl. Acad. Sci. U.S.A., published online April 11, 2017, DOI 10.1073/pnas.1703066114. Copyright 2017, E. A. Mullins, et al.



To further explore the possible interaction of AlkZ with DNA, the researchers created a computational model of the enzyme in complex with a DNA duplex bearing an AZB ICL. They found that two opposite binding poses were equally favored for the substrate, so that either side of the ICL could face the protein. The model predicts that helix αI of WH2 binds in the major groove of the DNA substrate, positioning it to serve as a DNA recognition helix, a known function for WH motifs. Notably, neither of the other WH motifs were oriented in a way that they could serve this function. Also of interest was the presence of a β11-β12 hairpin structure (contributed by the C-terminal β-barrel domain) in the minor groove of the substrate model. Similar structures in DNA glycosylases help stabilize the desired substrate conformation.


These data strongly supported the hypothesis that AlkZ contains the necessary structural components to act as a DNA glycosylase. To further test this hypothesis, the investigators searched the structure for appropriately positioned carboxylate- or amide-containing residues, such as are commonly found in the active sites of DNA glycosylases. They identified two glutamine residues, Q37 and Q39, near the active site of AlkZ. Their model suggested that Q37 probably binds the DNA backbone, but Q39 is properly aligned to help position a water molecule for nucleophilic attack on the glycosidic bond. Two glutamate residues, E221 and E45, are also located near the purported active site, but their position suggests that they are less likely to be directly involved in catalysis (Figure 4). The researchers explored the importance of each of these residues by creating site-directed mutants of AlkZ in which one of the residues had been converted to alanine. They also created a mutant protein in which the β-hairpin (residues 307-313) was replaced with two glycine residues. Testing the ability of the mutants to remove N7-methylguanine from a duplex oligonucleotide revealed that the Q37A, Q39A, and β-hairpin substitution mutations all completely abrogated DNA glycosylase activity. In contrast, the E221A and E45A mutations had no effect. These results support the hypothesis that Q37, Q39, and the β-hairpin have an important role in substrate binding and/or catalysis.



FIGURE 4. (A) Close-up view of a DNA duplex containing an AZB cross-link modeled into the AlkZ active site. The protein is color-coded as in Figure 3. The DNA backbone is in gray, as are the two adducted dG residues. AZB is in magenta. The residues that were subjected to site-directed mutageneis are labeled. (B) Same as (A) but viewed following a 90o rotation. (C) and (D) Same as (A) and (B), but the DNA is shown in the second binding orientation. The two orientations differ with regard to which adducted dG residue is subject to glycosidic bond hydrolysis. (E) Results from the enzyme activity assay of wild-type AlkZ and the various mutant proteins that were evaluated. In this assay, active enzyme converts a larger oligonucleotide to a smaller one, which migrates further on the gel. Figure reproduced with permission from E. A. Mullins, et al. Proc. Natl. Acad. Sci. U.S.A., published online April 11, 2017, DOI 10.1073/pnas.1703066114. Copyright 2017, E. A. Mullins, et al.


Most DNA glycosylases employ a mechanism in which the adducted base is flipped out of the double helix into the active site prior to glycosidic bond hydrolysis. This base-flipping mechanism explains why these enzymes are restricted to substrates containing small adducts. Clearly, such a mechanism is not possible in the case of a bulky, doubly bound ICL. Fortunately, in their prior studies of the DNA glycosylase AlkD, which can repair bulky DNA adducts, the Eichman lab identified an alternative mechanism that does not require base-flipping. An interesting additional question arises regarding whether AlkZ cleaves the glycosidic bond on only one or both sides of the ICL. The ability of the DNA to bind in both orientations in their model suggests that either option is possible. The investigators note that they could also construct a model in which AlkZ binds to both sides of the ICL simultaneously. In this model, the two opposing proteins contact each other along surfaces that appear to be quite compatible. There is no current evidence for this type of dimerization of AlkZ around its substrate, but exploring this possibility is an intriguing topic for further research.


AlkZ is not the only example of a DNA glycosylase produced as a self-defense mechanism by bacteria that synthesize toxic secondary metabolites. YtkR2, an analog of AlkD is produced by a species of Streptomyces to protect themselves against the large DNA alkylating secondary metabolite, yatakemycin, which they also synthesize. Enzymes similar to AlkZ and YtkR2 are widespread among bacteria species, suggesting that these genes have become broadly disseminated and offer a survival advantage. These findings suggest that expression of DNA repair enzymes may be a common means by which bacteria develop antibiotic resistance. Future research may uncover an array of heretofore unknown enzymes that serve this function.



View PNAS article: Structure of a DNA glycosylase that unhooks interstrand cross-links









The Vanderbilt Institute of Chemical Biology, 896 Preston Building, Nashville, TN 37232-6304, phone 866.303 VICB (8422), fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2014 by Vanderbilt University Medical Center