Vanderbilt Institute of Chemical Biology



Discovery at the VICB







New Insight into the Toxicity of a Potent Natural Product


By: Carol A. Rouzer, VICB Communications
Published: August 2, 2017


Studies of the structure of yatakemicin-DNA adducts and their repair by DNA glycosylase AlkD help to explain the high potency of this bacterial toxicant. 


Many natural and manmade toxicants are alkylating agents that exert their deleterious effects by covalently reacting with a base of DNA to form an adduct. An example of a natural alkylating agent is yatakemycin (YTM, Figure 1), which is produced by the bacterium Streptomyces sp. TP-A0356 as a defense against other bacteria. YTM is a very potent alkylating agent that reacts at the 3-position of adenine bases in DNA to form 3-yatakemycinyl-2′-deoxyadenosine adducts (YTMA, Figure 1). To protect itself, Streptomyces sp. TP-A0356 uses the base excision repair (BER) pathway (FIgure 2) to remove any YTMA adducts that might form with its own DNA. BER starts with the action of a DNA glycosylase that hydrolyzes the glycosidic bond joining an adducted base to the deoxyribose backbone of the DNA, yielding an apurinic (AP) site and adducted adenine. Streptomyces sp. TP-A0356 expresses the enzyme YtkR2, a specialized glycosylase that efficiently removes YTMA from adducted DNA. Recently, however, Vanderbilt Institute of Chemical Biology member Brandt Eichman and his laboratory reported that AlkD, a DNA glycosylase from Bacillus cereus, is also capable of removing YTMA adducts from DNA (Mullins et al. Nature, 2015, 527: 254-258) (Figure 1). In a new report, they further investigate the interaction of YTM with DNA and YTMA repair by AlkD. Their findings provide new insight into the cause of the extremely high toxicity of YTM and the catalytic mechanism of Alk D (E. A. Mullins, R. Shi, and B. F. Eichman. Nat. Chem. Biol., published online July 24, 2017, DOI 10.1038/nchembio.2439).



FIGURE 1. Alkylation of DNA at a deoxyadenosine residue by YTM to yield YTMA-DNA and hydrolysis of the YTMA glycosidic bond by AlkD to yield AP-DNA and YTMAde. Figure reproduced by permission from Macmillan, Ltd. from E. A. Mullins, R. Shi, and B. F. Eichman. Nat. Chem. Biol., published online July 24, 2017, DOI 10.1038/nchembio.2439). Copyright 2017.




FIGURE 2. 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.




AlkD originally came to the attention of the Eichman lab as a result of the enzyme's ability to remove large, bulky DNA adducts. This is unusual for a DNA glycosylase because most of these enzymes work by flipping the adducted base out of the DNA double helix into an active site pocket where protein residues are available to catalyze glycosidic bond cleavage. The size of the adduct that can be removed is limited by the size of the pocket (Figure 3). In contrast, AlkD can remove very large adducts as long as they carry a positive charge that helps to destabilize the glycosidic bond. In their earlier studies, the Eichman lab discovered that this unusual property of AlkD can be explained by its catalytic mechanism that does not depend on base flipping. Instead, AlkD binds to the minor groove of DNA and bends the helix to expose the glycosidic bond to three catalytic residues, aspartate-113, tryptophan-187, and trytophan-109. Together, these residues promote hydrolysis of the glycosidic bond by stabilizing the transition state of the reaction (Figure 3).



FIGURE 3. Top) Diagrammatic representation of the crystal structure of human alkyladenine DNA glycosylase (AAG, a base-flipping DNA glycosylase) in complex with a substrate. The damaged nucleobase (magenta) is no longer paired to its complement (green). Instead, it has flipped out of its position in the DNA duplex and occupies the AAG active site. An intercalating residue from the enzyme (gray) fills the space left by the damaged base. (Bottom) Diagrammatic representation of the crystal structure of Bacillus cereus AlkD in complex with a substrate. Note that the damaged nucleobase (magenta) and its complement (green) remain stacked in the DNA duplex. Figure kindly provided by Brandt Eichman (Copyright 2015, Brandt Eichman).




In their original studies, the Eichman lab showed that deletion of the gene that encodes AlkD (alkD) increased the sensitivity of Bacillus anthracis to YTM toxicity. They confirmed these findings, and also investigated the effects of deletion of the gene uvrA, which encodes a protein required for nucleotide excision repair (NER), the pathway most often used for repair of bulky DNA adducts. They found that deletion of either gene increased YTM sensitivity in the bacteria, though a larger effect was observed in the case of uvrA. These findings suggested that both BER and NER act to repair YTM-mediated DNA damage in B. anthracis. However, the investigators were surprised that the protective effect appeared to be much smaller than expected. In other words, bacteria expressing both genes were still highly susceptible to the toxicity of YTM.


To better understand why YTM is so toxic, the investigators explored the structure of a YTMA adduct in a DNA oligonucleotide (YTMA-DNA) by circular dichroism and X-ray crystallography. Although the fiber diffraction pattern from YTMA-DNA crystals did not afford an atomic resolution X-ray crystal structure, their data indicated that the DNA was in the B-form. Thus, they used computational approaches to model the YTM moiety into B-form DNA. Their model indicated that YTM fits neatly into the minor groove of DNA, with only its hydrophilic substituents contacting water and its three aromatic rings forming CH-π interactions with the deoxyribose groups of both strands (Figure 4). Although individually weak, this large number of CH-π interactions can combine to hold YTM very tightly to the DNA while also sealing the two DNA strands together and limiting access to the glycosidic bond. Consistently, addition of a YTMA adduct to an oligonucleotide increased its denaturation temperature by 36oC, and the half-life of spontaneous hydrolysis for the glycosidic bond in YTMA-DNA was 1.2 years versus 1.1 weeks for the 3-methyl-2′-deoxyadenosine adduct in DNA.





FIGURE 4. (A) Model of YTMA-DNA showing the hydrogen atoms (dotted lines) forming CH-π interactions with YTMA. (B) Close-up view of the YTMA-DNA model, showing how the glycosidic bond is protected from the aqueous environment. Figure reproduced by permission from Macmillan, Ltd. from E. A. Mullins, R. Shi, and B. F. Eichman. Nat. Chem. Biol., published online July 24, 2017, DOI 10.1038/nchembio.2439). Copyright 2017.



The structure of YTMA-DNA suggested that hydrolysis of the glycosidic bond would be a significant challenge. However, kinetic studies revealed that AlkD could increase the rate of hydrolysis by a factor of 107. A crystal structure of a complex of AlkD with YTMA-DNA showed that the glycosidic bond had been hydrolyzed to yield an apurinic site (AP-DNA) that was retained in the AlkD active site along with the other product of the reaction, yatakemycinyladenine (YTMAde) (Figure 5). Binding of the DNA in the AlkD active site induced a 30o bend in the helical axis, tilting it away from the protein and inducing a widening of the minor groove to expose the site where the glycosidic bond had been (Figure 6). The YTMAde remained bound in a cleft that follows the minor groove of the DNA. Only a few interactions between the substrate and the protein were evident, including hydrogen bonds to glutamine-38 and lysine-156 and CH-π interactions with tyrosine-27 and the catalytic residues tryptophan-109 and tryptophan-187. Additional water-mediated hydrogen bonds were also present. Mutation of tyrosine-27, glutamine-38, or lysine-156 to alanine had no significant effect on AlkD activity, suggesting that the protein-substrate interactions formed by these residues were not critical to catalysis. In contrast, mutation of any one of the three catalytic residues, aspartate-113, tryptophan-109, or tryptophan-187 to alanine substantially reduced but did not eliminate activity. These findings suggested some redundancy in the role of each of the catalytic residues. The investigators hoped to obtain a crystal structure of an AlkD-YTMA-DNA complex in which the glycosidic bond of the substrate remained intact. To this end, they created a substrate in which the adducted adenosine contained a 2′-fluroarabinose ring in place of 2′-deoxyribose. The fluoro group stabilizes the glycosidic bond, and this approach had proven successful in the past for obtaining structures of many DNA glycosylase/substrate complexes. However, in the case of AlkD, glycosidic bond cleavage occurred, even in the presence of the fluoro modification, providing further evidence that AlkD is a powerful enzyme against YTMA adducts.




FIGURE 5. Two views of the crystal structure of AlkD complexed with AP-DNA and YTMAde. Figure reproduced by permission from Macmillan, Ltd. from E. A. Mullins, R. Shi, and B. F. Eichman. Nat. Chem. Biol., published online July 24, 2017, DOI 10.1038/nchembio.2439). Copyright 2017.




FIGURE 6. (A) Close-up view of YTMA-DNA showing the position of the glycosidic bond within the DNA double helix. (B) Comparable view of the crystal structure of AP-DNA and YTMAde in complex with AlkD. AlkD distorts the DNA helix, enabling the catalytic residues to gain access to the glycosidic bond. Figure reproduced by permission from Macmillan, Ltd. from E. A. Mullins, R. Shi, and B. F. Eichman. Nat. Chem. Biol., published online July 24, 2017, DOI 10.1038/nchembio.2439). Copyright 2017.



All of their data suggested that AlkD should be a highly effective glycosylase for YTM-adducted DNA. However, these findings did not agree with the relatively weak protection AlkD afforded to B. anthracis in the toxicity studies. To attempt to understand this discrepancy, the researchers performed additional kinetic studies. They found that, although the cleavage of a single molecule of YTMA-DNA by AlkD occurred very rapidly, processing of subsequent substrate molecules proceeded at progressively slower rates. This suggested that AlkD is subject to product inhibition. Addition of an AP endonuclease, which carries out the second step in BER (Figure 2), to further metabolize the product AP-DNA only partially relieved this inhibition. Further studies demonstrated that AP-DNA remains tightly bound to AlkD following glycosidic bond hydrolysis. This both prevents further processing of the AP-DNA by endonucleases as well as metabolism of new YTMA-DNA by AlkD.

The results provide key insights into the high toxicity of YTM. It forms adducts with DNA that are highly stable due to its ability to simultaneously interact with both strands, forming what is functionally a noncovalent cross-link. This structure is difficult to repair by any of the known pathways. AlkD provides one option for repair, but its high affinity for AP-DNA limits its efficiency. The product inhibition of AlkD by AP-DNA is also a new finding that would seem to apply regardless of the nature of the adduct being repaired. This raises a new question regarding what function, if any, this inhibition might have in vivo,opening the door to further studies about the role of AlkD and similar glycosylases in DNA repair.




ViewNature Chemical Biology article: Toxicity and repair of DNA adducts produced by the natural product yatakemycin









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