Chem. Res. Toxicol. 2006, 19, 463-468
463
Reactivity of the C2′-Oxidized Abasic Lesion and Its Relevance to Interactions with Type I Base Excision Repair Enzymes Marc M. Greenberg,* Cortney R. Kreller, Summer E. Young, and Jaeseung Kim Department of Chemistry, Johns Hopkins UniVersity, 3400 North Charles Street, Baltimore, Maryland 21218 ReceiVed January 2, 2006
The C2′-oxidized abasic lesion (C2-AP) is produced in DNA that is subjected to oxidative stress. C2-AP is incised by phosphodiesterases, but is not a substrate for endonuclease III even though a Schiff base is formed (Greenberg, M. M., et al. (2004) Biochemistry 43, 15217). A chemically synthesized oligonucleotide was used to study C2-AP reactivity under alkaline conditions and with nitrogen nucleophiles chosen to mimic the lysine or N-terminal proline side chains present in the active site of Type I base excision repair enzymes. Alkaline cleavage of the C2-AP lesion produces 3′-phosphoglycoaldehyde and 3′-phosphate termini. The former is degraded further to 3′-hydroxyl groups. Cleavage at the C2-AP lesion is enhanced by small peptides, which form Schiff base intermediates with the lesion. C2-AP cleavage by Lys‚Trp‚Lys and Lys‚Trp‚Gly‚Lys suggests that the inability of endonuclease III to cleave the lesion is due to the absence of appropriately positioned functional groups to take advantage of formation of the covalent intermediate. These observations leave open the possibility that the C2-AP lesion may be a substrate for other Type I repair enzymes. The C2′-oxidized abasic site (C2-AP)1 is produced when DNA is subjected to γ-radiolysis and is also formed when nucleic acids containing 5-halopyrimidine radiosensitizing agents (5-bromodeoxyuridine, 5-iododeoxyuridine) are irradiated (16). The C2-AP lesion is smaller than other types of abasic sites, which may affect translesional synthesis (7). The C2-AP structure is further distinguished by the presence of a β-hydroxy aldehyde, which enables the lesion to undergo a retroaldol condensation reaction under alkaline conditions (Scheme 1) (4).
the lesion (11, 12). The reactivity of the C2-AP lesion in the presence of alkali, pyrrolidine, and small peptides was examined to determine its alkaline lability and to explore the possibility that other Type I enzymes might catalyze cleavage.
Scheme 1. Retroaldol Condensation Cleavage of C2-AP To Produce Phosphoglycoaldehyde Termini
Retroaldol condensation cleavage of the C2-AP lesion could be physiologically detrimental because the phosphoglycoaldehyde formed at the oligonucleotide termini may produce premutagenic exocyclic adducts via reaction with nucleobases within the biopolymer (8-10). Type I base excision repair enzymes such as endonuclease III (Endo III) and formamidopyrimidine DNA glycosylase (Fpg) do not catalyze the retroaldol condensation cleavage of C2-AP, but do form Schiff bases with * To whom correspondence should be addressed. Phone: 410-516-8095. Fax: 410-516-7044. E-mail:
[email protected]. 1 Abbreviations: C2-AP, C2′-oxidized abasic site; C4-AP, C4′-oxidized abasic site; AP, abasic site; L, 2-deoxyribonolactone; PAGE, polyacrylamide gel electrophoresis; Endo III, endonuclease III; Fpg, formamidopyrimidine DNA glycosylase; Exo III, exonuclease III; Endo IV, endonuclease IV.
The C2-AP lesion is efficiently incised by the Type II BER enzymes exonuclease III (Exo III) and endonuclease IV (Endo IV) (11). Other oxidized abasic lesions that are substrates for such enzymes also interact with Type I repair enzymes (1315). For instance, the C4′-oxidized abasic lesion (C4-AP) is a substrate for Endo III, and 2-deoxyribonolactone (L) cross-links to this enzyme. Unlike the AP and C4-AP lesions, C2-AP lacks a leaving group (phosphate) at the β-position relative to the aldehyde carbon. Consequently, C2-AP does not undergo the typical lyase reaction catalyzed by BER enzymes (16, 17). However, Type I BER enzymes contain other amino acid side chains in their active sites that can hypothetically assist a retroaldol condensation cleavage of the C2-AP lesion by participating in acid-base chemistry (18). From the perspective of organic chemistry, the Schiff base of C2-AP represents a late stage intermediate in the reaction between an aldehyde and an enamine (an enolate equivalent). The principle of microscopic reversibility implies that the Schiff base can fragment to produce the retroaldol condensation products, cleaved DNA, which contains phosphoglycoaldehyde at their termini. We explored this possibility using solution reaction conditions chosen to
10.1021/tx060001q CCC: $33.50 © 2006 American Chemical Society Published on Web 02/25/2006
464 Chem. Res. Toxicol., Vol. 19, No. 3, 2006
Greenberg et al.
Table 1. Oligonucleotides and Duplexes Employed in Experiments
Figure 1. Alkaline cleavage of C2-AP (5′-32P-1) and enzymatic 3′end group analysis using T4 polynucleotide kinase. Calibration standards containing 3′-hydroxyl (3) or 3′-phosphoglycoaldehyde (5) were independently synthesized.
model those that the C2-AP lesion may encounter in the active site of a BER enzyme.
Materials and Methods General Procedures. Oligonucleotide (Table 1) synthesis was carried out on an Applied Biosystems Incorporated 394 DNA synthesizer using standard protocols. Oligonucleotides containing C2-AP were synthesized as described (19). Commercially available oligonucleotide synthesis reagents were obtained from Glen Research (Sterling, VA). 3′-Phosphoglycoaldehyde containing DNA (5) was prepared via NaIO4 oxidation of the vicinal diol. The respective vicinal diol containing DNA was obtained by standard solid-phase oligonucleotide synthesis on glyceryl-CPG support (Glen Research) and aqueous ammonia deprotection. DNA manipulation, including enzymatic labeling, was carried out using standard procedures (20). Preparative and analytical oligonucleotide separations were carried out on 20% polyacrylamide denaturing gel electrophoresis (PAGE) (5% cross-link, 45% urea (by weight)). T4 polynucleotide kinase was obtained from New England Biolabs (Beverly, MA). Lys‚Trp, Lys‚Trp‚Lys, and Lys‚Trp‚Gly‚Lys were obtained from Bachem (Torrance, CA). [γ-32P]ATP was purchased from Amersham Pharmacia Biotech. Radioactive samples were quantitated by Cerenkov counting using a Beckman LS6500 liquid scintillation counter. Quantification of radiolabeled oligonucleotides was carried out using a Molecular Dynamics Storm 840 Phosphorimager equipped with ImageQuant Version 5.1 software. Cleavage of C2-AP (5′-32P-1) in 0.1 M NaOH. 5′-32P-1 (360 nM) was incubated with 0.1 M NaOH (total volume 140 µL) at 37 °C over a period of 55 h. At indicated time points (0, 0.5, 1, 1.5, 2, 4, 6, 8, 10, 12, 25, 31, 47, 55 h), a 10 µL aliquot was removed and quenched with 1 µL of 1 M HCl. Samples were stored at -80 °C until the experiment had reached its full time course. Upon completion of the time course, each aliquot was thawed, diluted with 9 µL of formamide loading buffer, separated by PAGE, and analyzed using a phosphorimager. Reactions were carried out in triplicate. Enzymatic 3′-End Group Analysis of NaOH Cleaved C2-AP. 5′-32P-1 (1 µM, 30 µL) was incubated in 0.1 M NaOH at 37 °C. Aliquots (15 µL) were removed after 5 and 24 h and neutralized by with 1.0 M HCl. A portion (5 µL) of each aliquot was stored at -80 °C. The remaining neutralized solution (10 µL) was diluted to 400 nM using 10× T4 polynucleotide kinase buffer and H2O to produce a final reaction mixture containing 70 mM Tris-HCl (pH 7.6), 10 mM MgCl2, and 5 mM dithiothreitol in a final reaction volume of 25 µL. The reaction was incubated at 37 °C for 2 h with T4 polynucleotide kinase (10 U). The kinase-treated sample was precipitated from 0.3 M NaOAc and calf thymus DNA (0.13 mM), resuspended in 50 µL of formamide loading buffer, and analyzed by PAGE.
5′- and 3′-End Group Analysis of the NaOH Cleavage of C2AP (1) Using NaBH4. 5′-32P-1 (125 nM, 20 µL) was treated with 0.1 M NaOH for 5 h at 37 °C and neutralized with 1.0 M HCl. One-half of the solution was reacted with an equal volume of 100 mM NaBH4 at 37 °C for 30 min. The reaction was diluted with formamide loading buffer (18 µL) and analyzed by denaturing PAGE. The observations were calibrated using 5′-32P-5, which was reacted with NaBH4 in an identical manner. Analysis of the 5′terminus of the 3′-C2-AP fragment was analyzed in a similar manner using 3′-32P-1. Reaction of C2-AP (5′-32P-2) with Pyrrolidine. Pyrrolidine (6 µL, 100 mM) was mixed with 30 µL of 400 mM Tris (pH 7.62 at 25 °C), H2O (9 µL), and 1 µM 5′-32P-2 in 100 mM KCl and 10 mM MgCl2 (15 µL). Control experiments were carried out in parallel by incubating 5′-32P-2 in 50 mM KCl and 5 mM MgCl2 (30 µL), in either 30 µL of 400 mM Tris (pH 7.62 at 25 °C) or in 30 µL of water. The reactions were incubated at 37 °C, and 10 µL aliquots were removed and quenched with 10 µL of formamide loading buffer at 24, 48, 84, and 132 h. Samples were kept at -80 °C for the duration of the experiment and analyzed by denaturing PAGE. Reaction of C2-AP (5′-32P-2) with Peptides. An equal volume of Lys‚Trp‚Lys (100 µM) in 100 mM NaCl (pH 6.3) was added to 5′-32P-2 (0.1 µM) in 100 mM phosphate buffer (pH 7.5) and 100 mM NaCl (final reaction volume 25 µL) and incubated at 37 °C for 96 h or 55 °C for 24 h. A control experiment was carried out in parallel by incubating 25 µL of 5′-32P-2 (50 nM) in 100 mM NaCl and 50 mM phosphate buffer (pH 7.5) at the same temperature. Aliquots (5 µL) were removed from each solution and quenched with 5 µL of formamide loading buffer at 24, 44, and 96 h. The aliquots were analyzed by denaturing PAGE. An identical procedure was carried out using Lys‚Trp and Lys‚Trp‚Gly‚Lys at 55 °C. Sodium Cyanoborohydride Trapping of Schiff Bases between C2-AP and Peptides. 5′-32P-2 (0.1 µM) was incubated separately for 24 h with either Lys‚Trp, Lys‚Trp‚Lys, or Lys‚Trp‚Gly‚Trp at 55 °C as described above. Each sample was divided into two portions, one of which was treated with 0.1 M NaBH3CN at 37 °C for 1 h. The other portion was incubated under the same conditions with no reducing agent. Formamide loading buffer was added to each portion, and incubation was continued for 1 h at 37 °C. The samples were analyzed by denaturing PAGE.
Results Alkaline Hydrolysis of the C2-AP Lesion. The lability of the C2-AP lesion was established previously (4, 19). Cleavage of 5′-32P-1 under alkaline conditions revealed the formation of three products, two of which are formed in relatively greater quantity than the third (Figure 1). The slowest moving product comigrates with independently synthesized oligonucleotide containing the anticipated retroaldol condensation product, 3′phosphoglycoaldehyde (5′-32P-5). Further support for the assignment of this product was gleaned from reaction with NaBH4 (Figure 2A). Oligonucleotide 5 and the slower moving product
ReactiVity of a C2′-Oxidized Abasic Lesion
Chem. Res. Toxicol., Vol. 19, No. 3, 2006 465
Figure 2. Alkaline cleavage of C2-AP (A, 5′-32P-1; B, 3′-32P-1) followed by NaBH4 treatment.
Scheme 2. Alkaline Hydrolysis of C2-AP (1)
Figure 3. Alkaline cleavage of C2-AP (5′-32P-1) as a function of time. (A) Yield of individual cleavage products as a fraction of total DNA in reaction; (B) yield of individual cleavage products as a fraction of all products in reaction (3′-end groups: hydroxyl, b; phosphoglycoaldehyde, 9; phosphate, 2).
obtained upon alkaline cleavage of C2-AP are transformed into a more rapidly migrating product upon reaction with sodium borohydride. This product comigrates with the respective 3′hydroxyl-containing oligonucleotide (3), but we did not confirm whether the reduction product was 3. Our primary concern was the structure of the original C2-AP cleavage product, which these experiments reveal to be 3′-phosphoglycoaldehyde (5). Treatment of the neutralized sample with polynucleotide T4 kinase revealed that the fastest moving product obtained from 1 contained a 3′-phosphate (4) (Figure 1). The nature of the 5′-terminus of the cleavage fragment was explored using 3′32P-1. Two products were detected from cleavage of 3′-32P-1 in roughly equal amounts. The slower moving product comigrates with the oligonucleotide containing a 5′-hydroxyl terminus but is converted to the faster moving 5′-phosphate product upon NaBH4 treatment, indicating that it is fortuitous that the initial cleavage product comigrates with the oligonucleotide containing a 5′-hydroxyl group (Figure 2B). On the basis of the reactivity of the slower moving product and the characterization of the 5′-labeled cleavage fragment (above), the slower moving product observed from 3′-32P-1 is the 5′-phosphoglycoaldehyde (6) (Scheme 2). Prolonged NaOH treatment suggested that the 3′-phosphoglycoaldehyde was converted to the minor product, which
comigrated with DNA containing a 3′-hydroxyl (Figure 1). The product distribution obtained upon C2-AP hydrolysis was followed as a function of time (Figure 3). These data suggest that 3′-phosphoglycoaldehyde (5) decomposes to 3′-hydroxyl (3) but that the minor product, 3′-phosphate (4), is a primary product (Scheme 2). Subjection of independently prepared 3′phosphoglycoaldehyde to the hydrolysis conditions confirms that this end group decomposes to one containing a 3′-hydroxyl. The rate constants describing cleavage of the C2-AP lesion to form 3′-phosphoglycoaldehyde and 3′-phosphate products are 2.7 ( 0.2 × 10-5 s-1 (t1/2 ) 7.1 h) and 1.9 ( 0.1 × 10-5 s-1 (t1/2 ) 10.1 h), respectively. Reaction of C2-AP with Nitrogen Nucleophiles. The oligonucleotide containing C2-AP (5′-32P-2) was reacted with buffered pyrrolidine (10 mM) in Tris buffer (pH 7.62) as a model for what one may expect to find in the active site of a Type I base excision repair enzyme such as Fpg (N-terminal proline) or Endo III (lysine). Less than 5% of the C2-AP was cleaved over the course of 5.5 days (data not shown). Moreover, the extent of C2-AP cleavage in the presence of pyrrolidine was within experimental error of that when the lesion was incubated in Tris buffer, but was approximately 2-fold greater than that when the lesion was incubated in water. These observations are consistent with the inability of Endo III and Fpg to effect C2-AP cleavage, despite the presence of a nitrogen nucleophile capable of forming a Schiff base. In contrast, measurable amounts of cleavage above background were produced when the Lys‚Trp‚Lys tripeptide was
466 Chem. Res. Toxicol., Vol. 19, No. 3, 2006
Greenberg et al.
Figure 5. Sodium cyanoborohydride trapping of Schiff base adducts between C2-AP and peptides.
Figure 6. Peptide-induced cleavage of C2-AP containing DNA (5′32 P-2) over 24 h at 55 °C. Bkgd., buffer only; KW, Lys‚Trp; KWK, Lys‚Trp‚Lys; KWGK, Lys‚Trp‚Gly‚Lys.
Figure 4. C2-AP (5′-32P-2) in the presence of Lys‚Trp‚Lys (50 µM). (A) Autoradiogram illustrating product formation with and without tripeptide. Arrows indicate the presence the following products: a, tripeptide adduct of cleavage product; b, 3′-hydroxyl cleavage product (3); c, 3′-phosphate cleavage product (4). (B) Cleavage of 5′-32P-2 as a function of time in the presence and absence of tripeptide.
present in significantly lower concentration (50 µM) (Figure 4). Analysis of 5′-32P-2 cleavage showed that the major product contained a 3′-phosphate terminus (“c”, Figure 4A). Only faint amounts of a product that comigrates with that containing a 3′-phosphoglycoaldehyde (“b”, Figure 4A) were detected. Small amounts of a significantly slower moving product (“a”, Figure 4A), which is not observed in the absence of the tripeptide, are also detected. The slower moving product (“a”, Figure 5) could correspond to a Schiff base adduct of the tripeptide with a cleavage product containing an aldehyde. Similarly, cleavage of 3′-32P-2 in the presence of Lys‚Trp‚Lys produced a product
that comigrated with the 5′-phosphate product (21). Comparison of the intact oligonucleotide in the presence or absence of Lys‚ Trp‚Lys suggests that an adduct forms between the tripeptide and C2-AP containing DNA, as evidenced by a slight retardation in the migration of the intact DNA. Additional evidence for the peptide-DNA adduct was obtained by treating the mixture with sodium cyanoborohydride, following a 24 h incubation at 55 °C (Figure 5). A slower moving product that was stable to incubation in formamide loading buffer for 1 h at 37 °C was detected. No adduct was observed in the aliquot that was not treated with reducing agent, following incubation in formamide. However, small amounts (
