| Nucleic Acids Research | Pages |
3,N4-Ethano-2[prime]-deoxycytidine: chemistry of incorporation into oligomeric DNA and reassessment of miscoding potential
Introduction
Materials And Methods
Chemicals
DNA synthesis and purification
Reaction with monomer
Mass spectrometer analysis
Enzymatic digestion
Primer extension reactions
Results
Physicochemical studies
Miscoding by ethano-dC and etheno-dC
Discussion
Acknowledgement
References
3,N4-Ethano-2[prime]-deoxycytidine: chemistry of incorporation into oligomeric DNA and reassessment of miscoding potential
Received August 30, 1999; Revised and Accepted October 29, 1999
ABSTRACT 3,N4-Ethano-2[prime]-deoxycytidine (ethano-dC) may be incorporated successfully into synthetic oligodeoxynucleotides by omitting the capping procedure used in the automated DNA synthetic protocols immediately after inserting the lesion and in all iterations thereafter. Ethano-dC is sensitive to acetic anhydride found in the capping reagent, and multiple oligomeric products are formed. These products were identified by examining the reaction of ethano-dC with the capping reagent, and several acetylated, ring-opened products were characterized by electrospray mass spectrometry and collision induced dissociation experiments on a tandem quadrupole mass spectrometer. A scheme for the formation of the acetylated products is proposed. In addition, the mutagenic profile of ethano-dC was re-examined and compared to that for etheno-dC. Ethano-dC is principally a blocking lesion; however, when encountered by the exo- Klenow fragment of DNA polymerase, dAMP (22%), TMP (16%), dGMP (5.3%) and dCMP (1.2%) were all incorporated opposite ethano-dC, along with an oligomer containing a one-base deletion (0.6%).
INTRODUCTION
Antitumor drugs in the nitrosourea family are powerful DNA alkylating agents, and many form exocyclic DNA adducts and inter-strand cross-links (1,2). N,N[prime]-bis(2-chloroethyl)-N-nitrosourea (BCNU) alkylates at N-3 of 2[prime]-deoxycytidine and then cyclizes to form 3,N4-ethano-2[prime]-deoxycytidine (ethano-dC; 1, see Scheme 1), a substance that may be the cause of the cytoxic and carcinogenic properties of BCNU (3). In a previous publication (4), we described the synthesis of 3,N4-ethano-2[prime]-deoxycytidine and its efficient incorporation into oligomeric DNA using solid-phase automated DNA synthesis and the 5[prime]-dimethoxytrityl-3[prime]-phosphoramidite derivative of 1.
Scheme 1.
However, at a later date when we came to synthesize a second series of oligomers containing 1 under what appeared to be an identical set of conditions, exceedingly poor yields of the desired materials were obtained, and purification proved to be very difficult. A careful evaluation of all of the reagents involved in the synthesis revealed that the quality of the acetic anhydride (Ac2O) in the capping step was the chief factor in determining yield. Contrary to what might be expected, the purer the Ac2O, the lower the yield of the desired oligomer. By contrast, the greater the concentration of acetic acid (AcOH) in the capping reagent, the better the yield and purity of the DNA.
Initially, we examined a series of capping agents containing various ratios of Ac2O to AcOH in the synthesis of the tetramer TTXT (where X = ethano-dC). However, even at low ratios, the yields of the oligomer were still unacceptable, although the HPLC chromatogram of the products became much less complicated. Thus, it became obvious that in the original synthetic work (4), the capping reagent probably contained very little acetic anhydride. We have now found that, from a practical point of view, the syntheses of oligomers containing an ethano-dC residue are best conducted by using a modified protocol in which the capping agent is omitted from the iterative procedure immediately after the insertion of the modified nucleoside and from all steps thereafter. Using this method, modified DNA of excellent quality was obtained, and oligomers truncated by one or more deoxynucleotides were not observed by gel electrophoresis or electrospray mass spectrometer analysis.
We have also examined the effect of acetic anhydride on DNA containing an ethano-dC residue and on an O-protected form of ethano-dC (2), in order to determine the patterns of transformation that occur. In addition, we have re-examined the miscoding potential of this lesion because it was not clear if a pure oligomer had been used in our previous work (5) given the difficulty of separating DNA containing an ethano-dC residue from that which contains a further modified ethano-dC group. The results of these studies are similar to those obtained previously (5).
MATERIALS AND METHODS
Chemicals
Ethano-dC (), its bis(O-TBDMS) derivative (), and the 5[prime]-O-DMT-3[prime]-O-phosphoramidite of ethano-dC and etheno-dC were synthesized according to the previously published methods (4). Other phosphoramidites and all other reagents, including the capping reagent, used in the synthesis of DNA were purchased from Perkin-Elmer/Applied Biosystems (Foster City, CA). Acetonitrile (HPLC grade) and triethylamine were purchased from Fisher Scientific (Springfield, NJ). The latter was redistilled before each use.
DNA synthesis and purification
DNA oligomers were prepared on a Perkin-Elmer/Applied Biosystems 394 DNA Synthesizer (Foster City, CA) using standard phosphoramidite protocols for the synthesis of the normal oligonucleotides and oligomers containing etheno-dC. However, for those oligomers containing an ethano-dC residue, the capping step was omitted after the incorporation of this moiety and for each iteration thereafter. At the conclusion of each synthesis, the product was deprotected and released from the resin by incubation in 1 ml of 28% ammonia solution for 16 h at 55°C. The aqueous phase was evaporated to dryness using a lyophilizer (Savant Instruments, Farmingdale, NY), and the crude product was re-dissolved in water.
A double-pass HPLC procedure was used for the purification of the DNA which employed a Waters HPLC system (Milford, MA) consisting of a Model 600 MS solvent delivery system and a µ-Bondapak C18 column (3.9 × 300 mm; 10 µ) coupled to a Waters 991 photodiode array detector. Solvent conditions consisted of a gradient of 16-35% acetonitrile and triethylammonium acetate buffer (0.1 M, pH 7.2) over a period of 30 min using a flow rate of 1 ml/min. Collected fractions were pooled, evaporated to dryness, and the residue was treated with 80% AcOH/water at ambient temperature for 30 min to remove the 5[prime]-terminal dimethoxytrityl moiety. The solution was again evaporated to dryness, then dissolved in water and filtered. The solution was rechromatographed using the same column and mobile phase as above, and a gradient of 5-20% acetonitrile over 30 min was employed to obtain the purified DNA. Aliquots were taken at both purification stages and dried for mass analysis.
Reaction with monomer
Compound 2 (50 mg) was allowed to react with the standard capping reagent (Applied Biosystems) at room temperature for 10 min. The mixture was evaporated to dryness under vacuum, and 1 ml of water was added to the residue. An aliquot was then diluted 1000-fold in 50% acetonitrile/water and used for analysis on a Quattro LC Mass Spectrometer. The remaining portion was purified by semi-preparative HPLC to isolate the major, stable reaction product for 1H NMR characterization. The 1H NMR spectra of compound 4 was recorded on a Bruker AC-250 spectrometer in CDCl3 solvent. Chemical shifts are reported in p.p.m. relative to trimethylchlorosilane.
Mass spectrometer analysis
The molecular weight of each oligonucleotide was determined on a Trio-2000 Mass Spectrometer System (Micromass, Ltd, Manchester, UK). Samples were diluted with 40% acetonitrile/water to a concentration of ~10 pmol/µl, then introduced into the electrospray ion source. The source was coupled to a syringe pump (Harvard Apparatus Inc., South Natick, MA) supplying 40% acetonitrile/water containing 1% triethylamine at 12 µl/min. Nitrogen was used both as a drying gas (250 l/h) and for nebulization (15 l/h). The instrument was operated in the negative ion spray mode with the high voltage set to -3.3 kV and the source temperature at 70°C. The sampling cone was fixed at -35 V, and the lenses were adjusted to maximize transmission. A MassLynx data system acquired signals over a range of m/z 200-1500 in 8 s. Approximately eight scans were made in the multi-channel averaging (MCA) mode with automated base-line adjustment and digital filtration.
Mass spectrometer analysis of compounds related to the monomeric bis(O-TBDMS)-ethano-dC (2) were conducted on a Quattro LC triple quadrupole mass spectrometer system (Micromass, Ltd, Manchester, UK). Infusion experiments were carried out with the sample diluted to ~1 µM in 50% acetonitrile/water. A syringe pump (Harvard Apparatus, Natic, MA) supplied a constant flow of 50% acetonitrile/water at 5 µl/min to the source. The instrument was operated in the positive ion mode with the probe voltage set to 3.5 kV and the cone voltage set to 34 V. The source temperature was set to 100°C while the desolvation temperature was 150°C. Nitrogen was used as the drying gas and for nebulization at 10 and 1.4 l/min, respectively. The range m/z 300-1600 was scanned in 5 s, averaging several scans for each spectrum. Tandem MS experiments used argon at 4.3 × 10-3 mbar for the collision gas, and the collision energy varied from 32 to 37 eV depending on the material under study. The instrument was scanned over m/z 150-1600 in 10 s.
Fast atom bombardment (FAB) mass spectra were acquired on a Kratos MS890 magnetic mass spectrometer. Xenon atoms (7 kV) were used to sputter sample ions from a glycerine/thioglycerine matrix, and data was acquired in the negative ion mode on a Kratos DS 50 data system.
Enzymatic digestion
The oligodeoxynucleotide (3.0 µg) containing an ethano-dC moiety was digested with nuclease PI (2 U) and alkaline phosphatase (3 U) as described previously (6). The digested sample was extracted with methanol, evaporated to dryness, then analyzed by HPLC using a Supelcosil LC-18S column (4.6 × 250 mm; Supelco Inc., Bellefonte, PA). The column was eluted with a linear gradient of 0-30% methanol and 50 mM sodium phosphate (pH 4.6) over 30 min at a flow rate of 1.0 ml/min.
Primer extension reactions
Primer extension reactions were carried out using a 32P-labeled primer d(AGAGGAAAGT) (10mer) with a 24mer template d(CCTTCXCTACTTTCCTCTCCATTT), X = dC, ethano-dC or etheno-dC. The reactions were catalyzed by the exo- or exo+ Klenow fragment of DNA pol I (USB Corp., Cleveland, OH) and were carried out at 25°C in 10 µl of a buffer containing 50 mM Tris-HCl (pH 8.0), 8 mM MgCl2, 5 mM 2-mercaptoethanol and the four dNTPs (100 µM each) (5). To quantify the miscoding specificity, the primer extension reactions were conducted at 25°C for 1 h. To quantify all base substitutions and deletions, the reaction products were subjected to gel electrophoresis on two-phase 20% polyacrylamide gels (15 × 72 × 0.04 cm) containing 7 M urea in the upper phase and no urea in the lower phase (7). Bands were identified by autoradiography, excised from the gel and the radioactivity of each was measured by a Wallace liquid scintillation counter.
RESULTS
Physicochemical studies
Oligomers. The failure to reproduce our earlier work (4) led us to examine simpler systems in order to characterize the aberrant chemistry occurring during automated synthesis. The stability of ethano-dC (1) towards all of the reagents used in the DNA synthesizer was checked. None of these agents seemed to effect 1 as determined by HPLC, but the action of the capping reagent was difficult to examine because of the acetylation of the OH groups on the deoxyribose residue. When the oligodeoxynucleotide d(TTXT) (X is ethano-dC) was synthesized using standard procedures and then analyzed by HPLC, the chromatogram of the 5[prime]-DMT-protected DNA and the deprotected material revealed multiple products. We then examined the synthesis of this oligomer as a function of the composition of the capping reagent by increasing the percentage of acetic acid in the commercial reagent. Four concentrations were tested including 100, 50, 30 and 10% capping reagent with the balance constituting AcOH. In the case of the synthesis with 50% capping reagent, the chromatogram of the 5[prime]-DMT protected DNA (Fig. 1A) displayed a single peak with several unresolved components while the deprotected oligomer (Fig. 1B) had many products present. The major component in Figure 1B [retention time (rt) ~24 min] was collected, dried and analyzed by FAB mass spectrometry, and the data indicated that this substance had a molecular mass of 1225 Da, 60 Da greater than the desired oligomer. The next most intense peak in the chromatogram (rt ~26 min) was found to be the desired product and had a molecular mass of 1165 Da. In the case where the capping reagent was 100% Ac2O, the abundance of secondary DMT-protected products in the HPLC chromatogram was considerably greater than that for the 50% reagent, while reaction with 10 and 30% Ac2O produced a product distribution similar to the data at 50% Ac2O (Fig. 1A). Taken together, these experiments revealed that a capping reagent containing highly pure Ac2O yielded multiple oligomeric products, whereas the capping reagent with little Ac2O (mostly AcOH) resulted in a higher percentage of the desired oligomer.
Figure 1. HPLC chromatograms of a DNA tetramer, d(TTXT) where X is ethano-dC, synthesized under conditions where the capping reagent consisted of a 1:1 mixture of acetic acid and acetic anhydride. (A) The 5[prime]-dimethoxytrityl protected products after release from the resin with 28% ammonia at 55°C for 16 h, and (B) the products after first pass HPLC purification and deprotection with 80% acetic acid at room temperature for 30 min.
At this point the syntheses of the dimer d(XT) and the trimer d(TXT) were undertaken with and without the use of the capping reagent in the synthetic protocol. HPLC analyses of the 5[prime]-dimethoxytrityl-protected trimer are shown in Figure 2. It is clear from the data that when the capping step is included in the synthetic protocol, a number of products are formed, and elimination of this step produces a single major component. The 5[prime]-dimethoxytrityl-protected trimer from the experiment in which the capping reagent was eliminated was analyzed by FAB mass spectrometry and showed the correct molecular mass (1163 Da). No other products were evident in the mass spectrum. Based on these results, a modified DNA synthesis protocol was developed in which the capping reagent was omitted immediately after the introduction of the ethano-dC residue and at all steps thereafter. Using this procedure, the following oligomers were synthesized:
| d(CCTTCXCTACTTTCCTCTCCATTT) | (I) |
| d(CCTTTXCTACTTTCCTCTCCATTT) | (II) |
| d(CATGCTGATGAATTCCTTCXCTACTTTCCTCTCCATTT) | (III) |
| d(CCATAXGTACTTC) | (IV) |
| d(TCCTCCTGXCCTCTC) | (V) |
Figure 2. HPLC chromatograms of a 5[prime]-dimethoxytrityl-protected trimer d(TXT) where X is ethano-dC after synthesis with a normal capping reagent (A), and after synthesis where the capping step was deleted after the addition of ethano-dC and from all iterations thereafter (B).
All of these oligomers, after the normal isolation and purification procedures, presented a single peak during each HPLC analysis. Oligomer I was selected for complete physicochemical characterization. The HPLC analysis (Fig. 3) and the electrospray mass spectrum (Fig. 4) both confirmed its high purity, and the latter provided the correct molecular mass, 7117.8 ± 4.0 Da (calc. 7117.7 Da). This 24mer containing a single ethano-dC lesion was also digested enzymatically in order to analyze the deoxynucleotide composition by HPLC. To determine both ethano-dC and etheno-dC using the same conditions (Fig. 5A), a linear gradient of 0-30% methanol and 50 mM sodium phosphate (pH 4.6) was used to separate all four deoxynucleosides and standards of ethano-dC (rt ~15.2 min) and etheno-dC (rt ~29.0 min). When the ethano-dC-modified 24mer was digested, ethano-dC was the only modified deoxynucleoside (Fig. 5B) that was detected, as shown by the retention time and the UV spectrum (286 nm). The ratio of deoxynucleosides (dC:ethano-dC:T:dA) recovered from the digested sample was 10.3:1:10.7:2.1 and was approximately equivalent to the theoretical ratio (10:1:11:2). Oligomer I was then used for the biological studies discussed later.
Figure 3. HPLC chromatogram of the deprotected 24mer d(CCTTCXCTACTTTCCTCTCCATTT) synthesized by omitting the capping step from the DNA synthesis protocol immediately after the addition of ethano-dC and from all iterations thereafter.
Figure 4. Electrospray ionization mass spectrum of a fraction of the oligomer collected from the major peak in the chromatogram in Figure 3. Molecular mass was determined to be 7117.8 ± 4.0 Da, and no significant peaks from any impurity were found in the spectrum.
Figure 5. HPLC analysis of the enzyme digest of d(CCTTCXCTACTTTCCTCTCCATTT). (A) Standard markers for the four deoxynucleosides (5 nmol each), ethano-dC and etheno-dC, and (B) the chromatogram of the product of enzymatic digestion.
Monomer (ethano-dC). In order to assess the effects of the capping reagent on the heterocyclic moiety of ethano-dC (1), the 3[prime],5[prime]-bis(O-TBDMS) derivative 2 appeared to be the best choice of substrate because it is soluble in organic solvents and it avoids the complication of deoxyribose acetylation that occurs with the parent compound. The electrospray (ESI) mass spectrum of the crude reaction mixture (Fig. 6) from the reaction of 2 with the capping reagent showed five molecular ion peaks representing the starting material at m/z 482 (M+H)+ and four major reaction products designated A, B, C and D at m/z 410, 524, 542 and 602, respectively. Other minor peaks in the ESI mass spectrum also appeared in the control reaction when 2 was omitted from the system. Each of the product ions was analyzed by MS/MS to obtain additional structural information. Two of the products were further characterized by collisional dissociation of the molecular ion in the high pressure region of the ESI source followed by further activation of the mass selected daughter ion in the collision cell of the mass spectrometer, a technique often referred to as (MS)3.
Two products observed in the crude reaction mixture at m/z 524 and 410 are related by the loss of a single TBDMS group from the deoxyribose. Figure 7 shows the (MS)3 mass spectrum obtained for the product utilizing the fragmentation: m/z 524 -> m/z 180 -> products. The simple addition of an acetyl group to the starting material is supported by the initial loss of 42 Da from the ion at m/z 180 (BH2+). The ion at m/z 138 and subsequent fragment ions that appear in the spectrum are identical to those found in the MS/MS spectrum of the starting material. The proposed structure for this acetylated product is 3 as shown in Scheme 2.
Scheme 2.
The MS/MS spectrum of the ion at m/z 542 (M + 60) is shown in Figure 8. The most abundant daughter ion (m/z 198, BH2+) is formed by cleavage of the glycosidic bond and represents the nucleobase portion of the molecule. This peak is 60 Da greater in mass than the corresponding ion in the MS/MS spectrum of the starting material. Less intense ions at m/z 156 and 139 represent the loss of 42 (acetyl) and 59 (acetate) from m/z 198. Other ions in the spectrum result from fragmentation of the deoxyribose moiety, and they are not present in the (MS)3 analysis when the ion at m/z 198 was selected by the first quadrupole and dissociated in the collision cell.
This substance was also isolated from the reaction mixture by semi-preparative HPLC and subjected to 1H NMR analysis. Chemical shifts are reported in p.p.m. relative to trimethylchlorosilane {[delta] 7.43 (d, J = 8 Hz, 1H, H-6), 7.40 (s, 1H, -NH), 6.21 (t, J = 7 Hz, 1H, H-1[prime]), 5.82 (d, J = 8 Hz, 1H, H-5[prime]), 4.33 (m, 1H, H-4[prime]), 3.96-3.66 (m, 7H, H-5[prime], H-3[prime], NCH2CH2NH), 2.18 (m, 1H, H-2[prime][beta]), 1.96 (m, 4H, H-2[prime][alpha], -COCH3), 0.84 [s, 9H, SiC(CH3)3], 0.82 [s, 9H, SiC(CH3)3], 0.03 [s, 6H, Si(CH3)2], 0.01 [s, 6H, Si(CH3)2]}. Structure 4 is assigned to this compound, and it represents the simple hydrolysis product of compound 3. From the molecular mass data, the final product having m/z 602 would appear to be related to 4 by the addition of a second molecule of acetic acid, and in fact the MS/MS spectrum (Fig. 9A) is similar to that of 4. Cleavage of the glycosidic bond and loss of the bis(O-TBDMS)-2[prime]-deoxyribose moiety gives rise to the ion at m/z 258, and this is followed by the loss of acetyl (m/z 216) and acetylamino (m/z 199) groups from the nucleobase. However, unlike 4 which is completely stable to ammonia, treatment of the reaction mixture from the action of the capping reagent on 2 with concentrated aqueous ammonia converts this substance to a compound with a molecular ion at m/z 601 (Fig. 9B). Thus, in addition to a second acetyl group, the compound must also contain a labile OH group capable of exchanging with ammonia. The structure assigned to compound 6 (Scheme 2) is in accord with this fragmentation pattern. Of interest is the fact that the spectra of both 6 and the ammonia exchange product, now assigned structure 7, show a peak at m/z 240 corresponding to the losses of H2O and NH3 from the ions at m/z 258 and 257, respectively, thus adding further evidence to the suggested structures.Miscoding by ethano-dC and etheno-dC
Primer extention reactions catalyzed by the exo- or exo+ Klenow fragment of DNA pol I were conducted in the presence of four dNTPs. Using an unmodified template, primer extention occurred rapidly to form the fully extended product (Fig. 10). However, with an ethano-dC-modified template, the primer extention was retarded at one base before the lesion and opposite the lesion. The extent of elongation past ethano-dC increased with incubation time. Approximately 45% of the product was the fully extended primer after 60 min when the exo- Klenow fragment was used (Fig. 10A), but the quantity of this product dropped to 6% when the exo+ Klenow fragment was used under the same conditions (Fig. 10B). In the latter reaction, the proofreading function of the exo+ fragment might minimize base insertion opposite the abnormal deoxynucleotide. Under similar experimental conditions, etheno-dC also allowed extension and formed 76 and 8.5% of the completed oligomer when the exo- or exo+ Klenow fragment was used respectively (data not shown). Thus, the efficiency for replication past ethano-dC was less than that for etheno-dC.
Figure 6. Electrospray ionization mass spectrum of a fraction of the crude reaction mixture after reaction of bis(O-TBDMS)-ethano-dC with the standard capping reagent (Applied Biosystems). Four products (A, B, C and D) and the starting material are labeled. Other peaks in the spectrum were also found in the control reaction.
Figure 7. Electrospray MS/MS spectrum of BH2+ (formed by glycosidic bond cleavage in the molecular ion and hydrogen atom transfer) from product B (Fig. 6). The ion was formed in the high-pressure region of the ion source, mass selected by the first mass spectrometer and dissociated in the collision cell of the MS/MS instrument. Masses of the products from the fragmentation of this ion were determined by the second mass spectrometer.
Figure 8. The collision-induced dissociation products obtained from the molecular ion of product C (m/z 542) in the reaction mixture. The BH2+ ion at m/z 198 is 60 Da greater than the corresponding ion in the starting material 2, indicating a modification to the nucleobase portion of the molecule.
Figure 9. Collision-induced dissociation products obtained from the molecular ion of product D (m/z 602) in the reaction mixture (A), and a related product formed by the addition of concentrated ammonia to the product mixture (B). In both spectra the BH2+ ion, formed by cleavage of the glycosidic bond, is the base peak.
Figure 10. Time course of primer extension reactions catalyzed by exo- (A) and exo+ (B) Klenow fragments. Using unmodified or ethano-dC-modified 24mer template primed with a 32P-labeled 10mer, primer extension reactions were determined at 25°C, using 0.01 U of exo- or exo+ Klenow fragment for the unmodified template and 0.05 U for the ethano-dC-modified template.
Fully-extended reaction products were analyzed by two-phase gel electrophoresis to determine the miscoding specificities of ethano-dC and etheno-dC (6). The standard mixture of six 32P-labeled oligodeoxynucleotides containing dC, T, dA or dG at the position of the lesion and one- or two-base deletions are completely resolved by this method (Fig. 11, lanes 2 and 5). When the exo+ (lane 1) or exo- (data not shown) Klenow fragment was used, DNA synthesis on an unmodified template leads to the expected incorporation of dGMP opposite dC. However, using the exo- Klenow fragment, dAMP (22%), TMP (16%), dGMP (5.3%) and dCMP (1.2%) were incorporated opposite ethano-dC, and an oligomer containing a one-base deletion (0.6%) (lane 3) was also observed. In comparison, on an etheno-dC-modified template, the incorporations of dAMP (65%) and TMP (11%) were observed together with a small amount of dCMP (1.3%, lane 6). These data are similar to those obtained previously (5). Although both ethano-dC and etheno-dC led mainly to dAMP and TMP insertion opposite the lesion, the incorporation of dAMP opposite ethano-dC was only slightly greater than that of TMP, whereas the incorporation of dAMP opposite etheno-dC was six times greater than that of TMP. In addition, dGMP incorporation and a one-base deletion were detected opposite the ethano-dC, but not in the case of the etheno-dC lesion.
Figure 11. Nucleotide incorporation opposite ethano-dC or etheno-dC lesions. Using a 24mer template primed with a 5[prime]-32P-labeled 10mer, the primer extension reaction was conducted for 1 h at 25°C, using 0.01 U of exo+ Klenow fragment for the unmodified template and 0.05 U of exo- or exo+ Klenow fragment for the ethano-dC and etheno-dC modified templates. One third of the reaction mixture was subjected to two-phase 20% polyacrylamide gel electrophoresis (15 × 72 × 0.04 cm). Mobilities of reaction products were compared with those of 18mer standards containing dC, dA, dG or T opposite the lesion and one-base ([Delta]1) or two-base ([Delta]2) deletions (lanes 2 and 5).
When the exo+ Klenow fragment was used, only TMP incorporation (6.0%) was detected opposite ethano-dC (lane 4). However, on the etheno-dC modified template, small amounts of dAMP (4.8%) and TMP (3.6%) incorporation were also detected (lane 7). Thus, the miscoding specificity of ethano-dC is different from that of etheno-dC.
DISCUSSION
An approach to the formation of the reaction products from ethano-dC and the capping reagent used in DNA synthesis is shown in Scheme 2. In retrospect, the fact that the nucleobase part of ethano-dC is sensitive to the Ac2O in the capping agent is not surprising, although we did not anticipate it. In the cytosine base, the steady-state polarization across the system O2-N3-C4-N4 is commonly accepted to take the form shown in formula 8 (Scheme 3) and accounts for the efficient hydrogen bonding that can take place with a dG residue in DNA. Any amide resonance from the N1 nitrogen atom is probably quite weak. However, in ethano-dC the electron distribution is quite different. Two different systems are present, the amide resonance between N1 and O2 and the amidino conjugation across N3-C4-N4 which renders the N4 nitrogen atom much more basic than the same atom in 8. This basicity stabilizes the formation of the quaternary acetyl compound 3, and conversion to the more stable imide system accounts for its facile hydrolytic ring-opening to give 4. The formation of either 6 or 7 was unexpected. These are related via compound 10 (Scheme 4) by the addition of either water or ammonia, respectively. It is not clear if this diacetylated product (10) arises by a second acetylation of 3 or if 2 first undergoes acetylation at C5 followed by acetylation at N4. Nevertheless, the fact that 10 is not seen in the quenched reaction mixture is perhaps not surprising given that both the hydrolytic ring opening and the addition of a nucleophile such as water (or ammonia) to the 5,6 double bond should both take place easily. The 5,6-double bond in 10 is certainly electrophilic in nature given the electron withdrawing groups that are attached to it. In addition, this hydration (or ammonia addition) permits the formation of the stabilizing enolic form of the 1,3-dicarbonyl system depicted in 11 or 12 (Scheme 4), tautomers of 6 and 7, respectively.
Scheme 3.
This unusual chemistry now allows us to resolve the irreproducible nature of the synthesis of oligomeric DNA containing an ethano-dC residue that we encountered after our initial but fortuitous success. Fortunately, the solution to the problem turned out to be exceedingly simple-the omission of the capping step at the point of introduction of the abnormal base and at every iterative step thereafter.
Scheme 4.
ACKNOWLEDGEMENT
This research was supported by a grant from the National Institutes of Health (CA47995).
REFERENCES
*To whom correspondence should be addressed. Tel: +1 631 632 8866; Fax: +1 631 632 7394; Email: francis{at}pharm.sunysb.edu
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||