Structural and Conformational Probing of 2-Acetylaminofluorene-Modified -1, -2 and -3 Slipped Mutagenic Intermediates

2-Acetylaminofluorene (AAF) is a prototype arylamine carcinogen that forms C8substituted dG-adduct (dG-C8-AAF) as a major DNA damage. The bulky dG-C8-AAF lesion is known to induce -1, -2, or -3 frameshift mutations depending on the base sequences around the lesion. We hypothesize that the stability of bulged-out structures slipped mutagenic intermediate (SMI) facilitates primer elongation, hence manifestation of frameshift mutations. The objective of the present study was to probe the structural/conformational basis of various dG-C8-AAF induced frameshift mutations. Here, we describe spectroscopic ( 19 F NMR and CD) and thermodynamic (UV-melting and DSC) studies of several dG-C8-FAAF-modified 16-mer DNA duplexes containing fully-paired, -2, and -3 deletion duplexes on the 5'-CTCTCGATG[FAAF]CCATCAC-3' sequence and -1 deletion duplexes on either the 5'-CTCTCGATG[FAAF]CCATCAC-3' or 5'-CTCTCGGCG [FAAF]CCATCAC-3' sequences. The results were analyzed to determine the conformational and thermodynamic basis of AAF-induced frameshift mutagenesis. We found that the dG-AAF lesion exists in a mixture of external binding B and inserted/bulge conformers and the population of the latter was in order of ‘GC’1(73%) > ‘AT’ -1 (72%) > full (60%) > -2 (55%) > -3 (37%). Thermodynamic stability was found to be in order of -1 deletion > -2 deletion > fully paired > -3 deletion duplexes. These results indicate that the stacked conformer especially in the deletion duplexes is thermodynamically more stable than the conformationally flexible external B-conformer. Previous primer extension results involving the human DNA polymerase η have shown that the frequency of deletion was in order of -1 > -2 > -3. Taken together, our results support the hypothesis that the conformational and thermodynamic stability of the SMI is a critical determinant for the induction of various frameshift mutations.


INTRODUCTION
Aromatic amines are a well-known group of mutagens/carcinogens that cause cancer initiation (1). These amines are widely present in the environment as byproducts of fossil fuel combustion, tobacco smoke, dyes, and charred meat. 2-Acetylaminofluorene (AAF) was developed originally as a pesticide, but the compound was never used for its intended purpose because of its potent liver carcinogenicity in rodents (2). Instead, AAF has become a prototype molecule for investigating the mechanisms of aromatic amineinduced carcinogenesis (3). Upon activation in the liver, AAF is converted into a highly electrophilic nitrenium ion, which in turn reacts directly with cellular DNA to form covalent DNA adducts. The most commonly formed are C8-substituted 2'deoxyguanosine adducts: the N-acetylated dG-C8-AAF, [N-(2′-deoxyguanosin-8-yl)-2acetylaminofluorene] and the N-deacetylated dG-C8-AF, [N-(2′-deoxyguanosin-8-yl)-2aminofluorene] (2,4). Although the two adducts are similar in chemical structures, they exhibit different mutagenic and repair properties. For example, dG-C8-AF is largely non-mutagenic and is correctly replicated by high-fidelity polymerases possibly. This may be due to its flexibility to accommodate both syn-and anti-glycosidic conformation.
The latter allows to forms the B-type DNA conformation, supporting Watson-crick base pairs at the lesion site (5). On the other hand, the bulky N-acetylated dG-C8-AAF lesion exists primarily in the highly distorting syn stacked (S)-or wedge (W)-conformation, resulting in a strong block of high fidelity polymerase (6). Bypass of the AAF-dG lesion leads to frame shift mutations (3,7).
DNA polymerases are the enzymes that catalyze the polymerization of 2´deoxyribonucleotides into DNA strands. Polymerases use steric and hydrogen-bonding interactions to identify the correct nucleotide for insertion opposite each base in the template strand. Duplication of DNA containing damaged bases is a challenge to DNA polymerases that normally replicate with high speed, high accuracy and high processivity in undamaged templates. When a replicative DNA polymerase encounters a chemically altered base that it is unable to copy, a process called TransLesion Synthesis (TLS) takes place during which the polymerase is transiently replaced by a specialized lesion bypass polymerase. The process by which the polymerase proceeds through the adduct in TLS may be either error-free or error pone. Error-free lesion bypass results in the preferential incorporation of a correct nucleotide opposite the damage, whereas error-prone lesion bypass leads to the preferential incorporation of an incorrect nucleotide opposite the damage (8). Consequently, error-free lesion bypass is a mutation-avoiding mechanism, whereas error-prone lesion bypass is a mutation-generating mechanism. For example, human DNA polymerase Pol η belongs to the class of Y-family DNA-polymerases.
These low fidelity polymerases have larger active sites that can accommodate bulky DNA adducts allowing them to bypass these lesions during replication. The activity of Pol η is critical in preventing the Xeroderma Pigmentosum (XPV) disease (9). Recently it was discovered that Pol η is involved in the in vivo bypass of dG-C8-AAF, which occurs by base pairing the lesion correctly with dC (10). However, dG-C8-AAF in the E. coli hot spot NarI sequence (5′-G 1 G 2 CG 3 CC-3′) is prone to induce a -1 or -2 frame shift mutation (11)(12)(13)(14)(15)(16).
A majority of AAF-induced mutations in E. coli are frameshift mutations (17) including additions and deletions of a single G:C base pair, deletions of two adjacent base pairs, and a few three-base-pair deletions (12,17). Frameshift mutations involve gain or loss of one or more base pairs relative to the original sequence, thereby causing a shift in the genetic reading frame, resulting in loss of the whole information content of the gene (18). Frameshift mutations induced by AAF can be best explained by a slippage mechanism. A specific mechanism has been proposed to explain the double-base frameshifts in the NarI sequence in which the 'GC' deletion results from a slipped mutagenic intermediate (SMI) formed during replication shown in Figure 1. In the first step, the correct nucleotide dC is incorporated opposite dG-C8-AAF. Hindering of the replication fork at the modified dG favors a misalignment, thereby causing slippage of the growing primer strand and two bases of the NarI sequence are looped out in the parental strand. Continued extension from the two nucleotide bulge in the template strand, leads to a newly synthesized strand two bases shorter than the template strand (19). Similarly, in -1 and -3 deletions polymerase extends the primer with a dC base opposite the lesion, however a substantial fraction of the primer misaligns to pair with the 5'-dG next to the lesion in -1 slippage and two bases further away in -3 slippage (shown in Figure 2 (A) (B) respectively) (20).
Using high resolution 1 H NMR, O'Handley et al (21) reported the presence of about 70% base displaced S-conformer in a AAF-modified fully paired 9-mer duplex in the CG*C sequence context. The Cho group has used 19 F NMR spectroscopy to show the exclusive existence of the S-conformer with a novel cis/trans equilibrium of the acetyl group for a fully paired 12-mer duplex in the TG*A sequence context (22). Milhe et al. (23,24) obtained 1 H NMR data on the AAF modified -1, -2 deletion duplexes on CG*C context. 70% of the AAF-population was external in the -1 deletion duplex, whereas in the -2 deletion duplex, 80% of the fluorene moiety was inserted in to the bulged double helix. However, the NMR data was difficult to interpret because of the conformational heterogeneity. Similar NMR studies were conducted on -1, -2 deletion duplexes by the N-deacetylated AF (25,26) and anti-benzo[a]pyrenediol epoxide (BPDE) (27,28). All these deletion duplexes with the exception of AAF -1 duplex, exist primarily in conformation, in which the carcinogenic moiety is inserted in to the helix (23). These modified duplexes showed consistent thermal stabilities (ΔTm= 11-15 ℃) relative to controls, which support well-stacked structures.
Recently, Schorr et al (20) studied the AAF-induced frame shift mutation in the NarI sequence (5-G 1 G 2 CG 3 CC-3) using the low fidelity human polymerase η. They found that AAF can induce -1, -2, or -3 frame-shift mutations depending upon the neighboring bases around the lesion. For example, changing of G 2 C to AT (5-G 1 G 2 CG 3 CC-3 5-G 1 ATG 3 CC-3) resulted in -3 deletion mutations, presumably due to the weakly paired A:T region around the lesion. However, the structural/conformational bases of these deletion mutations remain elusive.
In the present study we created fully paired, -3, -2 and -1 bulge duplexes using a fluorine containing AAF (FAAF) in a 16-mer template (5-CTCTCGATG*CCATCAC-3). This sequence is identical to the core sequence used by the abovementioned paper by We hypothesize that the AAF-lesion exists in a mixture of two major conformations: the stable intercalated and the unstable external conformers. The combined thermodynamic stabilities play a critical role in determining the different outcomes of deletion mutations. We have utilized dynamic 19 F NMR and CD spectroscopy to investigate the conformational profiles of several AAF-containing model deletion duplexes. Furthermore, we employed UV-melting and differential scanning calorimetry (DSC) to study their thermodynamic properties. The results present valuable structural and thermodynamic evidences for the manifestation of various frameshift mutations.

MATERIALS AND METHODS
Caution: 2-Acetylaminofluorene derivatives are mutagens and suspected human carcinogens and therefore must be handled with caution.

Preparation and characterization of FAAF modified ODNs
FAAF-modified 16-mer ODNs were prepared using the general procedures described previously (22,30). For example, 0.5 ~ 1 mg of N-acetoxy-N-2-(acetyl amino)-7fluorofluorene which was prepared from starting material 7-fluoro-2-nitrofluorene, was dissolved in absolute ethanol was added drop wise to a pH 6.0 sodium citrate buffer (10mM) containing unmodified oligomer (5-CTCTCG 1 ATG 2 CCATCAC-3) and placed in a water-bath shaker for 5 min at 37°C. Figure 5a shows a typical reverse-phase HPLC chromatogram derived from an aliquot of the reaction mixture. Unmodified oligomer appeared at 11.7 min and modified oligomers, two mono-adducts and one di-adduct, between 15 and 25 min. The two mono FAAF-modified oligos were separated and purified up to >97% purity by repeated injections on HPLC. The HPLC system consisted of a Hitachi EZChrom Elite HPLC unit with an L2450 diode array detector and a Phenomenex Luna C18 column (150×10 mm, 5.0 m). We employed a gradient system involving 3-15% Acetonitrile for 15 min followed by 15-30% acetonitrile for 25 min, respectively, in pH 7.0 ammonium acetate buffer (100 mM) with a flow rate of 2.0 ml/min. The FAAF-modified G 2 (5-CTCTCG 1 ATG 2 (FAAF)CCATCAC-3) was annealed with an appropriate complementary sequence (5'-GTGATGGCATCGAGAG-3') to form a fully paired duplex. Deletion duplexes (-3, -2, and -1) (Figure 4 a-d) and an identical set of unmodified control duplexes were similarly prepared.

Characterization of FAAF modified oligomers
FAAF-modified 16-mer oligos were characterized by initial enzyme digestions followed either by matrix assisted laser desorption ionization-time of flight (MALDI-TOF) or eletrospray ionization (ESI) LC/MS detection.

Enzymatic digestion/MALDI-TOF
A matrix solution was prepared by dissolving a 1:1 ratio of 3-hydroxypicolinic acid ( The digest solution was heated to 37 ℃ for the SVP digestion, but kept at room temperature for the BSP digestion. A 1 µL sample of the digest solution was removed at regular time intervals until the digestion rate was significantly reduced, and the reaction was quenched by mixing the aliquot with 1 µL of matrix (3-HPA and DHAC in 1:1 ratio). The sample was spotted on the MALDI plate and dried for immediate analysis. All MALDI-MS spectra were obtained using Shimadzu mass spectrometer in the Reflectronmode. Oligonucleotide molecular weight and detection limit determinations required 10 shots from the nitrogen laser (337 nm), whereas 150 laser shots were summed when the digest was analyzed.

Enzymatic digestion/ESI-LC/MS
We used electrospray ionization and quadrupole time-of-flight (TOF) mass spectrometry to verify the molecular weights and the position of FAAF attachment on the 16-mer oligomer. The ESI technique is used to produce intact highly charged molecular ions of oligo, as it can overcome the propensity of these molecules to fragment when ionized (31). For oligos with high molecular weight, this ionization results in a characteristic bellshaped distribution of multiply charged ions. Along with such a distribution, it is also accompanied by an adjacent major peak in the spectrum differing by one charge.
The FAAF-modified 16-mer oligomers were sequenced using 35 or 53 exonucleases as described previously for the analysis of modified 12-mers (31). Briefly, 1 µg of a 16-mer template was combined with 0.01 units of an exonuclease in a 1 mM solution of MgCl 2 and incubated for several hours. The digest was separated using a Phenomenex Aqua C18, 1.0×50 mm column (5 m; 120A˚). Solvent A was 5 mM in both ammonium acetate and dimethylbutyl amine. Acetic acid was added to solvent A to adjust the pH to 7.0. Solvent B was 0.1% formic acid in acetonitrile. The flow rate was 100µl/min and the total run time was 20 min. All LC/MS spectra were acquired using a Waters SYNAPT quadrupole time-of-flight mass spectrometer (Milford, MA, USA) operated in the negative ion and V-modes.

UV-melting
UV melting data were obtained using a Cary100 Bio UV/VIS spectrophotometer equipped with a 6×6 multi-cell block and 1.0 cm path length. Sample cell temperatures were controlled by an in-built Peltier temperature controller. Samples with a total concentration in the range of 0.5-10 µM were prepared in solutions containing 0.2 M NaCl, 10 mM sodium phosphate and 0.2 mM EDTA at pH 7.0. Thermomelting curves were constructed by varying temperatures of the sample cell (1 °C/min) and monitoring absorbances at 260 nm. A typical melting experiment consisted of forward/reverse scans and was repeated five times. Thermodynamic parameters were calculated using the program MELTWIN version 3.5, as described previously (32).

Circular Dichroism (CD)
CD measurements were conducted on a Jasco J-810 Spectropolarimeter equipped with a Peltier temperature controller. Typically, 6.5 µM of each strand were annealed with an equimolar amount of a complementary sequence. The samples were dissolved in 300 l of a pH 7.0 buffer (0.2 M NaCl, 10 mM sodium phosphate, 0.2 mM EDTA) and placed in a 1.0 mm path length cell. The samples were heated at 85 °C for 5 min and then cooled to 15 °C, over a 10 min period to ensure complete duplex formation. The spectropolarimeter was scanned from 200 to 400 nm at a rate of 50 nm/min. Spectra were acquired every 0.2 nm with a 2 s response time were the averages of 10 accumulations and were smoothed using 17-point adaptive smoothing algorithms provided by Jasco.

Dynamic 19 F Nuclear Magnetic Resonance ( 19 F-NMR) Spectroscopy
Approximately 20  MHz, respectively, using acquisition parameters described previously (33)(34)(35). 19 F-NMR spectra were acquired in the 1 H-decoupled mode and referenced relative to that of CFCl 3 by assigning external C 6 F 6 in C 6 D 6 at -164.9 ppm. One and two-dimensional 19 F-NMR spectra were measured between 5 and 78°C with increment of 5-10°C. Temperatures were maintained by a Bruker-VT unit with the aid of controlled boiling liquid N 2 in the probe. Computer line shape simulations were performed as described previously (36) using WINDNMR-Pro (version 7.1.6; J. Chem. Educ. Software Series; Reich, H.J., University of Wisconsin, Madison, WI, USA).

Differential Scanning Calorimetry (DSC)
Microcalorimetric measurements of all five FAAF-modified 16-mer duplexes were performed using a Nano-DSC from TA Instruments (Lindon, UT, USA). Prior to temperature scanning, samples were degassed for at least 10 min under vacuum in a closed vessel. Solutions were loaded, respectively, into the sample and reference cells using a pipette by attaching a small piece of silicone tube at the end of the tip and were purged several times to get rid of air bubbles. After both cells were filled, they were capped and a slight external pressure (~3 atm) was applied to prevent evaporation of the sample solution. Raw data were collected as microwatts versus temperature.  Chakrabarti et al. (37).

Matrix-Assisted Laser Desorption Ionization-Time of Flight (MALDI-TOF)
The molecular weights of the two FAAF mono-adducts (Peak 2 and Peak 3) were measured by MALDI-TOF prior to sequence verification by exonuclease digestions.
Exonucleases cleave terminal deoxynucleotides from the oligo chain until the FAAFmodified nucleotide is exposed at the end of the chain. At that point the digestion reaction slows down significantly. The position of modification is identified (in this case) when the fragment(s) formed by the loss of the unmodified guanine nucleotides is observed in the MALDI-TOF spectra. Peaks 2 and 3 were characterized as G 1 and G 2 , respectively.
a) 5ꞌ 3ꞌ exonuclease digestion of Peak 2: Figure 6 shows the 5ꞌ3ꞌ exonuclease digestion fragments for peak 2 at different time intervals. The ions observed at m/z 5016 in 6(a) represent the FAAF modified 16mer oligo (5'-CTCTCG 1 ATG 2 CCATCAC-3') before adding BSP enzyme. As shown in Figure 6 (b, c, d) increase in incubation time leads to the digestion of the subsequent unmodified bases. However, 5ꞌ-digestions was significantly slowed down at m/z 3,540.
The m/z 3540 ion at 60 min of 5ꞌ-digestion represents an ion formed from the FAAF d) 3ꞌ 5ꞌ exonuclease digestion of Peak 3: Figure 9 represents the 3ꞌ5ꞌ exonuclease digestion fragments for peak 3 at different time intervals. The ions observed at m/z 5,016 in 9(a) represent ions from the 16-mer FAAF modified 16mer oligo (5'-CTCTCG 1 ATG 2 CCATCAC-3') before adding SVP enzyme. As shown in Figure 9 (b, c, d) increase in incubation time leads to the digestion of the subsequent unmodified bases. The 3ꞌ-digestion was significantly slowed down at m/z 2,928 at 30 sec, 2 and 5 min. The ion m/z 2,928 of 3ꞌ-digestion represents ion formed from the FAAF modified 5'-CTCTCG 1 ATG 2 (FAAF)-3' fragment (see inset for theoretical MW values) suggesting the peak 3 is a G 2 position of the 16-mer template.

Electro Spray Ionization Quadruple Time of Flight (ESI-QTOF)-mass spectrometry
The molecular weights of FAAF-modified oilgos were measured by ESI-QTOF-MS prior to sequence verification by exonuclease digestion. Exonucleases cleave terminal deoxynucleotides from the oligo chain until the FAAF-modified nucleotide is exposed at the end of the chain. At that point the digestion reaction slows down significantly. This is shown in Figure 10(a) for the 5ꞌ3ꞌ digest of the 16-mer template 5ꞌ -CTCTCG 1 ATG 2 CCATCAC-3ꞌ. The ions observed at m/z 1178 is the (M-3H) 3ion from the fragment 5ꞌ-G 1 (FAAF)ATG 2 CCATCAC-3ꞌ. The observation of these ions confirms that peak 2 is modified on the G 1 closest to the 5ꞌ end. Figure 10  differences between the two could be due to difference in conformational characteristics. Figure 13(a-e) shows DSC plots of excess heat capacity C p ex vs. temperature for all five modified duplexes relative to their unmodified controls. Table 2  Kcal/mol) to produce an overall loss of free energy (ΔΔG 37°C =0.8 kcal/mol). As for the fully-paired duplex, 60% of S-conformations causes disturbance of Watson-crick base pairing, resulting in enthalpy reduction (ΔΔH = 13.9 Kcal/mol). Again, enthalpy-entropy compensation resulted in loss of free energy (ΔΔG 37°C = 3.5 kcal/mol) (39). The DSC thermograph in Figure 14 shows the differences in thermal and thermodynamic parameters of these deletion duplexes.

Differential Scanning Calorimetry (DSC)
As expected, both the "AT' and 'GC' -1 duplexes exhibited high stacked conformer population (72-73%) (see below 19 F NMR), thus displaying greater thermal and thermodynamic stability. Compared to 'AT', the 'GC' -1 deletion duplex displayed slightly higher ΔT m possibly due to the greater hydrogen bonding strength of G:C vs.
A:T. (Figure 17: c, d). Figure 15(a-e) shows the overlay CD spectra at 30 ℃ of all the duplexes in both unmodified and FAAF-modified. All duplexes exhibited a CD pattern characteristic for B-form DNA duplexes with positive and negative S-shaped ellipticity curve at around 275 and 250 nm, respectively. Modified duplexes displayed low intensity shoulders in the narrow 290-320 nm range, which is characteristic of FAAF modification (30,38,40).

Induced Circular Dichroism (ICD)
A slight increase in the positive intensity (hyperchromity) around 275 nm was observed for the FAAF modified (red) -2 deletion, 'AT' -1 deletion, 'GC'-1 deletion duplexes relative to their unmodified duplexes (blue) respectively. This may be due to adduct-induced stability caused by increase in stacking interactions in the bulged pocket.
However, the opposite (hypochromity) is true for the full and -3 deletion duplexes with the effect much greater for the latter. The difference in stability is likely due to the differences in conformational populations (inserted/stacked vs. external binding). These adduct-induced stabilization/destabilization was supported by the thermodynamic data described above.

CD Blue shifts:
In addition to the conformation specific ICDs above, the modified duplexes also displayed significant blue shifts relative to their respective unmodified duplexes ( Figure   15, Table 3). With the exception of -3 deletion duplex, all other modified duplexes exhibited significant blue shifts (up to 5 nm). The order of blue shift was in order of full duplex (Δ G*-G =5nm) > 'AT'-1 deletion ~ 'GC' -1 deletion (Δ G*-G =4nm) > -2 deletion (Δ G*-G =3nm). We previously reported that a FAAF-modified duplex in the CG*C context exists mostly (61%) in the stacked conformation. As a consequence, the greater CD blue shift in the full duplex could be due to the disturbance of Watson and Crick base pairing at the lesion site. These blue shifts could also indicate a lesion-induced distortion of DNA backbone contributing to bending (41). The next higher blue shifts seen in both 'AT' and 'GC' -1 deletion duplexes followed by -2 deletion duplex also could be based on the percentage of stacked conformations which are 72 and 73% in the 'AT' and 'GC' -1 duplexes respectively and 55% in the -2 deletion duplex.

CD Red shift:
As shown in Figure 15b, the FAAF modified -3 deletion duplex displayed a shift to longer wavelengths at 270 nm (Δ G-G*= 3nm). This red shift could have arisen due to a high percentage of external conformations (52%) compared to other deletion duplexes. Figure 16 shows an overlay of the CD spectra for the four FAAF modified (1 fulland -3 deletion) duplexes relative to fully unmodified duplex as a control. The idea was that unmodified deletion duplexes alone could not be used as a proper control. We observed similar order of blue shifts in all duplexes (Table 4). Here, -3 deletion duplex actually displayed a blue shift, not the red shift as in Figure 14b. The blue shift was in the order of full duplex (Δ G*-G =5nm) > 'AT' -1 deletion (Δ G*-G =4nm) > -2 deletion (Δ G*-G =3nm) ~ -3 deletion (Δ G*-G =3nm). A greater negative dip at 290-320nm range was seen for the full duplex followed by -1, -2 ~ -3 deletion duplexes.

Conformational Heterogeneity: 19 F NMR Signal Assignment
We previously reported 19 F NMR studies of fully paired FAAF-modified duplexes with various flanking sequence contexts (TG*A, CG*C, CG*G, GG*C) and in different length (11-and 16-mer). The results showed that FAAF-duplexes exist in a mixture of B, S and W conformations going from downfield to up field, i.e., -115.0 to -115.5 ppm, -115.5 to -117.0 ppm and -116.5 to -118.0 ppm, respectively (30). Using this general strategy, we assigned the two distinct signals at -115.6 and -117.9 ppm in the FAAFmodified fully paired 16-mer duplex at 5°C as 'S' and 'W' conformations (30). The 'AT' -1 deletion duplex exhibited a similar 19 F NMR pattern as the 'GC' one and was assigned accordingly, i.e., external B (-115.5 ppm) and inserted S (-116.5 ppm).
The signal at -116.1 ppm (marked in asterisk) is located in between S and B and coalesced with the stacked conformation above 40°C. This signal could be an intermediate conformer between B and S, but we were unable to pinpoint assignment.
Again, the greater stability of the 'GC' over 'AT' -1 duplex must be due to the better hydrogen-bonding capability of the 5'-G:C over A:T.
The -3 and -2 deletion duplexes also displayed two major 19 F signals in the -114 ~ -115 and -114.5 ~ -115.5 ppm range, respectively. However, these signals are collectively shifted to the downfield by about 1 ppm relative to those of -1 duplexes, suggesting an altered electronic environment. The major signals could be assigned as an external B (downfield) and stacked S (upfield) conformations according to their chemical shifts. However, we were unable to identify the additional signals marked as asterisk ( Figure 17). deletion site in an 11/10-mer duplex (25).

DISCUSSION
In this study, we investigated the role of conformational heterogeneity on thermal and thermodynamic stability of bulge duplexes modified with the prototype bulky arylamine FAAF. It is believed that the bulge structures are implicated in frameshift mutagenesis in the form of the so-called "slipped mutagenic intermediate (SMI)." A working hypothesis is that the conformational and thermodynamic stability of a SMI is a key factor for determining the extent of the frameshift mutations. We also hypothesized that deletion duplexes can exist in three major conformational motifs: external binding anti-glycosidic B, syn-glycosidic inserted stacked S, and others between B and S. To that end, we chose the three duplexes shown in Figure 2 as model systems for -1, -2, and -3 deletion mutations. As for -1 deletion, we studied two 5'-sequence context: AT vs. GC base pair. We conducted a variety of spectroscopic (CD, UV, NMR) and calorimetric (DSC) studies to obtain the conformational and structural information.
The 19 F-NMR results show that FAAF in the model duplexes exists in varying ratios of mixture of B and S conformers. A greater population of the bulge stacked conformation was observed in 'GC' (73%) and 'AT' (72%) -1 deletion duplexes than the -2 (55% S) and -3 (37% S) counterparts. As such, the highly stacked -1 bulged duplexes were greatly stabilized relative to -2 and -3 bulged duplexes, due to favorable stacking interactions between the intercalated fluorene and the bulge. These results were supported by increased positive CD at 275 nm, which indicated increased п-п stacking interactions between fluorene and DNA base pairs. We also observed a lesion induced blue CD shift, which is characteristic for lesion-induced DNA bending. 19 F NMR shows that the -3 deletion duplex exists in a mix of 55% B and 37% S conformers. Accordingly, the CD of the -3 duplex exhibited a red shift (Figure 15b were low compared to -1 bulged structure (72%). As a result, stacked conformation displays greater DNA bending in the full duplex, but promotes stability in the bulged structures.
As expected, FAAF in fully paired duplex resulted in thermal and thermodynamic destabilization relative to the unmodified control duplex. It has been documented that N-acetylated FAAF adducts in fully paired duplexes produce a mixture of complex S/B/W-conformers (30). The N-acetyl group is responsible for generating up to 40% W-conformers in the present 16-mer non NarI sequence (Figure 17a). We were unable to find the B conformation in the fully-paired duplex. The destabilizing effect of FAAF was related to the difference in the conformational populations. Thus, the presence of 60% S-conformations in full duplex promoted lesion stacking and disrupted the Watson-Crick base pairs, resulted in thermal (ΔT m = -8.0 °C) and enthalpy destabilization (ΔΔH = 13.9 Kcal/mol). Our previous studies of conformational and thermodynamic studies on NarI 16-mer full duplex also displayed 61% S-conformations, resulting in thermal (ΔT m = -8.3 °C) and enthalpy destabilization (ΔΔH = 24.7 Kcal/mol) (40). In contrast to the full duplex, the stacked conformer in bulged duplexes resulted in thermal and thermodynamic stabilization. Theoretical studies suggested that the synglycoside conformeric SMI is more stable than the external conformation and can be stabilized by many favorable interactions between a carcinogen and flanking base pairs inside the bulging pocket (19). In both 'GC'-1 deletion duplex and 'AT' -1 deletion duplexes, the highly stacked conformeric ( .

Biological implications:
Translesion DNA synthesis and subsequent elongation are most likely achieved by employing specialized lesion-bypass polymerases (15). Using gel based primer kinetics, Schorr et al have shown the mechanism of dG-C8-AAF induced frameshifts by the human DNA polymerase η. The lesion could base pair correctly with dC causing -1, -2 and -3 frameshifts depending upon the base pairs around the lesion ( Figure 22). The band representing the -1 frameshift appears to be more pronounced (Figure 22a), followed by the -2 frameshift which also showed several shorter sequences as well as the fully replicated 20-mer sequence. This is followed by -2 and -3 frameshift products (Figures 22 b and c). This is in accordance with our thermodynamic data where the -1 deletion duplex displayed high stability followed by -2 and -3 deletion duplexes. These results support the importance of lesion stacking which is directly correlated to mutational efficiencies.
Our working hypothesis is that the conformational and thermodynamic stability of adduct-induced bulge is a key determining factor for the propensity to form frameshift mutations. Fuchs et al studied the mechanism of AAF mutagenesis by DNA sequencing the spectrum of mutations in the E.coli lacI gene and found a greater frequency of -2 followed by -1 frameshifts, and few -3 deletions were observed (42). The 19 F NMR data showed that the AAF-modified opposite -1, -2 and -3 deletion duplexes have revealed 72%, 55% and 37% of stacked S conformation. We found that the -1 and -2 deletion duplexes are stacked better, having tight compactness of the lesion in the bulge, consequently displaying a greater stability. In the -3 deletion duplex, however, the lesion is not stacked well, resulting in more conformation flexibility. These results indicate that optimum space required to incorporate the AAF lesion is -1 followed by -2 bulge, in agreement with the extension assay data ( Figure 22).

F NMR Spectroscopy
The sensitivity of the 19 F nucleus to the macromolecular environment and the lack of background noise in dealing with complex conformations votes to opt this technique. 19 F NMR spectroscopy is an excellent tool to probe into the conformational heterogeneity of modified DNA and proteins. The fluorine technique was used in my thesis to analyze DNA-adduct conformation in various deletion DNA duplexes. 19 F act as a probe to determine the position of lesion in bulged duplexes. 19 F NMR spectra were acquired in the 1 H-decoupled mode and referenced relative to that of CFCl 3 by assigning external C 6 F 6 in C 6 D 6 at -164.9 ppm. 19 F NMR spectra were measured between 5 and 78°C with increment of 5-10°C. Temperatures were maintained by an FTS unit.
Usually 19 F-NMR spectra are obtained in a pH 7.0 aqueous NMR buffer in D 2 O.
For imino proton spectra, however, it is necessary to use a 10% D 2 O/90% H 2 O buffer, which prevents potential exchange of the imino protons involved in the Watson-Crick base pairs. Frequently, both 19 F-and imino NMR spectra are acquired at various temperatures in order to examine the adduct-induced heterogeneity around the lesion site and also to calculate the kinetic and thermodynamics parameters. Assignment of 19 F NMR signals is often difficult due to lack of related signals (i.e., NOE or scalar couplings, etc.) as well as conformational heterogeneity. However, general information could be obtained using chemical shift, isotope effect, dynamic NMR information as we have done previously (1,2). For example, dynamic 19 F NMR spectra of FAAF modified Guanine (G 1 ) in NarI 16mer sequence performed in our lab is shown below. Based on the chemical shifts at 5°C B, S, W conformations were assigned accordingly.

Circular Dichroism (CD) Spectroscopy
Circular Dichroism (CD) is a procedure that measures the ellipticities between the difference in the absorption of left and right handed circularly-polarized light. CD is sensitive to the changes occurring in chiral DNA and proteins. Chiral molecule characteristically produces a CD spectrum, depending on its spatial arrangement. Circular dichroism (CD) spectroscopy is a spectroscopic technique where the CD of molecules is measured over a range of wavelengths. CD spectroscopy is used extensively to study chiral molecules of all types and sizes. A primary use is in analyzing the secondary structure or conformation of macromolecules, particularly proteins as secondary structure is sensitive to its environment, temperature or pH, circular dichroism can be used to observe how secondary structure changes with environmental conditions or on interaction with other molecules. Structural, kinetic and thermodynamic information about macromolecules can be derived from circular dichroism spectroscopy (3).
Measurements carried out in the visible and ultra-violet region of the electro-magnetic spectrum monitor electronic transitions, and, if the molecule under study contains chiral chromophores then one CPL state will be absorbed to a greater extent than the other and the CD signal over the corresponding wavelengths will be non-zero. A circular dichroism signal can be positive or negative, depending on whether left-handed circularly polarized light (L-CPL) is absorbed to a greater extent than R-CPL (CD signal positive) or to a lesser extent (CD signal negative). An example circular dichroism spectrum of a sample with multiple CD peaks is shown below, demonstrating how CD varies as a function of wavelength, and that a CD spectrum may exhibit both positive and negative peaks. It is an ideal by carcinogen and the presence of bulges and mismatches can be studied effectively by using this technique. Induced Circular Dichroism (ICD) is a novel technique and has been used for understanding interactions between chromophores and DNA and protein molecules. CD can be used to study the adduct-induced conformational heterogeneity, in the near-UV absorption range providing information about the binding nature of the carcinogen moiety. The CD measures ellipticities (deg*cm 2 /dmol) as a function of wavelength (nm). Previously we carried out similar CD experiments in our lab on various carcinogens (AF and AAF modified duplexes), which gives us idea about the conformation of adduct-induced heterogeneity; disturbances such as distortion and disruption of Watson-Crick base pairs and bulge structures (2,4).
The figure shown below is a typical example for B-type DNA duplexes (Blue: unmodified; Red: Carcinogen modified) exhibiting CD ellipticity pattern characteristic for a S-shape curve at around 275 (+) and 250 (-) nm, respectively. An increase in the positive intensity around 275nm in modified duplexes indicate an increase interaction of the modified base in the bulged pocket and a decreased positive intensity represents a decreased interaction of the carcinogen in the bulged pocket with the neighboring base pairs. Negative ellipticity at 290-320 nm is an indication of FAAF modification. In addition, adduct-induced shift (blue, short wavelength and red, longer wavelength) at 275nm indicate adduct induced DNA bending (2). The main advantage of CD is it requires less sample and the sample can be used again as the instrument does not destroy the sample unlike NMR, X-ray techniques but the main disadvantages are it produce excess noise when high concentration of buffers are used or when there is any particulate matter that interfere with CD signals. It cannot give detail information about the structures as NMR and X-ray (5).