DNA Polymerases η and ζ Combine to Bypass O²-[4-(3-Pyridyl)-4-oxobutyl]thymine, a DNA Adduct Formed from Tobacco Carcinogens

Abstract

4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and N′-nitrosonornicotine (NNN) are important human carcinogens in tobacco products. They are metabolized to produce a variety 4-(3-pyridyl)-4-oxobutyl (POB) DNA adducts including O²-[4-(3-pyridyl)-4-oxobut-1-yl]thymidine (O²-POB-dT), the most abundant POB adduct in NNK- and NNN-treated rodents. To evaluate the mutagenic properties of O²-POB-dT, we measured the rate of insertion of dNTPs opposite and extension past O²-POB-dT and O²-Me-dT by purified human DNA polymerases η, κ, ι, and yeast polymerase ζ in vitro. Under conditions of polymerase in excess, polymerase η was most effective at the insertion of dNTPs opposite O²-alkyl-dTs. The time courses were biphasic suggesting the formation of inactive DNA-polymerase complexes. The k_pol parameter was reduced approximately 100-fold in the presence of the adduct for pol η, κ, and ι. Pol η was the most reactive polymerase for the adducts due to a higher burst amplitude. For all three polymerases, the nucleotide preference was dATP > dTTP ≫ dGTP and dCTP. Yeast pol ζ was most effective in bypassing the adducts; the k_cat/K_m values were reduced only 3-fold in the presence of the adducts. The identity of the nucleotide opposite the O²-alkyl-dT did not significantly affect the ability of pol ζ to bypass the adducts. The data support a model in which pol η inserts ATP or dTTP opposite O²-POB-dT, and then, pol ζ extends past the adduct.

Graphical Abstract

INTRODUCTION

Lung cancer is the most common cause of cancer death in the US. Cigarette smoking is the main risk factor for lung cancer.¹ Tobacco specific nitrosamines (TSNAs) are a significant class of tobacco carcinogens.^2–4 N′-Nitrosonornicotine (NNN) and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) are the two most potent TSNAs in causing lung tumors in animals.^2–4 NNK and NNN are considered to be among the causative agents of lung and oral cavity cancers in individuals who use tobacco products.

Both NNN and NNK are metabolically activated to alkylating agents. NNK produces both a methyl and a 4-(3-pyridyl)-4-oxobutyl (POB) diazonium ion, while NNN produces the POB-diazonium ion.⁵ The methylation pathway, with the formation of the mutagenic O⁶-methyl-2′-deoxyguanosine, is an important contributor to mutagenicity and carcinogenicity of NNK.^6,7 The pyridyloxobutylation pathway, although less well studied, also plays a role in carcinogenesis in NNK and NNN. Four POB adducts have been characterized in rodents, in which the POB-group is bound to the N7- and O⁶-positions of dG and the O²-positions of dC and dT.^13,34–36 The chemical toxicology of O⁶-POB-dG has been the most studied. It is repaired by O⁶-alkylguanine-DNA alkyltransferase^8–10 and is mutagenic in bacteria and human cells causing mostly G to A transitions with some A to T transversions.^11,12 DNA polymerase η is the most reactive Y-family polymerase toward O⁶-POB-dG.¹³

O²-[4-(3-Pyridyl)-4-oxobut-1-yl]thymidine (O²-POB-dT) may also play a role in carcinogenesis. Its formation is shown in Scheme 1. In long-term studies, in which rats are continuously administered NNK or NNN, O²-POB-dT is the most abundant POB-adduct.^14–17 NNK-treated transgenic mice produce approximately equal amounts of mutations at G/C and A/T base pairs.^18,19 If the mutagenesis were solely due to the methyl diazonium ion pathway, then the mutagenic spectrum would be dominated by G/C to A/T mutations caused by O⁶-Me-dG, as observed in mice treated with N,N-dimethyl-N-nitrosamine.²⁰ Therefore, O²-POB-dT is a potential cause of the A/T mutations in NNK-treated rodents.

Scheme 1.

Formation of O²-POB-dT from the Bioactivation of NNK

DNA polymerases are crucial in maintaining genome integrity. Seventeen human DNA polymerases are involved in high fidelity DNA replication, DNA repair, and replication of DNA damage.^21–23 Polymerases α, γ, δ, ε, and telomerase are involved in the high fidelity replication of the genome. The Y-family polymerases η, κ, ι, and Rev1 as well as the B-family pol ζ are involved in translesion DNA synthesis (TLS).^21,24 Several other polymerases such as λ,^25–27 ν,²⁸ θ,²⁹ and PrimPol³⁰ may also be involved in TLS. A current model for replicating past DNA damage is the polymerase switch mechanism.^31–35 In this model, the high fidelity polymerase stalls at the DNA damage and is replaced by a TLS polymerase that inserts a dNTP opposite the damage. This polymerase or perhaps a different polymerase then extends past the damage.³⁵ The propensity of a specific DNA polymerase to bypass damage is thought to depend on the relative activity of the polymerase on the DNA substrate, protein–protein interactions, and post-translational modifications. The unique active sites of the polymerases have led to multiple polymerases combining to bypass the damage.³⁵ In many cases, the Y-family polymerase inserts a dNTP opposite the damage and pol ζ extends past the damage a few nucleotides so that the high fidelity polymerase can take over.

The earliest studies on the mutagenic potential of O²-methyl-dT (O²-Me-dT) and O²-ethyl-dT (O²-Et-dT)^36–38 were conducted with noneukaryotic polymerases. While O²-Me-dT forms the most stable base pair with dG,³⁶ T7 DNA polymerase and the Klenow fragment of proofreading deficient E. coli DNA polymerase I (Kf(exo-)) preferentially incorporate dATP and dTTP opposite O²-Et-dT.^37,38 We previously found that both Kf(exo-) and Sulfolobus solfataricus DNA polymerase IV (Dpo4) bypass both O²-Me-dT and O²-POB-dT with low efficiency and low fidelity, primarily inserting both dATP and dTTP opposite the adduct.³⁹ More recent in vitro replication studies showed that O²-Me-dT is highly blocking for DNA synthesis catalyzed by Kf(exo-), human pol κ, and Saccharomyces cerevisiae DNA polymerase η (yPol η).⁴⁰ Steady-state kinetic measurements revealed that Kf(exo-) and yPol η preferentially incorporated the correct nucleotide (dAMP) opposite O²-Me-dT.⁴⁰ However, using a LC-MS/MS to analyze both insertion and extension, hPol κ-mediated the insertion of dG opposite O²-Me-dT. Similar results were obtained with the larger O²-Et-dT, which was bypassed slowly with human DNA polymerases η, κ, and ι, yeast DNA polymerase ζ (yPol ζ), and Kf(exo-) to generate full-length replication products.⁴¹

The replication of O²-alkyl-dTs were evaluated in E. coli.^42,43 Bypass of O²-Me-dT and O²-POB-dT, in the absence of SOS activation, are low but accurate.⁴² When SOS is induced, both bypass efficiency and mutagenicity are increased. Pol II, IV, and V are responsible for bypass, with pol V as the most mutagenic. While dA is inserted opposite O²-Me-dT and O²-POB-dT most often, the major mutagenic process is the incorporation of dC opposite the adducts.⁴² More recently, the bypass of O²-Me-dT and O²-POB-dT were evaluated in mammalian cells.⁴⁴ Both dATP and dTTP were incorporated opposite the adducts, and pol η, ξ, and REV1 are the primary polymerases involved. To further investigate the mechanisms of mutagenicity of O²-POB-dT in humans, we evaluated the in vitro reactivity of O²-POB-dT in a defined oligodeoxynucleotide substrate with purified human DNA polymerases η, κ, and ι. We found that pol η has the highest reactivity for the incorporation opposite the adducts, incorporating both dATP and dTTP, and that yPol ζ is able to extend the primer past this base pair.

EXPERIMENTAL PROCEDURES

General

[³²P]ATP (6000 Ci/mmol) was purchased from PerkinElmer and T4 polynucleotide kinase from USB/Affymetrix. The dNTPs (ultrapure grade) were purchased from GE Healthcare, and the concentrations were determined by UV absorbance.⁴⁵ yPol ζ was purchased from Enzymax (Lexington, KT).

Oligodeoxynucleotides

Oligodeoxynucleotides containing O²-Me-dT and O²-POB-dT were synthesized, purified, and characterized as described.^42,46 The oligodeoxynucleotide containing 6-carboxyfluorescein was purchased from Integrated DNA Technologies (Coralville, Iowa). The sequences of the oligodeoxynucleotides are shown in Chart 1. The concentrations of oligodeoxynucleotides were determined from the absorbance at 260 nm, using the method of Borer⁴⁷ in which it was assumed that the spectroscopic properties of O²-Me-dT and O²-POB-dT were identical to those of the dT. The primer was ³²P-labeled with γ-[³²P]ATP and annealed with a 20% excess of the template.⁴⁸

Chart 1.

Oligodeoxynucleotide Sequences

Polymerase Purification and Concentration

Polymerases κ, η, and ι were purified from Sf9 insect cells as described.^49–51 The polymerases are full length proteins with an N-terminal His-tag. The active concentration of the enzymes were determined by evaluating the magnitude of the burst using an undamaged template (data not shown). The polymerases were reacted with 300 nM DNA (P15/T24G) and 100 μM dCTP and fit to eq 2. The amplitude was set as the polymerase concentration.

Polymerase Kinetics

Enzyme reactions were initiated by mixing equal volumes of the DNA (P15/T24)/polymerase solution in buffer with dNTP and Mg²⁺ in H₂O at 37 °C. The final buffer concentration was 40 mM Tris-HCl (pH 8.0), 3 mM DTT, 10 μg/mL BSA, and 2.5% glycerol. The reactions were quenched with equal volumes of STOP solution containing 10% 0.5MNa₂EDTA, 90% formamide, 0.025% (w/v) xylene cyanol, and 0.025% (w/v) bromophenol blue. Rapid reactions were performed on a RQF-3 (Kin-Tek Corporation) and quenched with 300 mM EDTA (pH 8).

The time course of the reaction was analyzed by denaturing PAGE, and the radioactivity on the gel was visualized with a Typhoon 9200. The progress of the reaction was quantitated by dividing the total radioactivity in the product band(s) by the radioactivity in the product and reactant bands. Multiple product bands appeared when the incorrect dNTP was added to the reaction.

Polymerase-DNA Binding Experiments

The binding affinity of the DNA to the polymerases was evaluated with fluorescence anisotropy. The DNA consisted of the primer strand (P15) modified on the 5′-end with 6-carboxyfluorescein and the template strand containing dT, O²-Me-dT, or O²-POB-dT. The fluorescence anisotropy was measured with 1 nM DNA and 0–100 nM polymerase at 37 °C. Anisotropy was measured with a Tecan Safire 2 plate reader with half-volume 96 well plates. The excitation and emission wavelengths were 471 and 525 nm, respectively, with an emission bandwidth of 20 nm.

Data Analysis

Data were fitted by nonlinear regression using the program Prism, version 5, for Windows (GraphPad Software, San Diego, California, USA, www.graphpad.com). The V_max and K_m values were determined by fitting the data to eq 1; v₀ is the initial rate, E₀ the enzyme, and N₀ the dNTP concentrations.

$$v_o=\frac{V_{max}{N}_0}{N_0+K_m}$$ (1)

The time courses for the insertion enzyme in excess reactions with O²-Me-dT and O²-POB-dT were fitted to eq 2 in which P is the product concentration, A the burst amplitude, k the burst rate constant, k_ss the steady-state rate constant, and t the time. The dNTP concentration dependence was quantified by fitting the burst equation parameters to the hyperbolic eq 1 in which v₀ is the value of the parameter at a specific [dNTP], and V_max is the maximum value of that parameter. The time courses for the enzyme in excess extension reactions with O²-Me-dT and O²-POB-dT were fitted to eq 3 in which P is the product concentration, A the amplitude, and k the first-order rate constant. The dNTP concentration dependence on the kinetic parameters in eqs 2 and 3 was quantified by fitting the parameters to the hyperbolic eq 1.

$$P=A(1-{\mathrm{e}}^{-kt})+k_{ss}t$$ (2)

$$P=A(1-{\mathrm{e}}^{-kt})$$ (3)

The K_d^DNA was determined by fitting the measured anisotropy (A) to eq 4, in which K is K_d^DNA, E₀ and D₀ (set to 1 nM) are the total concentrations of the polymerase and DNA, respectively, and A_max is the increase in anisotropy caused by the polymerase binding to the DNA.

$$A=\frac{A_{max}}{{[E]}_0}\frac{({[E]}_0+{[D]}_0+K)-\sqrt{{({[E]}_0+{[D]}_0+K)}^2-4{[E]}_0{[D]}_0}}{2}$$ (4)

RESULTS

Running Start Incorporation of dNTPs Opposite O²-alkyl-dT

The incorporation of dNTPs into a DNA substrate containing O²-alkyl-dT was examined with the running start analyses as illustrated in Figure 1. The lowest bands represent the 12-mer primer, while the top band is the 24-mer product. The left-most lanes show that templates containing dT are rapidly replicated by all three polymerases. With O²-alkyl-dT in the template, the gels display a band corresponding to a 15-mer, indicating that the polymerases stalled prior to the adduct. A more faint band just above the 15-mer indicates that extension past the adduct is also inefficient.

Figure 1.

PAGE analysis of running start polymerase reactions. DNA substrates (10 nM, P12/T24 with dT, O²-Me-dT, and O²-POB-dT) were reacted with pol κ (a), pol η(b), and pol ι (c) and all four dNTPs for the indicated time (min). The DNA polymerase concentrations were 0.1 nM for dT and 0.5 nM for O²-alkyl-dT substrates. The dNTP concentrations were 5 μM for dT and 50 μM for O²-alkyl-dT substrates.

DNA-Polymerase Binding to Oligodeoxynucleotides Containing O²-alkyl-dT

The binding affinities of the DNA substrates to the polymerases were measured by fluorescence anisotropy using a fluorescein-labeled primer strand. The increase in anisotropy was plotted against polymerase concentration and fitted to the quadratic equation as shown in Figure 2. The results, presented in Table 1, indicate that the modification, either methyl or POB, does not affect the affinity of the DNA to the polymerase.

Figure 2.

Polymerase-DNA affinity. Fluorescence anisotropy was determined with 1 nM DNA in which the 5′-terminus of the primer strand was modified with fluorescein with variable concentrations of (A) pol η, (B) pol ι, and (C) pol κ. The DNA contained dT (open circle), O²-Me-dT (black circle), O²–POB-dT (square). The data points are the mean ± standard deviation of three determinations. The lines are the best fit to eq 3.

Steady-State Incorporation of dNTPs Opposite O²-alkyl-dT

The Michaelis–Menten kinetic parameters were determined with polymerase concentration from 0.01 to 0.1 nM, 10 nM DNA, and 50 μM to 1 mM dNTP. The kinetic parameters are presented in Tables 2–4 for the different polymerases. For each enzyme, we also evaluated the correct incorporation of dATP opposite dT and the misincorporation of dTTP opposite dT. The relative k_cat/K_m values are visualized in Figure 2. Fluorescence anisotropy was determined with 1 nM DNA in which the 5′-terminus of the primer strand was modified with fluorescein with variable concentrations of (A) pol η, (B) pol ι, and (C) pol κ. The DNA contained dT (open circle), O²-Me-dT (black circle), and O²–POB-dT(squrate). The data points are the mean ± standard deviation of three determinations. The lines are the best fit to eq 3.

Table 2.

Steady-State Parameters for the Pol η Catalyzed Insertion Opposite O²-alkyl-dT^a

dNTP	template	kcat (min−1)	Km (μM)	kcat/Km (mM−1 min−1)
dATP	dT	49 ± 2	5.3 ± 0.3	9230 ± 230
dTTP	dT	0.140 ± 0.004	77.8 ± 5.5	1.8 ± 0.1
dATP	O2-Me-dT	0.068 ± 0.003	18.5 ± 2.4	3.7 ± 0.4
dCTP	O2-Me-dT	0.068 ± 0.003	130.2 ± 11.4	0.52 ± 0.02
dGTP	O2-Me-dT	0.039 ± 0.001	38.4 ± 1.8	1.05 ± 0.04
dTTP	O2-Me-dT	0.119 ± 0.004	166 ± 9	0.72 ± 0.02
dATP	O2-POB-dT	0.098 ± 0.004	21.7 ± 2.5	4.50 ± 0.37
dCTP	O2-POB-dT	0.141 ± 0.008	148.6 ± 16.2	0.95 ± 0.05
dGTP	O2-POB-dT	0.121 ± 0.004	126.2 ± 8.0	0.96 ± 0.03
dTTP	O2-POB-dT	0.126 ± 0.005	103.3 ± 8.6	1.22 ± 0.06

^aInitial rates were conducted with 0.25 to 0.5 nM pol, 5 nM DNA, and 0 to 200 μM dNTP. The values are the mean ± standard deviation of three determinations.

Table 4.

Steady-State Parameters for the Pol κ Catalyzed Insertion Opposite O²-alkyl-dT^a

dNTP	template	kcat (min−1)	Km(μM)	kcat/Km
dATP	T	4.3 ± 0. 2	1.4 ± 0.1	3200 ± 200
dTTP	T	0.11 ± 0.01	120 ± 20	1.0 ± 0.1
dATP	O2-Me-dT	0.087 ± 0.008	35 ± 8	2.5 ± 0.3
dCTP	O2-Me-dT	0.067 ± 0.006	160 ± 30	0.41 ± 0.03
dGTP	O2-Me-dT	0.055 ± 0.004	64 ± 12	0.86 ± 0.10
dTTP	O2-Me-dT	0.067 ± 0.001	34 ± 2	1.99 ± 0.07
dATP	O2-POB-dT	0.047 ± 0.001	33 ± 2	1.41 ± 0.05
dCTP	O2-POB-dT	0.022 ± 0.002	29 ± 7	0.81 ± 0.16
dGTP	O2-POB-dT	0.13 ± 0.01	66 ± 12	2.0 ± 0.2
dTTP	O2-POB-dT	0.14 ± 0.03	215 ± 79	0.64 ± 0.10

^aInitial rates were conducted with 0.25 to 0.5 nM pol, 5 nM DNA, and 0 to 200 μM dNTP. The values are the mean ± standard deviation of three determinations.

As shown in Figure 3, the incorporation of dATP opposite dT was >1000-fold faster than that of dTTP opposite dT for all three polymerases. The increase in k_cat/K_m was due to both k_cat and K_m effects. The k_cat/K_m values for the incorporation of the dNTPs opposite the O²-alkyl-dTs are similar to the misincorporation k_cat/K_m values. The selectivity for a particular dNTP is low. The greatest selectivity is for the incorporation of dATP opposite O²-Me-dT by pol η in which the k_cat/K_m for dATP is 7-, 3.5-, and 5-fold greater than that for dCTP, dGTP, and dTTP, respectively. This selectivity is due to K_m differences. A similar selectivity is found with O²-POB-dT in which the k_cat/K_m for the pol η catalyzed insertion of dATP is about 4-fold greater than that for the other dNTPs. Pol ι is the least selective polymerase with dATP, dGTP, and dTTP having similar k_cat/K_m values with dCTP having k_cat/K_m valuees 5-fold less. The k_cat/K_m values for each enzyme are also very similar. Upon the basis of the steady-state kinetic analysis, we would predict that O²-alkyl-dTs would be bypassed with low fidelity with pol η, κ, and ι.

Figure 3.

k_cat/K_m values ± standard errors for the incorporation of dNTPs opposite dT, O²-Me-dT, and O²–POB-dT catalazed by pol κ (black circle), ι (square), and η (open circle).

Incorporation of dNTPs Opposite O²-alkyl-dT with Polymerase in Excess

The TLS bypass of DNA adducts does not occur under steady-state kinetic conditions in the cell. The TLS polymerase would be brought to the DNA, perform an insertion reaction, and then hand the DNA off to a subsequent polymerase. This process can be kinetically driven by the relative reactivity of the polymerases or actively driven by protein–protein interactions.³¹ While k_cat/K_m values and consequently the selectivity of the polymerase under steady-state conditions is not affected by nonproductive binding complexes, the relative reactivity during single-turnover reaction will be. Consequently, we analyzed the single-nucleotide incorporation reactions with enzyme in excess conditions with polymerase concentrations of 250 nM and a DNA concentration of 25 nM. The DNA was P15/T24 in which the template base was dT, O²-Me-dT, or O²-POB-dT. The time courses for pol η catalyzed incorporation of each dATP opposite O²-POB-dT are shown in Figure 4A, with the early time points in Figure 4B. The data were fit to the burst equation, and the burst amplitude (A), burst rate constant (k), and steady-state rate constant (k_ss) were determined. Depending on the polymerase and the dNTP, the amplitude, burst rate constant, and/or steady-state rate constants were dependent on the dNTP concentration. If the parameters were dependent on the dNTP concentration, the parameters were fit to the hyperbolic eq 1. Figure 4C and D show the graphs for the amplitude and burst rate constant for pol η with each dNTP. The complete set of time course plots and the dNTP concentration dependence of the kinetic parameters for pol η, ι, and κ are presented in Figures S1–S12. The fitted parameters are presented in Tables 5–7. The parameters in the tables are the A^max, the maximum amplitude, K_A, and the dNTP concentration at half maximal amplitude, k_pol, the maximum burst rate constant, and K_d^NTP (app) the apparent dNTP dissociation constant, k_ss^max, the maximum k_ss, and K_ss, the dNTP concentration at half maximal k_ss^max. In some cases, the parameter did not exhibit a [dNTP] dependence, in which case the K value was not listed.

Figure 4.

Pol η catalyzed insertion of dNTPs opposite O^2–POB-dT. (A) Pol η (250 nM) and O²–POB-dT DNA (25 nM) were reacted with 10 μM (open circle), 30 μM (open square), 50 μM (square), 100 μM (up triangle), 200 μM (down triangle), 500 μM (diamond), and 1000 μM (circle) dATP. (B) Early time points from panel A. The lines are the best fit to the burst equation. The data points are the mean of three experiments ± standard deviation. dNTP concentration dependence on the (C) amplitude and burst rate constant (D) of the pol η catalyzed insertion of dNTPs opposite O²–POB-dT. The error bars are the standard errors. The lines (solid for dATP and dGTP, and dotted for dCTP and dTTP) are the best fit to eq 1 for dATP (black circle), dCTP (open square), dGTP (black square), and dTTP (open square).

Table 5.

Kinetic Parameters for the dNTP Dependence of the Pol κ Catalyzed Insertion Opposite O²-alkyl-dT^a

template	dNTP	amplitude			burst rate constant			steady-state rate constant
		Amax/[pol]	KA (μM)	Amax/[pol]/KA (mM−1)	kpol (s−1)	Kd dNTP (μM)	kpol/Kd(s−1 mM−1)	kssmax (s−1)	Kss (μM)
dT	dATP	0.69 ± 0.03	1.9 ± 0.4	363	155 ± 7	7.9 ± 0.8	19600	0.31 ± 0.03	8.0 ± 2.0
O2-Me-dT	dATP	0.72 ± 0.04	22 ± 3	33	7.4 ± 0.4	168 ± 20	44	0.023 ± 0.003	73 ± 41
O2-Me-dT	dCTP	0.22 ± 0.02	100 ± 29	2.2	3.5 ± 0.28	32 ± 15	109	0.014 ± 0.002	130 ± 70
O2-Me-dT	dGTP	0.25 ± 0.02	62 ± 17	4	3.9 ± 0.23	41 ± 12	95	0.013 ± 0.001	41 ± 13
O2-Me-dT	dTTP	0.72 ± 0.04	97 ± 12	7.4	2.6 ± 0.4	96 ± 60	27	0.020 ± 0.002
O2-POB-dT	dATP	0.68 ± 0.04	11 ± 2	62	5.8 ± 0.4	78 ± 16	74	0.026 ± 0.002	55 ± 17
O2-POB-dT	dCTP	0.18 ± 0.02	101 ± 29	1.8	3.89 ± 0.26	51 ± 16	76	0.010 ± 0.002	78 ± 45
O2-POB-dT	dGTP	0.20 ± 0.03	149 ± 67	1.3	3.35 ± 0.45	6 ± 15	555	0.011 ± 0.001	37 ± 14
O2-POB-dT	dTTP	0.72 ± 0.04	97 ± 12	7.4	2.6 ± 0.4	96 ± 60	27	0.020 ± 0.002

^aReactions were conducted with 250 nM pol, 25 nM DNA, and 0 to 1000 μM dNTP. The individual time courses were fit to eq 2 to obtain amplitude (A), burst rate constants (k_b), and steady-state rate constants (k_ss) as a function of dNTP concentration. These values were fit to the hyperbolic equation to obtain the parameters in the table. The values are the mean ± standard error of three determinations. If the K values are missing, then the parameter was not dependent on the [dNTP].

Table 7.

Kinetic Parameters for the dNTP Dependence of the Pol ζ Catalyzed Insertion Opposite O²-alkyl-dT^a

template	amplitude				burst rate constant			steady-state rate constant
	dNTP	Amax/[pol]	KA (μM)	Amax/[pol]/KA (mM−1)	kpol (s−1)	Kd dNTP (μM)	kpol/Kd (s−1 mM−1)	kssmax (s−1)	Kss (μM)
dT	dATP			0.3 ± 0.01	62 ± 2	8 ± 1	7750	0.86 ± 0.15	14 ± 5
O2-Me-dT	dATP	0.49 ± 0.11	340 ± 190	1.4	3.3 ± 0.5			0.033 ± 0.005	202 ± 84
O2-Me-dT	dCTP	0.18 ± 0.01	320 ± 27	0.6	3.2 ± 0.7			0.013 ± 0.003	438 ± 215
O2-Me-dT	dGTP	0.14 ± 0.015	195 ± 21	0.7	4.8 ± 0.6	210 ± 80	22	0.010 ± 0.001	152 ± 30
O2-Me-dT	dTTP	0.87 ± 0.47	1120 ± 960	0.7	3.2 ± 0.5			0.039 ± 0.006	364 ± 139
O2-POB-dT	dATP	0.29 ± 0.03	85 ± 37	3.4	3.3 ± 0.3	20 ± 7	165	0.029 ± 0.006	260 ± 150
O2-POB-dT	dCTP	0.25 ± 0.07	860 ± 410	0.3	2.9 ± 0.4	60 ± 40	48	0.014 ± 0.002	490 ± 170
O2-POB-dT	dGTP	0.80 ± 0.67	5200 ± 4800	0.2	2.3 ± 0.1	10 ± 4	230	0.011 ± 0.002	380 ± 200
O2-POB-dT	dTTP	0.51 ± 0.03	350 ± 60	1.5	2.8 ± 0.3			0.027 ± 0.004	110. ± 50

^aReactions were conducted with 150 nM pol, 15 nM DNA, and 0 to 1000 μM dNTP. The individual time courses were fit to eq 2 to obtainamplitude (A), burst rate constants (k_b), and steady-state rate constants (k_ss) as a function of dNTP concentration. These values were fit to the hyperbolic equation to obtain the parameters in the table. The values are the mean ± standard error of three determinations. If the K values are missing, then the parameter was not dependent on the [dNTP].

The relative reactivity of the polymerase/DNA/dNTP pairings are shown in the time course plots in Figure 5 in which the dNTP concentration was 50 μM. This dNTP concentration is shown because cellular dNTP concentrations typically range from 5 to 50 μM.⁵² The left panels have O²-Me-dT, and the right panels have O²-POB-dT as the template. Each panel also shows the correct incorporation of dATP opposite dT (black cricle). Several general observations can be made. First, the rate of incorporation of dATP opposite dT is 100–1000-fold faster than incorporation opposite the O²-alkyl-dTs. Second, pol η is the most reactive polymerase toward the alkylated substrates. Third, the incorporation of dATP and dTTP are faster than the incorporation of dCTP and dGTP opposite the O²-alkyl-dTs. These observations are discussed further below.

Figure 5.

Comparisons of the relative reactivity of polymerases with DNA and dNTP. The pol and O²-alkyl-dT DNA pairing in each panel is as follows: (A) pol η and O²-Me-dT; (B) pol η and O²-POB-dT; (C) pol ι and O²-Me-dT; (D) pol ι and O²-POB-dT; (E) pol ι and O²-Me-dT; (F) pol ι and O²-POB-dT. Each panel shows the insertion of 25 μM dATP opposite dT (black circle). The polymerase (250 nM) and DNA (25 nM) were reacted with 50 μM dATP (black square), dCTP (open diamond), dGTP (black diamond), dTTP (open square); dT, O²-Me-dT, or O²-POB-dT. The solid lines are the best fit to the burst equation with the purine dNTP represented by the solid line and the pyrimidine dNTP represented by the dashed line.

O²-alkyl-dT Are Poorer Substrates than dT

The decreased reactivity of the incorporation of dATP opposite the alkylated substrates is due to decreased k_pol and increased K_d^dNTP and K_A values. This is evident in Tables 5–7, in which the k_pol values for dATP decreases from 160, 72, and 62 s⁻¹ to less than 10 s⁻¹ with O²-alkyl-dT for all three enzymes. For pol η, the K_d^dNTP(app) increased ~10-fold, but for ι and κ, the K_d^dNTP(app) did not appreciably change. The A^max for pol η remained at ~0.7 for both dT and the O²-alkyl-dTs, while the K_A and K_d^dNTP increased 5–10-fold for the O²-alkyl-dTs.

Pol η Is More Reactive than Pol ι or κ with Both O²-Me-dT and O²-POB-dT

In Figure 5, the increased reactivity of pol η is evident by comparing the half-lives of the reactions. The half-lives of the pol η-catalyzed incorporation of dATP is ~1 s, while that for pol ι and κ are ~30 s. The increased reactivity of pol η is due to an increased burst amplitude for pol η relative to pols ι and κ. These conclusions are evident in Tables 5–7 in which the A^max is the highest and the K_A is the lowest for pol η when compared with those of pols ι and κ. For example, for dATP, the A^max values are 0.72 and 0.68, and K_A values are 25 and 14 μMfor O²-Me-dT and O²-POB-dT, respectively. The A^max values drop to between 0.24 and 0.49, and the K_A values rise to 90–340 μMfor pols ι and κ. In contrast, the k_pol and K_d^dNTP(app) parameters are very similar for each polymerase. For example, for pol η and O²-alkyl-dT, the k_pol ranges from 5.8 to 7.3 s⁻¹, and very similar values are observed for pol ι (3–3.7 s⁻¹) and pol κ (3 s⁻¹). The K_d^dNTP(app) values are also very similar: η (67–162 μM), ι (8–43 μM), and κ (8–11 μM).

dATP and dTTP Are the Most Reactive Nucleotides

The increased reactivity of dATP and dTTP over dCTP and dGTP is due primarily to the burst amplitudes. Examination of Table 5 for pol η shows that the A^max/[pol] for dATP and dTTP is ~0.7, while that for dCTP and dGTP is ~0.2. In contrast, the K_A, k_pol, and K_d^dNTP(app) parameters are quite similar for each dNTP.

Extension Past O²-alkyl-dT

We examined the ability of human DNA pols η, κ, and ι, and yeast pol ζ to extend past X/Y base pairs using the P16/T24 DNA substrates. Figure 6 shows polyacrylamide gels for the extension past dA/dT (left), dA/O²-Me-dT (middle), and dA/O²-POB-dT (right) by pols η, κ, ι, and ζ. In each panel, the lower band is the 16-mer starting material, and the upper bands are product bands. The polymerase concentrations were set so that the reactivity with the dA/dT DNA substrate was similar for each polymerase. As can be seen in the left panels, the polymerases react with the DNA almost to completion within 1 min. Also evident is the robust activity of yeast pol ζ (lower panels), while the human Y-family polymerases show little if any activity. These experiments were duplicated for pol ζ with dC, dG, and dT as the base pair partner for the O²-alkyl-dTs. Plots showing the quantification of these data are shown in the Supporting Information. These experiments indicate that pol ζ catalyzes the extension past the dN/O²-alkyl-dT more efficiently than the Y-family polymerases and that the extension is not dependent on the base pair partner or the alkyl chain.

Figure 6.

Relative bypass of O²-alkyl-dT by human (a–c) pol η, (d–f) pol κ, (g–i) pol ι, and (j–l) yeast pol ζ. The concentration of the next incoming dNTP was 50 μM, and the concentration of the DNA containing a dA/dT (left), dA/O²-Me-dT (middle), and dA/O²-POB-dT (left) base pair was 3 nM. The concentration of pol η was 0.05 nM, pol κ and ι were 0.1 nM, and pol ζ was 6 μg/μL.

To quantitate the relative reactivity of pol ζ in extending undamaged versus NNK-damaged base pairs, we performed steady-state kinetic analysis. The initial rate kinetics of the bypass of dN/O²-alkyl-dT were performed at various concentrations of the next correct dNTP. The data were fit to eq 1, and the resulting Michaelis–Menten parameters are presented in Table 8. The V_max/K_m parameters are summarized in Figure 7. Several observations are made. First, the presence of mispairs in the terminal base pair (X/Y) does not inhibit the insertion of the correct dNTP. This result is consistent with previous reports that pol ζ is effective at extending mispairs.⁵³ Second, the presence of the O²-alkyl-dT only inhibits replication by a factor of 3. Third, the identity of the alkyl chain, being methyl or POB, does not affect the V_max/K_m parameter. Finally, the base pair opposite the O²-alkyl-dT does not affect the rate of extension.

Table 8.

Steady-State Kinetic Parameters for Pol ζ Catalyzed the Extension Past O²-alkyl-dT^a

Y	Z	dNTP	X	Vmax nmol s−1 g−1	Km μM	Vmax/Km μ L/(s g)
A	T	A	dT	0.0674 ± 0.0067	20.9 ± 2.4	3.22 ± 0.16
A	T	C	dT	0.0472 ± 0.0047	35.7 ± 8.2	1.35 ± 0.21
A	T	G	dT	0.0674 ± 0.0040	25.2 ± 1.3	2.68 ± 0.28
A	T	T	dT	0.0678 ± 0.0090	23.4 ± 6.5	2.98 ± 0.40
T	A	A	O2-Me-dT	0.0421 ± 0.0078	38 ± 18	1.28 ± 0.52
T	A	C	O2-Me-dT	0.0411 ± 0.011	60 ± 31	0.73 ± 0.16
T	A	G	O2-Me-dT	0.0424 ± 0.00035	46 ± 14	0.96 ± 0.21
T	A	T	O2-Me-dT	0.0444 ± 0.0103	48 ± 26	1.03 ± 0.28
T	A	A	O2-POB-dT	0.0431 ± 0.0075	44 ± 16	1.02 ± 0.25
T	A	C	O2-POB-dT	0.0401 ± 0.0048	66 ± 16	0.62 ± 0.08
T	A	G	O2-POB-dT	0.0416 ± 0.0013	51 ± 11	0.83 ± 0.17
T	A	T	O2-POB-dT	0.0371 ± 0.0038	36 ± 11	1.06 ± 0.24

^aData are expressed as the mean ± SD of three independent experiments. The DNA substrate was P16/T24 as in Chart 1.

Figure 7.

Pol ζ catalyzed extension of O²-alkyl-dT base pairs. The relative V_max/K_m values for dA, dC, dG, and dT as at the primer terminus (X) with dT, O²-Me-dT, and O²-POB-dT as base pair partner (Y). The steady-state kinetics were performed with 30 nM DNA (P16/T24) and 1 μg/μL pol ζ. The data points are the mean ± SD of three independent determinations.

Since pol ζ is effective in extending dN/O²-POB-dT base pairs, we briefly examined if the enzyme was capable of inserting a nucleotide opposite the adducts. As is shown in Figure S18, the enzyme is very inefficient at inserting dATP opposite O²-alkyl-dT.

Extension Past O²-alkyl-dT with Polymerase in Excess

As discussed above, the steady-state kinetic parameters may not accurately reflect the activity of the Y-family polymerases during the incorporation of a single nucleotide at the replication fork. We therefore examined the extension past the adducts with enzyme in excess. We found that extension past the adducts was slower than the insertion opposite the adducts for the three Y-family polymerases studied. The reactions with 3 nM DNA, 30 nM pol η, and 50 μM dNTP are shown in Figure 8. The black triangles show the rapid insertion of dATP opposite dT, and the open triangles show the slower insertion of dATP opposite O²-alkyl-dT. The extension past the dN/O²-POB-dT base pairs are represented by the solid and open circles and squares. As is evident, the pol η catalyzed extension is much slower than the insertion reaction. The extension past the adducts by pols ι and κ are shown in Figure S19.

Figure 8.

Insertion and extension reactivity of pol η past O²-Me-dT (A) and O²-POB-dT (B). Pol η (30 nM) and DNA (3 nM) were reacted with 50 μM dNTP. Each panel shows the insertion of dATP opposite dT (triangle) and O²-alkyl-dT (open triangle) and the extension past dA (square), dC (open circle), dG (black circle), and dT (open square) opposite O²-alkyl-dT. The solid lines are the best fit to the burst equation, and the dashed lines are fit to a first-order equation.

DISCUSSION

NNK is a potent human lung carcinogen.⁴ It is unique among lung carcinogens in that irrespective of the route of administration, rodents given NNK develop lung tumors.⁵ Understanding the molecular mechanisms underlying its mutagenicity may shed some light on its remarkable organ-specificity. NNK is a bifunctional alkylating agent, producing both methyl and POB-DNA adducts. While similar sites on the DNA are alkylated, the relative proportions are different. In particular, O²-POB-dT is the most abundant POB-DNA adduct in rats chronically treated with NNK or NNN.^14,54–56 In contrast, the corresponding O²-Me-dT is a minor adduct from the corresponding methylating agent. Rodents given NNK have increased levels of mutations at AT base pairs when compared with rodents given a methylating agent,^19,20 supporting the role of NNK in mutagenesis at AT base pairs.

Recently, we found that siRNA knockdown of pol η, ζ, and Rev1 impacted the bypass of O²-Me-dT and O²-POB-dT in human cells.⁴⁴ These results, along with known reactivities of the enzymes, led to the hypothesis illustrated in Scheme 2 in which pol η is involved in the insertion of dNTPs opposite the adducts, pol ζ is involved in the extension past the adduct, while Rev1 is a structural protein. To test the hypotheses that catalytic activities of pols η and ζ are critical to bypass and mutagenesis of O²-POB-dT, we evaluated the in vitro kinetics of insertion of dNTPs opposite and bypass of O²-POB-dT by pols η, ι, κ, and ζ.

Scheme 2.

Model for the Bypass of O²-alkyll-dT

We initially examined the steady-state kinetics for the incorporation of each dNTP opposite both O²-Me-dT and O²-POB-dT. The kinetic parameters in Tables 2–4 show that pols η, κ, and ι are equally poor at insertion opposite the adducts. This result is inconsistent with a role for pol η in bypass. However, TLS polymerases do not operate under steady-state kinetics in vivo. A current model of TLS bypass is the handoff model in which polymerases are recruited for insertion opposite the adduct, while another polymerase can perform the extension past the adduct.^21,34 In this model, a polymerase acts once, and consequently, steady-state kinetics do not accurately describe the reactivity.

Under conditions of polymerase in excess, the time courses for the insertion of dNTPs opposite the adducts are biphasic with a rapid burst followed by a slower reaction. As illustrated in Figure 5, we show that pol η is more effective than pols ι and κ in the insertion of dNTPs opposite O²-POB-dT. All three enzyme are more effective than Kf(exo-) and Dpo4.³⁹ The biphasic kinetic behavior under conditions with polymerase in excess can be explained by two mechanisms: (1) formation of nonproductive enzyme–substrate complexes^39,57–61 and (2) a rate limiting step after rapid phosphodiester bond formation.^62,63 The formation of nonproductive complexes has been observed in DNA damage bypass.^39,57–61 In addition, pol κ exhibits this behavior during the formation of a correct base pair.⁶⁴

The substrate specificity of polymerases have been reported by their relative k_cat/K_m and presteady-state k_pol/K_d^dNTP values.^65,66 However, this relationship does not hold in this system. For example, as shown in Table 5, the k_pol/K_d value for pol η-catalyzed incorporation of dGTP (555 mM⁻¹ s⁻¹) is greater than that for dATP (74 mM⁻¹ s⁻¹). However, it is clearly evident in Figure 5B that the incorporation of dATP (solid square) is faster than the incorporation of dGTP (solid diamond). Therefore, the relative reactivity of the substrates is not simply the k_pol/K_d ratios but must take into account the amplitude and the steady-state parameters.

Pol κ and its E. coli analogue pol IV accurately bypass N²-dG adducts of various sizes with good efficiency.^51,67–70 For example, human pol κ bypasses the very bulky N²-benzopyrene adduct, while the prokaryotic analogue (pol IV/dinB) bypasses the less bulky N²-furfuryl-dG and N²-(1-carboxyethyl)-2′-dG adducts. The crystal structure of pol κ displays an open area on the minor groove of DNA.⁷¹ Thus, one can envision that pol κ could bypass O²-POB-dT by binding the POB-group in the minor groove pocket of pol κ, while dGTP would bind to dT in a Watson–Crick-like conformation in Figure 9d. However, we found that neither O²-POB-dT nor the smaller O²-Me-dT is a good substrate for pol κ. This result agrees with Andersen et al., who reported similar findings for O²-Me-dT and O²-Et-dT.^40,41

Figure 9.

Potential base pair structures.

The ability of pol κ to accommodate N²-alkyl-dG DNA damage is not a strictly steric issue. For example, the replacement of Phe13 by the smaller valine eliminates the ability of pol IV to bypass N²-furfuryl-dG.⁶⁹ This result appears to suggest that van der Waals interactions play a role in pol κ’s substrate specificity. In addition, functional/kinetic studies show that pol κ requires Watson–Crick hydrogen bonds for rapid catalysis.⁷² Adifference between N²-dG adducts and O²-POB-dT is that O²-POB-dT cannot form Watson–Crick hydrogen bonds with the incoming dNTP. While O²-POB-dT has a bulky adduct in the minor groove, perhaps the inability of O²-POB-dT to form Watson–Crick base pairs makes this a poor substrate for pol κ. This rationale was proposed to explain why pol IV is not involved in the bypass of O²-alkyl-dT adducts in E. coli.⁷³ In this article, we extend this rationale to human pol κ.

Pol ι flips template purines from the anti- to the syn-configuration thereby displaying the Hoogsteen hydrogen bonding face to the incoming dNTP.^74–77 This property allows for the replication of adducts in which the Watson–Crick hydrogen bonding face is blocked.⁷⁸ In addition, O⁶-Et-dG and N²-Et-dG adducts are bypassed in this manner.^79,80 However, pol ι is inefficient at bypassing more bulky N²-dG adducts.⁵⁰ Template pyrimidines react with pol ι in the anti-conformation. Kinetic evidence indicates that dGTP/dC is replicated with Watson–Crick base pairs, while dGTP/dT occurs via wobble base pairing.⁸¹ Upon the basis of these properties, there is no reason to expect that pol ι would be efficient at bypassing O²-POB-dT.

Pol η evolved to accurately bypass cis-syn cyclobutane pyrimidine dimers. Inactivation of the POLH gene leads to the XPV form of xeroderma pigmentosum in which individuals are predisposed to skin cancer.^82–84 Pol η is also proficient at bypassing other adducts such as N-(deoxyguanosin-8-yl)-1-aminopyrene,⁶¹ O⁶-alkyl-dG,¹³ and 8-oxo-dG.^85–88 Synthesis opposite the bulky N²-dG adducts is slow,⁶⁷ although pol η may play a role in the mutagenic bypass of N²-BP-dG.^68,89,90 These studies do not provide any mechanistic rationale for the ability of pol η to insert dTTP and dATP opposite O²-alkyl-dT. Our in vitro kinetic data do agree with the mammalian cell culture studies in which pol η is crucial to the bypass.⁴⁴

The mechanism underlying the pol η catalyzed insertion of dTTP and dATP opposite O²-alkyl-dT is not clear. The ability to form Watson–Crick-like structures has been proposed to explain the mutagenic potential of DNA adducts. This type of simplistic structural analysis worked well with O⁶-Me-dG in which the Watson–Crick-like structure (Figure 9a) initially proposed in 1976⁹¹ explains the mutagenic incorporation of dTTP.^92–94 However, an attempt to predict the base pairing of O²-alkyl-dT is not as satisfying. Figure 9b and c illustrates the potential hydrogen-bonding interactions between dA and dT with O²-alkyl-dT. While the dA/O²-alkyl-dT structure requires protonation, the alkyl group can also make a hydrophobic interaction with the 2-position of adenine stabilizing the structure.⁹⁵ Figure 9c displays a base-pair structure between dT and O²-alkyl-dT that is similar to the wobble structure that was observed in an oligodeoxynucleotide duplex containing a dT/dT mispair.⁹⁶ Figure 9b and c does not exhibit any chemical interactions that would explain the preference for pol η to preferentially insert dATP and dTTP opposite O²-alkyl-dT more often than dGTP. Figure 9d shows a potential Watson–Crick-like structure between dG and O²-alkyl-dT. This structure is very similar to that in Figure 9a, except that the alkyl group is in the minor groove. Perhaps this structure explains the mutagenic incorporation of dGTP opposite O²alkyl-dT by pol V in E. coli.^42,73 However, the preference for the incorporation of dT and dA opposite O²-alkyl-dG in humans cells is not evident. Structural and functional studies must be undertaken to elucidate the interactions that control the identity of the nucleotide inserted opposite O²-alkyl-dTs.

While pol η is the most efficient Y-family polymerase to incorporate dNTPs opposite O²-alkyl-dG, neither it nor pol ι or κ were effective at further extending the primer. We did find that yeast pol ζ is able to extend past the O²-alkyl-dT adducts. Pol ζ is a B-family polymerase that does not insert dNTPs opposite DNA damage but readily extends past mismatches,^97,98 abasic sites,⁹⁹ γ-hydroxy-1,N²-propano-dG,¹⁰⁰ 8-oxo-dG,¹⁰¹ O⁶-Me-dG,¹⁰¹ thymine glycol,¹⁰² and mismatched N²-BP-dG/T.^103,104 The polymerase we employed was the heterodimer between yeast Rev3 and Rev7. Human Rev3 is twice as large as the yeast protein.^105,106 Human and yeast Rev7 are homologous, and the activity of the catalytic subunit, Rev3, is increased 10-fold by Rev 7.¹⁰⁶ In vivo, the likely active species is a tetramer among Rev3, Rev7, and the pol δ subunits, pol31 and pol32.^107–110 We found that yeast pol ζ readily extends dN/O²-alkyl-dT base pairs, with k_cat/K_m values one-third that of undamaged DNA. There is no sequence specificity with respect to the base pair partner.

Our model for the bypass of O²-POB-dT is illustrated in Scheme 1. A replicative polymerase, such as pol δ, synthesizes up to the adduct. We found that the high fidelity polymerase Kf(exo-) can synthesize up to the adduct, but have very little activity at the insertion of a dNTP opposite the adduct.³⁹ The replicative polymerase stalls at the adduct and is replaced by a Y-family polymerase. Pol η has higher activity toward O²-POB-dT than κ or ι, and based upon relative kinetics, it will be the polymerase that inserts the correct dA or an incorrect dT opposite the adduct.^40,41 Pol η, as well as ι and κ, has low activity extending past O²-POB-dT. Following insertion, pol η will be replaced by pol ζ, which can extend the primer past the adduct. Subsequently, a high fidelity polymerase can continue DNA synthesis. The activity of pol ζ does not depend on the identity of the nucleotide that is inserted opposite the adduct. In our model, we have Rev1 acting as a scaffold protein. The involvement of Rev1 is supported from cell studies using siRNA to deplete polymerases. ⁴⁴ Rev1 is highly selective for inserting C opposite normal and adducted template G because it uses arginine as a template for dCTP binding.^111,112

Table 1.

Dissociation Constants for Normal and Damaged DNA to Polymerases^a

X	pol η (nM)	pol κ (nM)	pol ι (nM)
dT	7.9 ± 0.7	8.3 ± 0.7	8.4 ± 0.7
O2-Me-dT	10.1 ± 0.7	9.1 ± 0.5	9.7 ± 0.5
O2-POB-dT	8.9 ± 0.8	9.4 ± 0.9	9.7 ± 0.8

^aCalculated value with standard error. The fluorescence anisotropy was determined with 1 nM DNA with 0–100 nM polymerase in the appropriate buffer. The experiment was performed three times, and the data were fitted to the quadratic equation.

Table 3.

Steady-State Parameters for the Pol ι Catalyzed Insertion Opposite O²-alkyl-dT^a

dNTP	template	kcat (min−1)	Km (μM)	kcat/Km (mM−1 min−1)
dATP	dT	8.0 ± 0.6	2.6 ± 0.4	3070 ± 270
dTTP	dT	0.103 ± 0.003	70 ± 4	1.46 ± 0.06
dATP	O2-Me-dT	0.076 ± 0.002	24 ± 1	3.2 ± 0.1
dCTP	O2-Me-dT	0.061 ± 0.004	93 ± 14	0.65 ± 0.06
dGTP	O2-Me-dT	0.114 ± 0.002	46 ± 2	2.5 ± 0.1
dTTP	O2-Me-dT	0.121 ± 0.005	42 ± 5	2.9 ± 0.2
dATP	O2-POB-dT	0.086 ± 0.003	35 ± 3	2.4 ± 0.1
dCTP	O2-POB-dT	0.059 ± 0.003	112 ± 13	0.53 ± 0.03
dGTP	O2-POB-dT	0.094 ± 0.003	28 ± 3	3.3 ± 0.2
dTTP	O2-POB-dT	0.121 ± 0.001	37 ± 1	3.27 ± 0.08

^aInitial rates were conducted with 0.25 to 0.5 nM pol, 5 nM DNA, and 0 to 200 μM dNTP. The values are the mean ± standard deviation of three determinations.

Table 6.

Kinetic Parameters for the dNTP Dependence of the Pol ι Catalyzed Insertion Opposite O²-alkyl-dT^a

template	dNTP	amplitude			burst rate constant			steady-state rate constant
		Amax/[pol]	KA (μM)	Amax/[pol]/KA (mM−1)	kpol (s−1)	Kd dNTP (μM)	kpol/Kd (s−1 mM−1)	kssmax (s−1)	Kss (μM)
dT	dATP			3.0 ± 0.4	22 ± 6			0.25 ± 0.03	10 ± 5
O2-Me-dT	dATP	0.420 ± 0.047	190 ± 60	8.9	3.3 ± 0.3			0.027 ± 0.267	78 ± 37
O2-Me-dT	dCTP	0.180 ± 0.020	601 ± 130	0.3	2.3 ± 0.6			0.013 ± 0.003	250 ± 160
O2-Me-dT	dGTP	0.093 ± 0.007	213 ± 56	0.4	5.5 ± 1.1	217 ± 121	25	0.017 ± 0.006	592 ± 426
O2-Me-dT	dTTP	0.380 ± 0.033	104 ± 34	3.6	3.6 ± 0.2			0.023 ± 0.003	110 ± 40
O2-POB-dT	dATP	0.313 ± 0.020	90 ± 20	3.5	3.7 ± 0.2	43 ± 12	86	0.017 ± 0.001	80 ± 20
O2-POB-dT	dCTP	0.187 ± 0.013	300 ± 53	0.6	2.53 ± 0.05	27 ± 3	94	0.004 ± 0.001	8 ± 15
O2-POB-dT	dGTP	0.147 ± 0.013	421 ± 105	0.3	2.42 ± 0.24	45 ± 22	54	0.007 ± 0.001	186 ± 63
O2-POB-dT	dTTP	0.307 ± 0.033	82 ± 31	3.7	3.7 ± 0.3	42 ± 18	88	0.024 ± 0.004	170 ± 90

^aReactions were conducted with 150 nM pol, 15 nM DNA, and 0 to 1000 μM dNTP. The individual time courses were fit to eq 2 to obtain amplitude (A), burst rate constants (k_b), and steady-state rate constants (k_ss) as a function of dNTP concentration. These values were fit to the hyperbolic equation to obtain the parameters in the table. The values are the mean ± standard error of three determinations. If the K values are missing, then the parameter was not dependent on the [dNTP].

ABBREVIATIONS

7-POB-dG

7-[4-(3-pyridyl)-4-oxobut-1-yl]-2′-deoxyguanosine

Dpo4

Sulfolobus solfataricus DNA polymerase IV

Kf(exo-)

Klenow fragment of proofreading deficient E. coli DNA polymerase I

NNK

4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone

NNN

N′-nitrosonornicotine

O²-Me-dT

O²-methylthymidine

O²-POB-dC

O²-[4-(3-pyridyl)-4-oxobut-1-yl]-2′-deoxycytidine

O²-POB-dT

O²-[4-(3-pyridyl)-4-oxobut-1-yl]-thymidine

O⁶-POB-dG

O⁶-[4-(3-pyridyl)-4-oxobut-1-yl]-2′-deoxyguanosine

POB

4-(3-pyridyl)-4-oxobut-1-yl

pol

polymerase

yPol

Saccharomyces cerevisiae DNA polymerase