Critical role of DNA intercalation in enzyme-catalyzed nucleotide flipping

Nucleotide flipping is a common feature of DNA-modifying enzymes that allows access to target sites within duplex DNA. Structural studies have identified many intercalating amino acid side chains in a wide variety of enzymes, but the functional contribution of these intercalating residues is poorly understood. We used site-directed mutagenesis and transient kinetic approaches to dissect the energetic contribution of intercalation for human alkyladenine DNA glycosylase, an enzyme that initiates repair of alkylation damage. When AAG flips out a damaged nucleotide, the void in the duplex is filled by a conserved tyrosine (Y162). We find that tyrosine intercalation confers 140-fold stabilization of the extrahelical specific recognition complex, and that Y162 functions as a plug to slow the rate of unflipping by 6000-fold relative to the Y162A mutant. Surprisingly, mutation to the smaller alanine side chain increases the rate of nucleotide flipping by 50-fold relative to the wild-type enzyme. This provides evidence against the popular model that DNA intercalation accelerates nucleotide flipping. In the case of AAG, DNA intercalation contributes to the specific binding of a damaged nucleotide, but this enhanced specificity comes at the cost of reduced speed of nucleotide flipping.

S3 onto a 15% polyacrylamide gel. Gels were scanned with a Typhoon Imager (GE Trio+ Healthcare) to detect the fluorescein label by exciting at 488 nm and measuring emission with a 520BP40 filter. The gel bands were quantified using ImageQuant TL (GE Healthcare). The data were converted to fraction product [Fraction Product = product / (product + substrate)] and then fit by a single exponential using Kaleidagraph (Synergy Software) (Eq. 2). Single--turnover rates were independent of the concentration of AAG, indicating that the maximal rate constant was measured (k obs = k max ; Fig. S1).
(2) To measure competition between εA--DNA and undamaged DNA, single--turnover reactions containing 200 nM fluorescein--labeled εA DNA and 0-50 µM undamaged DNA, in which the εA was replaced by a normal A, were initiated with the addition of 300 nM AAG. Data for Y162A AAG was fit by an IC 50 equation (Eq. 3); WT and Y162F AAG were fit by a line with a slope of zero. (3) Stopped--Flow Kinetics. Pre--steady state kinetic experiments were performed on a Hi--Tech SF--61DSX2, controlled by Kinetic Studio (TgK Scientific). The fluorescence of εA was monitored using an excitation wavelength of 313 nm and a WG360 long--pass emission filter (3,4). At least three traces were averaged together at each concentration. We found that the fluorescence amplitudes varied from day to day with different instrument settings, but that the observed rate constants were highly reproducible. Therefore, we allowed the amplitudes to float in the curve fitting with Kaleidagraph. These amplitudes were not used in the data analysis. The traces for WT and Y162F AAG were fit by a triple exponential (Eq. 4) where F is the fluorescence as a function of time, C is the fluorescence of free DNA, X, Y, and Z are the changes in fluorescence of the intermediates, and t is the time. The traces for Y162A AAG were fit by a single exponential (Eq. 2) and k obs is equal to k flip + k unflip .
(4) Observed rate constants were plotted versus concentration and fit by a straight line. k 1,obs showed a linear concentration dependence, and the slope is equal to k on (M --1 s --1 ). Negative y-intercepts were observed under these conditions. Similar negative intercepts were reported for another DNA binding protein with high affinity for DNA (5). The value of k 2,obs was essentially independent of concentration and is equal to k flip + k unflip . It should be noted that in some cases the lowest concentrations gave slightly lower values of k 2,obs . The kinetic scheme predicts that a hyperbolic dependence would be observed at lower concentrations, as the initial recognition complex becomes saturated. However, there was not sufficient fluorescence signal to explore this region of the concentration dependence. Fitting of the data to hyperbolic fits gave essentially identical maximal values of k 2,obs , confirming that the initial recognition complex was saturated by the highest concentrations of DNA and protein that were tested (data not shown).
As an independent test of the kinetic parameters that were determined by combining association and dissociation kinetics, we fit the observed association kinetics obtained with either excess protein or excess DNA with the kinetic model using Berkeley Madonna (Berkeley Madonna, Inc.). The analysis with excess protein is given in Figure S3, according to the mechanism and script provided in Figure S5. Note that we must explicitly consider multiple bound proteins to explain the differences between excess protein and excess DNA. The analysis with excess DNA is given in Figure S4, according to the mechanism and script provided in Figure  S6.
Pulse--Chase Assay to Measure Substrate Dissociation. The macroscopic rate constant for dissociation of WT and mutant AAG from εA--containing DNA was measured by pulse--chase in the standard reaction buffer at 25 ⁰C as previously described for WT AAG (3,4). In 20 μL reactions, 50 nM fluorescein--labeled TEC DNA was mixed with 300 nM or 600 nM AAG for 20 seconds, and then a chase of 10 μM pyrrolidine--containing DNA was added. Pyrrolidine binds tightly to AAG as a transition state analogue (6,7). At various time points, a sample from the reaction was removed to evaluate the partitioning of a bound E•S complex between N-glycosidic bond hydrolysis and dissociation with the standard glycosylase assay. The fraction of product was fit by a single exponential (Eq. 2). Control reactions in which no chase was added provided the single--turnover rate constant, k max , and confirmed that these concentrations of AAG were saturating. From these values, the dissociation rate constant, k off , for AAG dissociating from εA--DNA was calculated as previously described. The rearranged equation is indicated by Eq. 5, in which A is the burst amplitude (the fraction of product formed in the burst phase of the experiment), k max is the single--turnover rate constant for formation of product, and k off,obs is the macroscopic rate constant for dissociation from the flipped--out complex.
Since the εA--DNA binds in two steps, the observed rate constant for dissociation of substrate (k off, obs ) could be limited by the unflipping rate (k unflip ) or dissociation from nonspecific DNA (k off ). According to Fig. 1C, and assuming that the flipped--out complex is stable (i.e., k flip >> k unflip ), this observed dissociation rate constant can be expressed in terms of the microscopic rate constants [Eq. 6; (8)]. Given rapid dissociation from nonspecific DNA, the observed rate of dissociation from the εA--DNA•AAG complex is approximately equal to the reverse rate constant for flipping (k off,obs ≈ k unflip ). flipped--out AAG--DNA complex, 2.8 µM AEA DNA was mixed with 2 µM Y162A. After an aging time of 1 second, 18 µM or 36 μM pyrrolidine--DNA was added as a competitor. The final concentrations after mixing were 700 nM AEA DNA, 500 nM Y162A, and 9 µM or 18 μM pyrrolidine--DNA. The reaction was followed for 10 seconds; no significant excision of εA occurs during this time. Three traces were averaged together and the change in fluorescence was fit by a single exponential (Eq. 2). As described for the pulse--chase assay, the observed rate constant for dissociation is equal to the rate constant for unflipping. Figure S1. Single--turnover excision of εA for WT and mutant AAG. Essentially identical rate constants were obtained for excision of εA from TEC DNA (50 nM) and either 300 nM or 600 nM AAG. These data indicate that 300 nM enzyme is saturating for all of the AAG variants tested.   Figure  S4.