Saccharomyces cerevisiae replication protein A binds to single-stranded DNA in multiple salt-dependent modes

Abstract

We have examined the single-stranded DNA (ssDNA) binding properties of the Saccharomyces cerevisiae replication protein A (scRPA) using fluorescence titrations, isothermal titration calorimetry, and sedimentation equilibrium to determine whether scRPA can bind to ssDNA in multiple binding modes. We measured the occluded site size for scRPA binding poly(dT), as well as the stoichiometry, equilibrium binding constants, and binding enthalpy of scRPA-(dT)L complexes as a function of the oligodeoxynucleotide length, L. Sedimentation equilibrium studies show that scRPA is a stable heterotrimer over the range of [NaCl] examined (0.02-1.5 M). However, the occluded site size, n, undergoes a salt-dependent transition between values of n = 18-20 nucleotides at low [NaCl] and values of n = 26-28 nucleotides at high [NaCl], with a transition midpoint near 0.36 M NaCl (25.0 degrees C, pH 8.1). Measurements of the stoichiometry of scRPA-(dT)L complexes also show a [NaCl]-dependent change in stoichiometry consistent with the observed change in the occluded site size. Measurements of the deltaH(obsd) for scRPA binding to (dT)L at 1.5 M NaCl yield a contact site size of 28 nucleotides, similar to the occluded site size determined at this [NaCl]. Altogether, these data support a model in which scRPA can bind to ssDNA in at least two binding modes, a low site size mode (n = 18 +/- 1 nucleotides), stabilized at low [NaCl], in which only three of its oligonucleotide/oligosaccharide binding folds (OB-folds) are used, and a higher site size mode (n = 27 +/- 1 nucleotides), stabilized at higher [NaCl], which uses four of its OB-folds. No evidence for highly cooperative binding of scRPA to ssDNA was found under any conditions examined. Thus, scRPA shows some behavior similar to that of the E. coli SSB homotetramer, which also shows binding mode transitions, but some significant differences also exist.

Figure 1

Schematic representation of scRPA hetero-trimer with its six OB-folds. Four OB-folds are contained within RPA70 (OB-folds, A, B, C and F), one within RPA32 (OB-fold D), and one within RPA14 (OB-fold E).

Figure 2. Stability of the scRPA heterotrimer as a function of [NaCl]

(A) - sedimentation equilibrium results at 20mM NaCl (buffer T, 25°C) for two scRPA concentrations (0.7 μM and 1.4 μM) and three rotor speeds of 10 krpm (red), 12 krpm (green), and 14 krpm (magenta). The solid curves represent the best global fits to a single species model with weight average molecular mass, M= 115 ± 4.7 kDa. (B) - sedimentation equilibrium results at 1.5 M NaCl (buffer T, 25°C) for three scRPA concentrations (2.4 μM, 1.2 μM, 0.6 μM) and three rotor speeds 10 krpm (red), 14 krpm (olive), 17 krpm (magenta). The solid curves represent the best global fits to a single species model with weight average molecular mass, M= 122 ± 3.8 kDa. The residuals for the fits are also shown below each data set.

Figure 3. Salt dependence of occluded site size for scRPA binding to poly(dT)

Results of titrations of scRPA with poly(dT) (buffer T, pH 8.1, 25°C) monitoring the quenching of scRPA tryptophan fluorescence (λ_excit= 295 nm and λ_emis =345 nm) at (A) 20 mM NaCl, and (B) 1.5 M NaCl. Each panel shows the results of two titrations obtained at two scRPA concentrations: (●) - 0.1 μM, and (○) - 0.2 μM. The occluded site size (n_app) was determined from the intersection of the two lines as indicated and is 19 ± 1 nucleotides at 20 mM NaCl and 27 ± 2 nucleotides at 1.5 M NaCl. (C) – The occluded site size (n_app) for scRPA binding to poly(dT) plotted as a function of [NaCl] (buffer T, pH 8.1, 25.0 °C). The solid line simulates the transition between a low salt binding mode with n = (18 ± 1) nucleotides and a high salt binding mode with n = (27 ± 2) nucleotides with a transition at ∼ 0.36 M NaCl.

Figure 4. Stoichiometry of scRPA binding to (dT)₃₀ is influenced by [NaCl]

Results of ITC titrations of scRPA with (dT)₃₀, plotted as normalized heat per mole of injectant, (panels A, B and D) and of (dT)₃₀ with scRPA (panel C). (A) - titration of scRPA (0.72 μM) with (dT)₃₀ (6.4 μM) at 1.5 M NaCl (buffer T, pH 8.1, 25°C). The solid line represents the best fit of the data to a single site binding model (eq 7) yielding: K=(1.21±0.14)×10⁸ M⁻¹ and ΔH_obs = −46.2±0.5 kcal/mol. (B) - titration of scRPA (0.6 μM) with (dT)₃₀ (6.2 μM) at 20 mM NaCl (buffer T, pH 8.1, 25°C). The solid line represents the best fit to a two site binding model (eq 8), yielding: K₁=(1.6 ± 0.8)×10¹⁰ M⁻¹; ΔH₁=−56.5±0.5 kcal/mol and K₂=(2.4±1.5)×10⁶ M⁻¹; ΔH₂=−32.7±10.0 kcal/mol. (C) - titration of (dT)₃₀ (0.43 μM) with scRPA (15.1 μM) at 0.20 M NaCl (buffer T, pH 8.1, 10% (v/v) glycerol, 25°C). The solid line represents the best fit to a two site binding model (eq 8), yielding: K₁=(9.3 ± 5.6)×10⁸ M⁻¹; ΔH₁=−55.4±1.1 kcal/mol and K₂=(6.0±5.2)×10⁵ M⁻¹; ΔH₂=−37.1±19.0 kcal/mol. (D) - titration of scRPA (1.1 μM) with (dT)₃₀ (21.1 μM) performed in the same buffer condition as in (C). The solid line represents the best fit to a two site binding model (eq 8) yielding: K₁=(8.9±4.5)×10⁸ M⁻¹; ΔH₁=−51.6±1.0 kcal/mol and K₂=(5.3±5.5)×10⁵ M⁻¹; ΔH₂=−30.2±22.0 kcal/mol.

Figure 5. Stoichiometries of scRPA-5′Cy3-(dT)₂₉ complexes examined by sedimentation equilibrium

(A) – Results of experiments performed at a 1:1 molar ratio of scRPA (1.27 μM) to 5′Cy3-(dT)₂₉ (1.27 μM) at 20 mM NaCl (buffer T, pH 8.1, 25°C) at three rotor speeds 10 krpm (green), 13 krpm (magenta), 17 krpm (dark yellow). These data fit well to a single species model (eq 1), yielding a molecular mass, M = 113± 1 kDa, which is close to that expected for a 1:1 complex (123 kDa). (B) – Results of experiments performed at a four-fold molar excess of scRPA (2.4 μM) over 5′Cy3-(dT)₂₉ (0.6 μM) at 20 mM NaCl (buffer T, pH 8.1, 25°C) at three rotor speeds 10 krpm (green), 13 krpm (magenta), 17 krpm (dark yellow). The data fit well to a single species model (eq 1) yielding a weight average molecular mass, M = 222.8 ± 2.0 kDa, close to the expected value of 218 kDa for a complex of two scRPA bound to one 5′Cy3-(dT)₂₉. (C) – Results of experiments performed at a four fold molar excess of scRPA (2.4 μM) over 5′Cy3-(dT)₂₉ (0.6μM) in 1.5 M NaCl (buffer T, pH 8.1, 25°C) at three rotor speeds 10 krpm (green), 13 krpm (magenta), 17 krpm (dark yellow). The data fit well to a single species model (eq 1) yielding a molecular mass, M = 127± 3 kDa, which is close to that expected for a 1:1 complex (123 kDa).

Figure 6. Fluorescence Binding isotherms show that 1:1 complexes of scRPA with (dT)₂₈ and (dT)₂₀ are formed at 1.5 M NaCl

(A)- Results of equilibrium fluorescence titrations of scRPA with (dT)₂₈ at two scRPA concentrations: (●) 6.0 × 10⁻⁸ M; (□) 1.2 × 10⁻⁷ M (buffer T, pH 8.1, 25°C, 1.5 M NaCl), monitoring the quenching of the scRPA tryptophan fluorescence. Global fitting of the two titrations to a single site binding model (eq 5) yields K_obs = 6.72±0.42 × 10⁷ M⁻¹, and Q_max = 0.38 ± 0.01. The solid lines are simulations based on eq 5 and the best fit parameters. (B)- Analysis of the data in panel (A) using the model-independent binding density function method shows that the fluorescence quenching of scRPA is directly proportional to the average number of (dT)₂₈ bound per scRPA heterotrimer. Linear extrapolation of the data to the known maximum fluorescence quenching at saturation yields a stoichiometry of 1.0 (dT)₂₈/scRPA. (C)- Results of equilibrium fluorescence titrations of scRPA with (dT)₂₀ at three scRPA concentrations: (□) 3.0 × 10⁻⁷ M (●), 1.0 × 10⁻⁷ M; (○) 0.5 × 10⁻⁷ M (buffer T, pH 8.1, 25°C, 1.5 M NaCl). Global fitting of the two titrations to a single site binding model (eq 5) yields K_obs = 4.02±0.21 × 10⁷ M⁻¹, and Q_max = 0.31 ± 0.01. The solid lines are simulations based on eq 5 and the best fit parameters. (D) - Dependence of scRPA fluorescence quenching on the average number of (dT)₂₀ bound per scRPA, obtained from analysis of the data in panel (C) using the binding density function method. The solid line shows a linear extrapolation of the data to the known maximum fluorescence quenching at saturation yields a stoichiometry of 1.0 (dT)₂₀/scRPA.

Figure 7. Equilibrium binding isotherms show that two molecules of (dT)₁₈ can bind per scRPA at high salt

(A) – Equilibrium titrations monitoring the scRPA tryptophan fluorescence quenching upon binding (dT)₁₈ obtained at two scRPA concentrations: (○)1.0 × 10⁻⁷ M (●) 3.6 × 10⁻⁷ M at 1.5 M NaCl (buffer T, pH 8.1, 25°C). The continuous lines represent simulations based on a model for two molecules of (dT)₁₈ binding per scRPA (eq 6). A global NLLS fit of both titrations yields K_1,obs = 2.08±0.22 × 10⁷ M⁻¹ and K_2,obs = 1.29±0.26 × 10⁵ M⁻¹ for the macroscopic binding constants for binding of the first and the second (dT)₁₈ to scRPA. (B) - Dependence of relative tryptophan fluorescence quenching of scRPA on the average number of (dT)₁₈ molecules bound per scRPA obtained from analysis of the titrations in panel (A). The relationship for the second phase of the titrations could not be obtained due to the weak binding affinity. However, a linear extrapolation from the points at [(dT)₁₈]/[scRPA] = 1 to the known maximum fluorescence quenching at saturation (Q_max = 0.33 ± 0.01) yields a maximum stoichiometrry of 2 moles of (dT)₁₈ bound per scRPA. (C) - Dependence of the maximum stoichiometry (maximum number of (dT)_L molecules that can bind per scRPA) as a function of the ss-oligodeoxythymidylate length, L (1.5 NaCl, buffer T, pH 8.1, 25° C), showing a clear transition between L = 18 and 20 nucleotides.

Figure 8. Determination of the scRPA contact size at 1.5 M NaCl

(A) – Results of an ITC titration of scRPA (0.24 μM in the cell) with (dT)₂₈ (5.0 μM in the syringe) (buffer T, pH 8.1, 1.5 M NaCl, 25°C). (B) - integrated heat responses from data in (A) are plotted as injection heats (kcal) per mole of injected DNA. The smooth line represents the best fit to a 1:1 binding model (eq 7) with parameters K_obs = 4.8 × 10⁷ ± 0.5 and ΔH_obs=−47±1 kcal/mol. (C) - Dependence of ΔH_obs on ssDNA length, L, determined for scRPA binding to a series of oligodeoxynucleotides, (dT)_L (buffer T, pH 8.1, 1.5 M NaCl, 25°C). A contact size of m=28 nucleotides is estimated from the intercept of the line describing the linear dependence of ΔH_obs on L, with the plateau value of ΔH_obs = −46 ± 2 kcal/mol at L ≥30. (D) - Dependence of the observed equilibrium binding constant, K_obs, on ss-oligodeoxynucleotide length, L. (∎) - data from ITC titrations, (○) - data from fluorescence titrations. A contact size of m=27−28 nucleotides is determined from the intercept of a linear extrapolation of the data for L ≥ 28 nucleotides to K_obs = 0.

Figure 9

Dependence on NaCl and NaBr concentrations of the equilibrium constant for scRPA binding to (dT)₂₆. Equilibrium binding was monitored by the quenching of scRPA (50−90 nM) tryptophan fluorescence upon binding (dT)₂₆ (buffer T, pH 8.1, 25°C). (A) - Binding isotherms obtained as a function of [NaCl]: 0.90 M (●); 1.10 M (◆); 1.30 M (◇); 1.50 M (○) and 1.70 M (▽). Solid lines represent best fits of the data to a single site binding model (eq 5). (B) - Binding isotherms obtained as a function of [NaBr]: 0.50 M (●); 0.60 M (○); 0.70 M (△), and 0.80 M (▲).Solid lines represent best fits of the data to a single site binding model (eq 5). (C) - Dependence of K_obs (obtained from analysis of the data in panels A and B) on salt concentration (log-log plots); ∂log K/∂log[NaCl] = −3.42 ± 0.07, and ∂log K/∂[NaBr] = −3.87 ± 0.15.

Figure 10

Schematic representation showing that the two scRPA modes of binding to long ssDNA differ in the number of OB-folds that interact with the ssDNA. (A) - low salt binding mode in which scRPA uses only three of its OB-folds (A, B, C) contained within the RPA70 subunit to bind ssDNA with an occluded site size of 18−20 nucleotides. (B) - high salt binding mode in which scRPA uses the three OB-folds within the RPA70 subunit plus an additional OB-fold (D) from the RPA32 subunit to bind ssDNA with an occluded site size of 26−28 nucleotides.