A nucleotide-gated molecular pore selects sulfotransferase substrates

Abstract

Human SULT2A1 is one of two predominant sulfotransferases in liver and catalyzes transfer of the sulfuryl moiety (-SO(3)) from activated sulfate (PAPS, 3'-phosphoadenosine 5-phosphosulfate) to hundreds of acceptors (metabolites and xenobiotics). Sulfation recodes the biologic activity of acceptors by altering their receptor interactions. The molecular basis on which these enzymes select and sulfonate specific acceptors from complex mixtures of competitors in vivo is a long-standing issue in the SULT field. Raloxifene, a synthetic steroid used in the prevention of osteoporosis, and dehydroepiandrosterone (DHEA), a ubiquitous steroid precusor, are reported to be sulfated efficiently by SULT2A1 in vitro, yet unlike DHEA, raloxifene is not sulfated in vivo. This selectivity was explored in initial rate and equilibrium binding studies that demonstrate pronounced binding antisynergy (21-fold) between PAPS and raloxifene, but not DHEA. Analysis of crystal structures suggests that PAP binding restricts access to the acceptor-binding pocket by restructuring a nine-residue segment of the pocket edge that constricts the active site opening, or "pore", that sieves substrates on the basis of their geometries. In silico docking predicts that raloxifene, which is considerably larger than DHEA, can bind only to the unliganded (open) enzyme, whereas DHEA binds both the open and closed forms. The predictions of these structures with regard to substrate binding are tested using equilibrium and pre-steady-state ligand binding studies, and the results confirm that a nucleotide-driven isomerization controls access to the acceptor-binding pocket and plays an important role in substrate selection by SULT2A1 and possibly other sulfotransferases.

Figure 1

The structures of DHEA and raloxifene.

Figure 2. An initial-rate study of raloxifene and DHEA sulfation

Rates were determined at each of the 16 conditions defined by a 4 × 4 matrix of substrate concentrations that varied from 0.2 – 5.0 × K_m in equal increments in double-reciprocal space. Reactions were initiated by the addition of PAPS to solution containing [³H]-acceptor. Rates were obtained from slopes of 4-point progress curves in which less than 5% of concentration-limiting reactant consumed at the reaction endpoint was converted to product. Conditions were as follows: SULT2A1 (0.20 μM), MgCl₂ (5.0 mM), KPO₄ (25 mM, pH 7.4), 25 ± 2 °C. Raloxifene and DHEA concentrations are indicated in the figures. PAPS concentrations were as follows: (A) 30, 3.3, 1.8 and 1.2 μM; (B) 2.5, 0.28, 0.15 and 0.1 μM. Each point represents the average of two independent determinations. The lines through the points represent behavior predicted by a weighted least-squares fit to a sequential Bi-Bi model (48). Quenching, separation and quantitation protocols are described in Materials and Methods.

Figure 3. Equilibrium binding of RAL and DHEA to SULT2A1

(A) Raloxifene binding to E and E·PAP. Binding is monitored by changes in intrinsic enzyme fluorescence (λ_ex = 290, λ_em = 340). Solution composition is as follows: SULT2A1 (0.20 μM), PAP (0 or 125 μM), MgCl₂ (5.0 mM), KPO₄ (25 mM, pH 7.4), 25 ± 2 °C. Fluorescence intensity (ΔI) is normalized to total fluorescence change (ΔI_total). Each point is the average of two independent determinations. The line through the data is the behavior predicted by least-squares fitting using a model that assumes a single binding site per subunit. Further details are described in Materials and Methods. (B) DHEA binding to E and E·PAP. Conditions were identical to those described in Panel A except that PAP was 25 μM. (C) Quantitating ternary-complex binding interactions. Raloxifene solubility prevents saturation of the ternary complex. To quantitate the ternary-complex interactions, titrations with acceptor were carried out at series of five PAP concentrations, and K_d values were obtained by simultaneously fitting the data from the five titrations. Y-axis values report fold change in the acceptor affinity constant, and are given as the ratio of K_d at a fixed non-zero PAP concentration to that at zero PAP (K_{d (PAP = X)}/K_{d (PAP = 0)}). PAP concentrations were selected to range from 0 – 25 times its ternary-complex K_d; hence, X-axis values are given as [PAP] divided by its affinity constant ([PAP]/K_{d (E} · _Acceptor)). The absolute PAP concentrations were: 0, 5.0, 25, 50, and 125 μM (raloxifene titration); and 0, 1.0, 5.0, 10, and 25 μM (DHEA titration). Further details are given in Materials and Methods.

Figure 4. Nucleotide-linked gate closure discriminates substrates

(A) DHEA positions well for chemistry in the open and closed complexes. (B) Raloxifene is sterically prevented from accessing the acceptor binding pocked in the closed, but not the open complex. Open and closed models were constructed from the SULT2A1·DHEA (1J99) and SULT2A1·PAP (1EFH) binary structures, and ligands were docked using an evolution-based algorithm (Materials and Methods).

Figure 5. Presteady state binding of RAL to SULT2A1

(A) The binding of raloxifene to E. Binding reactions were initiated by rapidly (1:1) mixing a solution containing RAL (2.0 μM) with a solution containing SULT2A1 (0.10 μM). Binding was monitored by changes in intrinsic enzyme fluorescence (λ_ex = 290 nm, λ_em ≥ 330 nm). Fluorescence changes are given relative to the intensity (I/I₀) at t = 0. Each point represents the average of three independent determinations. The curve through the data represents the behavior predicted by the best fit to a single exponential model. Solution conditions: MgCl₂ (5.0 mM), KPO₄ (25 mM, pH 7.4), 25 ± 2 °C. (B) k_obs vs [raloxifene]. Progress curves were obtained at four concentrations of raloxifene and conditions are given in (A). Reactions were pseudo-first order in raloxifene in all cases and apparent rate constants were obtained by fitting with a single exponential model. Similar studies were performed for the binding of DHEA and raloxifene to E and E·PAP. Results are compiled in Table 3.

Figure 6. Presteady state binding of PAPS to SULT2A1·raloxifene

(A) Progress curve for the binding of PAPS to E·raloxifene. A solution containing PAPS (100 μM) and raloxifene (50 μM) was mixed rapidly (1:1) with a solution containing SULT2A1 (0.05 μM) and raloxifene (50 μM). The raloxifene concentration (50 μM) is saturating with respect to the binary complex (45 × K_d) and subsaturating with respect to the ternary complex (2.0 × K_d). The addition of PAPS forms ternary complex (causing the early-phase decrease in signal) and initiates dissociation of raloxifene (causing the subsequent increase). PAPS concentrations were high enough to be pseudo-first order and to achieve good separation of the phases. Conditions: MgCl₂ (5.0 mM), KPO₄ (25 mM, pH 7.4), 25 ± 2 °C. Fluorescence was excited at 290 nm and detected above 330 nm using a cutoff filter. Each point represents the average of three independent determinations. The curve through the data is the behavior predicted by the best fit obtained using a double exponential model (see Materials and Methods). (B) Expanded view of Panel A. The time axis of the Panel A data is expanded to highlight the separation of PAPS-binding and raloxifene-dissociation reactions. The 100 μM PAPS used in this experiment was the lowest concentration used in constructing the k_obs vs [PAPS] plot from which rate constants were obtained (Table 4).

Figure 7. The closed structure encapsulates PAPS

The active-site surface of the closed form of SULT2A1 is shown “wrapped” around the nucleotide. The limited access of nucleotide to solvent is highlighted by the dashed line circumscribing the only accessible solvent interface. Structural change is required to release the nucleotide.