Nucleotide release sequences in the protein kinase SRPK1 accelerate substrate phosphorylation

Abstract

Protein kinases are essential signaling enzymes that transfer phosphates from bound ATP to select amino acids in protein targets. For most kinases, the phosphoryl transfer step is highly efficient, while the rate-limiting step for substrate processing involves slow release of the product ADP. It is generally thought that structural factors intrinsic to the kinase domain and the nucleotide-binding pocket control this step and consequently the efficiency of protein phosphorylation for these cases. However, the kinase domains of protein kinases are commonly flanked by sequences that regulate catalytic function. To address whether such sequences could alter nucleotide exchange and, thus, regulate protein phosphorylation, the presence of activating residues external to the kinase domain was probed in the serine protein kinase SRPK1. Deletion analyses indicate that a small segment of a large spacer insert domain and a portion of an N-terminal extension function cooperatively to increase nucleotide exchange. The data point to a new mode of protein kinase regulation in which select sequences outside the kinase domain constitute a nucleotide release factor that likely interacts with the small lobe of the kinase domain and enhances protein substrate phosphorylation through increases in ADP dissociation rate.

Figure 1. Diffusion Limits of ATP and SRSF1 in the Active Site of Wild-Type & Mutant Forms of SRPK1

A) Enzyme & Substrate Constructs. B,C) Viscosity Effects on Steady-State Kinetic Parameters. Relative values for k_cat/K_ATP (B), and k_cat/K_SR (C) in the absence and presence of varying sucrose concentrations are measured as a function of relative solvent viscosity (η^rel) for wild-type SRPK1 (○), SRPK1(ΔN) (•) and SRPK1(ΔS) (▲). Dotted lines reflect theoretical slope limits of 0 and 1. D) Bar graph showing changes in the association rate constants for SRSF1 and ATP.

Figure 2. N-Terminus & SID Promote ADP Dissociation from SRPK1

A) Viscosity effects on k_cat for wild-type SRPK1 (○), SRPK1(ΔN) (•) and SRPK1(ΔS) (▲). Dotted lines reflect theoretical slope limits of 0 and 1. B) Kinetic Mechanism. Enzyme (E) and SRSF1 (SR_n) are preequilibrated in the absence or presence of ADP and the reaction is started with ATP (k₁). SRSF1 (SR_n) is phosphorylated at the first site (k₂), ADP is released (k₃) and the product (SR_n+1) becomes the substrate for the next phosphorylation event (dotted line). C-E) C_ATT_RAP Experiments. SRSF1 and wild-type SRPK1 (C), SRPK1(ΔN) (D) and SRPK1(ΔS) (E) are prequilibrated in the absence (•) and presence (▲) of 120 ΔM ADP (in syringe) before the reaction is initiated with 600 ΔM ³²P-ATP. Insets show extended time frames for the reactions (2-15 sec). The data are simulated to obtain values for k_off, k₂, k₃ and [E*SR_n] (Table 2). F) Bar graph showing changes in k_off for ADP and k_cat.

Figure 3. Minimal Sequence Requirements in the SID for Efficient SRSF1 Phosphorylation

A) SID Deletion Constructs. Green and blue regions in the SID represent sequences spanning 223-249 and 484-493. B-D) Viscosity Effects on Steady-State Kinetic Parameters. Relative values for k_cat (B), k_cat/K_ATP (C) and k_cat/K_SR (D) in the absence and presence of varying sucrose concentrations are measured as a function of relative solvent viscosity (η^rel) for SRPK1[ΔS_INT] (•), SRPK1[ΔS_α1] (○), and SRPK1[ΔS_α2] (▲). Dotted lines reflect theoretical slope limits of 0 and 1. E, F) C_ATT_RAP Experiments for SRPK1[ΔS_INT] (E) and SRPK1[ΔS_α1] (F). SRSF1 and enzyme are preequilibrated in the absence (•) and presence (▲) of 120 ΔM ADP (in syringe) before the reaction is initiated with 600 ΔM ³²P-ATP. The data are simulated to obtain values for k_off, k₂, k₃, and [E*SR_n] (Table 2).

Figure 4. Effects of Mutations in S_α1 on Phosphorylation Kinetics

A) Molecular Contacts Between the S_α1 and Kinase Domain. S_α1 and small lobe of kinase domain are colored green and gray. N-terminal extension from residues 63-73 is shown in red (dotted line for 3 residues with insufficient electron density). B) Viscosity Effects on Turnover. Relative k_cat for SRPK1(I228G) (•) and SRPK1(I228K) (▲) in the absence and presence of varying sucrose concentrations are measured as a function of relative solvent viscosity (η^rel). Dotted lines reflect theoretical slope limits of 0 and 1. C) Pre-steady-state kinetic phosphorylation of SRSF1 using SRPK1(I228G). Final concentrations of enzyme, SRSF1 and ATP are 0.65, 2 and 50 ΔM, respectively. Data are simulated using the kinetic mechanism in Fig.2A and values are displayed in Table 2. D) Pre-steady-state kinetic phosphorylation of SRSF1 using SRPK1(I228K) (▲) and wt-SRPK1 (•). Final concentrations of wt-SRPK1, SRPK1(I228K), SRSF1 and ATP are 0.16, 0.2, 2 and 100 ΔM, respectively. Data for wt-SRPK1 are fit to equation (1) to obtain values of 0.9 ± .05, 9 ± 1 sec⁻¹, and 0.7 ± 0.1 sec⁻¹ for α, k_b and k_L, respectively. Data for SRPK1(I228K) are fit to a line function with a slope of 0.5 ± .05 sec⁻¹.

Figure 5. N-Terminal Residues Responsible for Efficient SRSF1 Phosphorylation

A) N-terminal Deletion Constructs. B-D) Viscosity Effects on Steady-State Kinetic Parameters. Relative values for k_cat (B), k_cat/K_ATP (C) and k_cat/K_SR (D) in the absence and presence of varying sucrose concentrations are measured as a function of relative solvent viscosity (η^rel) for SRPK1(ΔN_NT) (•) and SRPK1(ΔN_CT) (▲). Dotted lines reflect theoretical slope limits of 0 and 1. E) Pre-steady-state kinetic phosphorylation of SRSF1 using SRPK1(ΔN_NT) (•) and SRPK1(ΔN_CT) (▲). Final concentrations of enzyme, SRSF1 and ATP are 0.5, 2 and 50 ΔM, respectively. Data for SRPK1(ΔN_NT) are simulated using the kinetic mechanism in Fig. 2A and values are displayed in Table 2. Data for SRPK1(ΔN_CT) are normalized to total enzyme concentration and are fit to a linear function with a slope of 0.3 sec⁻¹.

Figure 6. Cooperative Interaction Between the N-Terminus & SID

A) Deletion Construct. B) C_ATT_RAP Experiment for SRPK1[ΔNS_α1]. SRSF1 (2 ΔM) and enzyme (0.5 ΔM) are prequilibrated in the absence (•) and presence (▲) of 120 ΔM ADP (in syringe) before the reaction is initiated with 600 ΔM ³²P-ATP. The data are simulated to obtain values for k_off, k₂, k₃, and [E*SR_n] (Table 2). C) Bar graph showing the effects on k_off for ADP and k_on for SRSF1. The bars corresponding to SRPK1[ΔN] + SRPK1[ΔS_α1] reflect theoretical values expected for SRPK1[ΔNS_α1] assuming both individual deletions are energetically additive. D) Steady-State Progress Curve. SRSF1 (1 ΔM) phosphorylation is monitored using 60 nM SRPK1[ΔNS_α1] (▲) or wt-SRPK1 (•) and 100 ΔM ATP. Phosphoproduct is normalized to the total concentration of SRSF1. The data for SRPK1 are fit to a double exponential function with amplitudes of 10 and 5 and rate constants of 0.5 and 0.05 min⁻¹. The data for SRPK1[ΔNS_α1] are fit to a single exponential function with an amplitude and rate constant of 14 and 0.027 min⁻¹.

Figure 7. Sequences Constituting the Nucleotide Release Factor in SRPK1

A) Mechanism for the acceleration of ADP release through the N-terminus and S_α1. B) Sequence homology within the SRPK family.