Calcium/calmodulin stimulates the autophosphorylation of elongation factor 2 kinase on Thr-348 and Ser-500 to regulate its activity and calcium dependence

Abstract

Eukaryotic elongation factor 2 kinase (eEF-2K) is an atypical protein kinase regulated by Ca(2+) and calmodulin (CaM). Its only known substrate is eukaryotic elongation factor 2 (eEF-2), whose phosphorylation by eEF-2K impedes global protein synthesis. To date, the mechanism of eEF-2K autophosphorylation has not been fully elucidated. To investigate the mechanism of autophosphorylation, human eEF-2K was coexpressed with λ-phosphatase and purified from bacteria in a three-step protocol using a CaM affinity column. Purified eEF-2K was induced to autophosphorylate by incubation with Ca(2+)/CaM in the presence of MgATP. Analyzing tryptic or chymotryptic peptides by mass spectrometry monitored the autophosphorylation over 0-180 min. The following five major autophosphorylation sites were identified: Thr-348, Thr-353, Ser-445, Ser-474, and Ser-500. In the presence of Ca(2+)/CaM, robust phosphorylation of Thr-348 occurs within seconds of addition of MgATP. Mutagenesis studies suggest that phosphorylation of Thr-348 is required for substrate (eEF-2 or a peptide substrate) phosphorylation, but not self-phosphorylation. Phosphorylation of Ser-500 lags behind the phosphorylation of Thr-348 and is associated with the Ca(2+)-independent activity of eEF-2K. Mutation of Ser-500 to Asp, but not Ala, renders eEF-2K Ca(2+)-independent. Surprisingly, this Ca(2+)-independent activity requires the presence of CaM.

Figure 1. Characterization of enzymatic activity

Buffers used are described under ‘Experimental Procedures’. Kinase activity was determined by measuring the rate of phosphorylation of the peptide (µM.s⁻¹). (A) Calcium dependence assays were made with 0.5 nM eEF-2K and 0–3 µM free calcium. Data were fitted with equation 1, where n = 1.4 ± 0.03, $K_c^{\text{app}}$ = 0.14 ± 0.003 µM and $k_{\text{cat}}^{\text{app}}$ = 25.5 ± 0.2 s⁻¹. (B) Calmodulin dependence assays were made with 0.5 nM eEF-2K and 0–1 µM CaM. The data were fitted to equation 2, where $k_c^{\text{app}}$ = 76 ± 5 nM and $k_{\text{cat}}^{\text{app}}$ = 21.1 ± 0.2 s⁻¹. (C) Enzyme concentration dependence assays were made with 0–10 nM eEF-2K. (D) Magnesium dependence assays were made with 2 nM eEF-2K and 0–10 mM free magnesium. The data were fitted to equation 2, where $k_c^{\text{app}}$ = 0.33 ± 0.01 mM and $k_{\text{cat}}^{\text{app}}$ = 37.8 ± 0.3 s⁻¹. (E) Salt dependence assays were made with 2 nM eEF-2K and several concentrations (0–500 mM) of KAcO.

Figure 2. Autophosphorylation of eEF-2K

(A–C) eEF-2K (500 nM) was allowed to autophosphorylate in the presence of 5 µM CaM and 50 µM free Ca²⁺. At the indicated times, 10 pmol of eEF-2K were removed and the reaction quenched with hot SDS-PAGE sample loading buffer. The samples were then analyzed as described under ‘Experimental Procedures’. (A) Coomassie-stained gel. (B) Autoradiograph. (C) Phosphate incorporation as a function of autophosphorylation time – stoichiometry of the autophosphorylation of eEF-2K. Inset: Expansion of the data for 0–10 min. (D) Rate of phosphate incorporation (mole ³² P incorporated per min) as a function of enzyme concentration. To analyze the mechanism of autophosphorylation of recombinant human eEF-2K, varying concentrations of the purified enzyme (0–500 nM) were allowed to autophosphorylate in the presence of 2 µM CaM and 50 µM free Ca²⁺. The reaction was carried out under conditions in which linear incorporation of ³² P was achieved (1 min incubation) and quenched by addition of hot SDS-PAGE sample loading buffer. The samples were then analyzed as described under ‘Experimental Procedures’. The experiment was duplicated with similar results. (E) Effect of autophosphorylation on kinase activity. eEF-2K (20 nM) was allowed to autophosphorylate in the presence of 2 µM CaM and 50 µM free Ca²⁺. At the indicated times (0–180 min), the effect of autophosphorylation on kinase activity against the peptide substrate was determined by assaying the autophosphorylated enzyme (2 nM) in the presence of 55 µM free Ca²⁺ and 2.2 µM CaM (●) as described under ‘Experimental Procedures’. The rate of phosphorylation of the peptide (µM.s⁻¹) was determined using the general kinetic assay, and a graph of $k_{\text{obs}}^{\text{app}}$ (s⁻¹) as a function of the autophosphorylation time (min) was plotted. Activity of the unautophosphorylated control (without ATP) was also determined (■).

Figure 3. Mass spectrometry analysis sequence coverage of eEF-2K

(A) Sequence coverage of the purified recombinant eEF-2K from E. coli is ~ 90%, indicated by the residues in blue (). (B) Sequence coverage of the autophosphorylated enzyme (3 h incubation with CaM/Ca²⁺ /MgATP) is ~ 86%, indicated by the residues in red (). Both samples were resolved by SDS-PAGE, subjected to tryptic and chymotryptic in-gel digestion and the peptide digests then used for mass spectrometry analysis as described under ‘Supplementary Experimental Procedures’.

Figure 4. Purification and kinetic analysis of wild type eEF-2K and autophosphorylationsite mutants expressed in E. coli

(A) Samples purified by Ni-NTA affinity, CaM-agarose affinity and gel filtration chromatography were resolved by SDS-PAGE. (B) Kinase activity of the autophosphorylation-site mutants. Buffers used are described under ‘Experimental Procedures’. Assays were performed with 2 nM eEF-2K enzyme, 2 µM CaM and 50 µM free Ca²⁺. Kinase activity of the autophosphorylation-site mutants was determined by measuring the rate of phosphorylation of the peptide (µM.s⁻¹). Activities of the mutants are reported as the percentage of the wild type activity. The assays were performed in triplicate and error bars represent the standard deviation. (C) Activity of the autophosphorylation-site mutants against 4 µM wheat germ eEF-2, using 2 nM eEF-2K enzyme, 2 µM CaM and 50 µM free Ca²⁺ over an incubation time of 1 min. Upper panel: Coomassie-stained gel. Lower panel: Autoradiograph. (D) Autophosphorylation of wild type eEF-2K, and T348A and T348D eEF-2K mutants, using 1 µM eEF-2K enzyme in the presence of 5 µM calmodulin and 50 µM free Ca²⁺, over an incubation time of 10 min. Upper panel: Coomassie-stained gel. Lower panel: Autoradiograph.

Figure 5. Analysis of phosphate incorporation at Thr-348

(A) Characterization of antiphospho-eEF-2K (Thr-348) antibody. The antibody was characterized against 50 ng recombinant eEF-2K by immunoblotting as described under ‘Experimental Procedures’. Lanes: 1 – untreated eEF-2K WT; 2 – autophosphorylated eEF-2K WT; 3 – untreated eEF-2K T348A; 4 – autophosphorylated eEF-2K T348A; 5 – untreated eEF-2K T348D; 6 – autophosphorylated eEF-2K T348D; 7 – untreated eEF-2K WT; 8 – autophosphorylated eEF-2K WT; 9 – λ-phosphatase treated eEF-2K WT. (B) Time course of incorporation of phosphate at Thr-348. eEF-2K (500 nM) was allowed to autophosphorylate in the presence of 5 µM CaM and 50 µM free Ca²⁺. At the indicated times, 50 ng of eEF-2K were removed and the reaction quenched with hot SDS-PAGE sample loading buffer. The samples were then analyzed by Western blotting using the anti-phospho-eEF-2K (Thr-348) antibody as described under ‘Experimental Procedures’, (C) Graphical representation of (B). Western blots were quantified using ImageJ, and data then plotted as the percent phosphorylation of Thr-348 against autophosphorylation time. Inset: Expansion of the data for 0–12 min. Experiments were performed in duplicate, and error bars represent the standard deviation. (D) Average percent phosphorylation of residue Thr-348 of eEF-2K based on monitoring the TILR peptide by LCMS/MS. eEF-2K was allowed to autophosphorylate in the presence of CaM, Ca²⁺ and MgATP. After 0 min (no ATP added), 1 min, 10 min and 3 h, the reaction was quenched and the sample subjected to tryptic in-gel digestion followed by analysis by mass spectrometry as described under ‘Supplementary Experimental Procedures’. Inset: Expansion of the data for 0–12 min. Runs for each sample were performed in triplicate and error bars represent the standard deviation.

Figure 6. Analysis of phosphorylation at Ser-500

(A) Characterization of anti-phospho-eEF-2K (Ser-500) antibody. The antibody was characterized against 50 ng recombinant eEF-2K by immunoblotting as described under ‘Experimental Procedures’. Lanes: 1 – untreated eEF-2K WT; 2 – autophosphorylated eEF-2K WT; 3 – untreated eEF-2K S500A; 4 – autophosphorylated eEF-2K S500A; 5 – untreated eEF-2K S500D; 6 – autophosphorylated eEF-2K S500D. (B) Time course of incorporation of phosphate at Ser-500. eEF-2K (500 nM) was allowed to autophosphorylate in the presence of 5 µM CaM and 50 µM free Ca²⁺. At the indicated times, 50 ng of eEF-2K were removed and the reaction quenched with hot SDS-PAGE sample loading buffer. The samples were then analyzed by Western blotting using the anti-phospho-eEF-2K (Ser-500) antibody as described under ‘Experimental Procedures’. (C) Graphical representation of (B). Western blots were quantified using ImageJ, and data then plotted as the percent phosphorylation of Ser-500 against autophosphorylation time. Inset: Expansion of the data for 0–70 min. Experiments were performed in duplicate, and error bars represent the standard deviation. (D) Buffers used are described under ‘Experimental Procedures’. Assays were performed with eEF-2K enzyme, ± 50 µM free Ca²⁺ and ± 2 µM calmodulin. EGTA (1 mM) was added to all assays conducted in the absence of Ca²⁺. For eEF-2K WT, S500A and S500D assayed in the presence of both Ca²⁺ and CaM, and eEF-2K S500D assayed in the presence of only CaM, activities were much higher than the basal level of kinase activity, and hence only 5 nM of kinase was used. For all the other assays, 50 nM eEF-2K was used in order to detect an increase in kinase activity over the basal level. Kinase activity was determined by measuring the rate of phosphorylation of the peptide (µM.s⁻¹). Activities of the mutants are reported as the percentage of the wild type activity.

Scheme 1. Regulation of eEF-2K activity by multisite phosphorylation

Summary of the various phosphorylated residues on eEF-2K. Components are color coded as follows: ( - red) – suggested to be involved in the negative regulation of eEF-2K activity through an inhibitory phosphorylation (these sites include Ser-78, Ser-359, Ser-366 and Ser-396). Regulation through the mTOR pathway involves the phosphorylation of Ser-366 by p70 S6 kinase, and the phosphorylation of Ser-359 and Ser-78 by at least two additional unknown kinases (–24). It has been postulated that the Ser-78 phosphorylation acts to hinder the binding of CaM to eEF-2K (24). The cdc2-cyclin B complex has been shown to modulate eEF-2K activity via Ser-359 in a manner that is dependent on the cell cycle as well as amino acid availability, and is perhaps controlled by mTOR (25). Regulation through the MAPK cascade occurs via the phosphorylation of Ser-366 by p90^RSK1 in an ERK-dependent fashion (22). In addition, the stress-activated protein kinases p38α and p38δ inhibit eEF2K via phosphorylation on Ser-396 (23). p38δ is also known to phosphorylate eEF-2K on Ser-359 (21); ( - green) – suggested to be involved in the positive regulation of eEF-2K activity through an activating phosphorylation (these sites include Ser-398 and Ser-500). Phosphorylation of Ser-398 by the energy-supply regulator AMPK is known to activate eEF-2K (29). The cAMP-dependent PKA has also been shown to activate eEF-2K via a phosphorylation on Ser-500, and in the process imparts Ca²⁺-independent activity to the kinase (–28); ( - blue) – involved in autophosphorylation of eEF-2K (these sites include Thr-348, Thr-353, Ser-445, Ser-474 and Ser-500). Of the 5 autophosphorylation sites, only Thr-348 appears to be essential for activity against its substrate. Ser-500 is an autophosphorylation site and is also known to be phosphorylated by PKA, and could be the key residue responsible for autophosphorylation-induced Ca²⁺-independent (CaM-dependent – this work) activity (16, 17). The role of the phosphorylation at Ser-377 by MAPKAP-K2 has not yet been determined (23).