Allosteric enhancement of MAP kinase ​p38α's activity and substrate selectivity by docking interactions

Journal name:
Nature Structural & Molecular Biology
Volume:
21,
Pages:
704–711
Year published:
DOI:
doi:10.1038/nsmb.2861
Received
Accepted
Published online

Abstract

Mitogen-activated protein kinases (MAPKs) are essential to intracellular signal transduction. MAPKs anchor their pathway-specific substrates through so-called 'docking interactions' at locations distal from the active site. Docking interactions ensure efficient substrate recognition, but their contribution to the kinase reaction itself remains unclear. Herein, we use solution NMR to analyze the interaction between dually phosphorylated, active human ​p38α and the C-terminal fragments of its substrate ​MK2. ​p38α phosphorylation and ​ATP loading collaboratively induce the active conformation; subsequently, ​p38α accommodates ​MK2 phosphoacceptor residues in its active site. The docking interaction enhances binding of ​ATP and the phosphoacceptor to ​p38α, accelerating the phosphotransfer reaction. Thus, the docking interaction enhances ​p38α's enzymatic activity toward pathway-specific substrates allosterically as well as by the anchor effect. These findings clarify how MAPK cascades are organized in cells, even under ​ATP-depleted conditions often associated with environmental stress.

At a glance

Figures

  1. Functional sites in p38α and the substrate peptides used in this study.
    Figure 1: Functional sites in ​p38α and the substrate peptides used in this study.

    (a) Surface (left) and ribbon (right) representations of a model structure of the dually phosphorylated ​p38α. The model was generated by the SWISS-MODEL server50, with the crystal structure of ​p38γ-2P in complex with ​AMP-PNP (PDB 1CM8 (ref. 17)) as the template. The structure of bound ​AMP-PNP was omitted in the figure. The N and C lobes are colored green and pink, respectively. At right, Thr180 and Tyr182 in the activation loop, which are phosphorylated by MAPKK, are indicated by green stick representations, and the ATP-binding site, the P+1 site and the docking site are indicated by orange, cyan and magenta circles, respectively. (b) Schematic representation of the ​p38α substrate, ​MK2 and the peptides derived from ​MK2 used in this study.

  2. Effects of dual phosphorylation and ATP-analog binding on p38α NMR spectra.
    Figure 2: Effects of dual phosphorylation and ATP-analog binding on ​p38α NMR spectra.

    (a) Overlay of the annotated ​methionine and ​isoleucine (δ1) methyl 1H-13C band-selective optimized flip-angle short transient (SOFAST)-HMQC spectra of ​p38α-2P in the absence (black) and presence (red) of 5 mM ATP analog. Large chemical-shift changes are indicated by blue arrows. (b) Overlay of the annotated ​methionine and ​isoleucine (δ1) methyl 1H-13C SOFAST-HMQC spectra of ​p38α-2P (blue) and unphosphorylated ​p38α (black) in the apo state.

  3. Chemical-shift changes induced by ATP-analog binding to p38α-2P and dual phosphorylation.
    Figure 3: Chemical-shift changes induced by ATP-analog binding to ​p38α-2P and dual phosphorylation.

    (a) Bar graphs showing the methyl CSPs induced by ATP-analog binding to ​p38α-2P (left) and the chemical-shift differences between ​p38α-2P and unphosphorylated ​p38α (right). The residues in the ATP-binding, docking and P+1 sites are shaded in orange, magenta and cyan, respectively. Unassigned methyl sites are indicated by asterisks. CSP values larger than 0.4 p.p.m. are indicated. For ​leucine and ​valine, the values for δ1/γ1 are shown above the values for δ2/γ2. (b) Mapping of the CSPs induced by ATP-analog binding to ​p38α-2P (a, left) on the crystal structure of ​apo-p38α (PDB 1A9U51). ILVM-methyl groups are indicated as spheres, and those with CSPs larger than 0.05 p.p.m. are colored red. The ATP-binding site is highlighted by the orange oval. The structure is displayed in the same orientation as in Figure 1a.

  4. Effects of substrate binding on p38α NMR spectra.
    Figure 4: Effects of substrate binding on ​p38α NMR spectra.

    (a) Overlay of the annotated ​methionine and ​isoleucine (δ1) methyl 1H-13C SOFAST-HMQC spectra of ​p38α-2P without the ATP analog in the absence (black) and presence (red) of an equimolar amount of the 334D peptide. Large chemical-shift changes are indicated by blue arrows. (b) As in a, except that ​p38α-2P is in complex with the ATP analog.

  5. Site-resolved analyses of substrate binding to p38α-2P.
    Figure 5: Site-resolved analyses of substrate binding to ​p38α-2P.

    (a) Bar graphs showing the methyl CSPs induced by 334D-peptide binding to ​p38α-2P in the absence (left) and presence (right) of the ATP analog. The residues in the ATP-binding, docking and P+1 sites are shaded in orange, magenta and cyan, respectively. Unassigned methyl sites are indicated by asterisks. CSP values larger than 0.1 p.p.m. are indicated. The CSPs at 0.02 p.p.m. and 0.04 p.p.m. are indicated by dashed gray lines. For ​leucine and ​valine, the values for δ1/γ1 are shown above the values for δ2/γ2. (b,c) Mapping of the CSPs induced by 334D-peptide binding on the crystal structure of ​p38α (PDB 2OKR29) without (b) and with (c) the ATP analog. The overlays of the 1H-13C spectral regions of selected methyl resonances of the hinge region (Met109), the docking sites (Ile116 and Val158) and the P+1 site (Met194, Ile229 and Ile259) are also indicated. ILVM-methyl groups that exhibited CSPs above 0.04 p.p.m. or 0.02 p.p.m. are represented as red and yellow spheres, respectively. Black spheres in c represent methyl groups with unassigned resonances in the 334D peptide–bound and/or the ATP analog–loaded ​p38α-2P. The ATP-binding and P+1 sites are indicated by orange and cyan ovals, respectively.

  6. Allosteric positive modifications of p38α-2P kinase-reaction steps by the docking interaction.
    Figure 6: Allosteric positive modifications of ​p38α-2P kinase-reaction steps by the docking interaction.

    (a) ATP-unbound population of ​p38α-2P. (b) 334 peptide concentration–dependent, normalized CSPs of Ile259 resonances. a.u., arbitrary units. (c) Lineweaver-Burk plots of the phosphorylation of the 334 peptide by ​p38α-2P. Black and magenta lines in ac represent data with and without the D peptide, respectively. For a and b, data representative of two independent experiments are shown. For c, error bars indicate the fitting errors of reaction velocities in independent experiments (n = 3). Experimental details are in Online Methods. (d) Schematic representation of ​p38α-2P kinase reaction process. As the initial step of activation, ​p38α is dually phosphorylated by the upstream MAPKKs (left). For genuine ​p38α substrates containing the docking sequence (top), ​p38α-2P recognizes the substrate via the docking interaction, binds ​ATP and then binds to the phosphoacceptor site of the substrate to lead to substrate phosphorylation. The docking-induced enhancements of the affinities and kinetics, in comparison to the (pseudo)substrate without the docking sequence (bottom) are indicated. Although omitted for clarification, ​ATP might bind before substrate docking. In that case, the affinities of the docking fragments to ​p38α-2P are enhanced by the ​ATP preloading (Supplementary Table 1). (e) ​ATP concentration dependence of the normalized reaction efficiencies (E) with (magenta) and without (black) formation of the docking interaction (details in Online Methods). ​ATP concentrations under the normal and the depleted conditions are indicated by the black and blue dashed lines, respectively.

  7. Characterization of p38α and the model substrates.
    Supplementary Fig. 1: Characterization of ​p38α and the model substrates.

    (a) Phos-tagTM SDS-PAGE of ​p38α, which is dually phosphorylated by ​MKK6DD (lane 2). The sample was then dephosphorylated by either ​HePTP, a phosphatase that is specific to phospho-Tyr182 (lanes 3 and 4), or ​PPM1A, a phosphatase that is specific to phospho-Thr180 (lanes 7 and 8). In lanes 5, 6, 9, and 10, the ​p38α-2P was treated with both of the phosphtases in different orders. The mobility of ​p38α-2P treated with ​HePTP increased (​p38α-1PT), indicating the dephosphorylation of Tyr182 (lanes 3 and 4). Thr180, however, remained phosphorylated, as the mobility is still slower than unphosphorylated ​p38α. Further treatment of ​p38α-1PT with ​PPM1A led to the complete regression to unphosphorylated ​p38α (lanes 5 and 6). This indicated that ​p38α was successfully dually phosphorylated by ​MKK6DD. Although ​p38α-2P first treated with ​PPM1A exhibited almost the same mobility as ​p38α-2P (​p38α-1PY; lanes 7 and 8), the subsequent treatment with ​HePTP resulted in the same mobility as unphosphorylated ​p38α. It seems that the mobilities of ​p38α-1PY and ​p38α-2P are indiscernible. The details of the reaction conditions are described in the Online Methods. (b-i) Characterization of the phosphorylation states of ​p38α based on the spectral pattern of the Thr-γ2 region of the 1H-13C HSQC spectra. Panels (b) to (e) show the spectra of ​p38α, ​p38α-2P, ​p38α-1PT, and ​p38α-1PY, respectively. The ​p38α-1PT and ​p38α-1PY proteins were prepared by treating ​p38α-2P with ​HePTP and ​PPM1A, respectively. Methyl resonances with distinctive chemical shifts in the respective phosphorylation states are highlighted by the blue rectangles. Panels (f) to (i) are the spectra of ​p38α phosphorylated with 1/70 mol. eq. of ​MKK6DD at 14 °C for the indicated times. A 20 hour reaction rendered most of the ​p38α dually phosphorylated. (j) Phos-tagTM SDS-PAGE of ​MK2 (47–400) phosphorylated by ​p38α-2P. Lane 1 is the reference which is not treated with ​p38α-2P, and lanes 2 to 7 show the timecourse of the phosphorylation of 5 μM ​MK2 (47–400) by 0.5 μM ​p38α-2P in the presence of 2 mM ​ATP at 25 °C. The time-dependent, discrete decrease in the mobility is indicative of the multisite phosphorylation of ​MK2 (47–400) by ​p38α-2P. (k) The retention volumes in the SEC analyses of ​p38α-2P in the absence and presence of 1 mM ​ATP. Samples were prepared as 240 μL solution of 25 μM ​p38α and were analyzed by chromatography on a Superdex 200 GL 10 300 column connected to AKTA Explorer 10S (GE Healthcare) at 4 °C. The retention volumes and the error bars are the averages and the standard deviations of five repeated runs. The larger error bar in the presence of ​ATP is due to the rough baseline caused by the high background absorbance at 280 nm by ​ATP. Nevertheless, the retention volume in the presence of ​ATP is larger than that in the absence of ​ATP, suggesting that the conformation of ​p38α-2P is more compact (i.e. “closed”) when ​ATP is loaded. (l) Competition experiment between the 334D-peptide and ​MK2 (47-400). Western blot of ​p38α-2P, detected with 1:1000 dilution of the anti-p38 monoclonal antibody. The N-terminally GB1-fused 334D-peptide was immobilized on IgG Sepharose beads, and then ​p38α-2P was further immobilized on them (input; lane 6). After washing the beads, ​p38α-2P was eluted with the longer ​MK2 fragment (47–400) (0.3, 0.6, and 1.2 μM; lanes 2–4). The intensity of the ​p38α-2P band increased in proportion to the ​MK2 concentration, which indicates that the longer ​MK2 fragment competes with the GB1-334D-peptide for the same site on ​p38α-2P. The dissociation of ​p38α-2P from the beads by washing is negligible, as the band is very weak when the elution buffer without the longer ​MK2 fragment is used (lane 1). The non-specific binding of ​p38α-2P on the beads is also negligible, as shown in the experiment using GB1 instead of the GB1-334D-peptide (lane 5). The detailed experimental conditions are provided in the Online Methods. (m) Phosphorylations of the 334D-peptide and the 334-peptide by ​p38α-2P were monitored by time-dependent increase of the phospho-334D-peptide (red circles) and the phospho-334-peptide (cyan squares). The reactions were tracked by successive measurements of 1H-15N SOFAST-HMQC spectra. The concentration of the phosphorylated product was estimated from the intensity of the phospho-Thr334 resonance for each peptide. The peptides (100 μM) were phosphorylated by 20 nM of ​p38α-2P, in the presence of 2 mM ​ATP at 25 °C. The error bars correspond to the level of noise in each spectrum.

  8. Effect of phosphorylation and ATP titration on p38α.
    Supplementary Fig. 2: Effect of phosphorylation and ​ATP titration on ​p38α.

    (a) Overlay of the annotated ​Leu/​Val methyl 1H-13C SOFAST-HMQC spectra of ​p38α-2P in the absence (black) and presence (red) of 5 mM ATP analog. Large chemical shift changes are indicated by blue arrows. (b) Left, center: Overlay of the ​Ile(δ1) and ​Leu/​Val methyl 1H-13C SOFAST-HMQC spectra of 20 μM [ILV-methyl] ​p38α-2P in the presence (red) and absence (black) of 5 mM ​ATP. Large chemical shift changes in the ​Ile(δ1) region are indicated by blue arrows. Spectra were measured with an 800 MHz 1H resonance frequency spectrometer at 10 °C. Right: Methyl sites that exhibited chemical shift changes larger than the linewidth or were not assigned only in the ATP-bound state are indicated as red spheres on the crystal structure (PDB code: 1A9U51). The ATP-binding site is highlighted in the orange oval. (c) Overlay of the annotated ​Leu/​Val methyl 1H-13C SOFAST-HMQC spectra of ​p38α-2P (blue) and unphosphorylated ​p38α (black) in the apo state. (d) Comparison of the ​Ile(δ1) and ​Leu/​Val methyl regions of 1H-13C HMQC spectra of 80 μM [ILVM-methyl] unphosphorylated ​p38α, in the presence (green) and absence (red) of 4 mM ​ATP. Spectra were measured at 800 MHz (1H frequency) at 25 °C. For the methyl resonances of ​Leu and ​Val residues, only those dispersed in the spectra are labeled, for clarification. (e) ​Met methyl regions of 1H-13C HMQC spectra of unphosphorylated ​p38α titrated with various concentration of ​ATP. The resonance of Met109, which was traced during the titration to determine the affinity of unphosphorylated ​p38α for ​ATP, is expanded as an inset, in which all of the titration points are overlaid. The dissociation constant of unphosphorylated ​p38α for ​ATP was estimated to be larger than 15 mM, based on the CSPs of the Met109 methyl 1H and 13C resonances. Since the chemical shift changes did not reach a plateau at the highest ​ATP concentration (15 mM), a unique dissociation constant was not obtained.

  9. Changes in 1H-15N TROSY spectra of p38α upon dual phosphorylation.
    Supplementary Fig. 3: Changes in 1H-15N TROSY spectra of ​p38α upon dual phosphorylation.

    Comparison of 1H-15N TROSY spectra of unphosphorylated ​p38α (​p38α-0P; black) and ​p38α-2P (red) in the absence of ​ATP. The regions containing the resonances of Met179, Thr180, Gly181, Tyr182, and Val183, which are located in the activation loop, are expanded to indicate that these resonances disappeared upon dual phosphorylation. In the lower left panel, the signal from ​p38α-2P, which overlaps with the Met179 signal from ​p38α-0P, does not belong to Met179 and is presumably from the unassigned resonance (n.a.) below.

  10. Effects of substrate binding on p38α NMR spectra.
    Supplementary Fig. 4: Effects of substrate binding on ​p38α NMR spectra.

    (a) Overlay of the annotated ​Leu/​Val methyl 1H-13C SOFAST-HMQC spectra of ​p38α-2P without the ATP analog, in the absence (black) and presence (red) of an equimolar amount of the 334D-peptide. Large chemical shift changes are indicated by blue arrows. (b) The same as (a), except that ​p38α-2P is in complex with the ATP analog.

  11. Interaction between the substrate phosphoacceptor site and p38α-2P upon ATP loading.
    Supplementary Fig. 5: Interaction between the substrate phosphoacceptor site and ​p38α-2P upon ​ATP loading.

    (a) Extensive interaction between the 334D-peptide and ​p38α-2P upon the ATP analog loading. Comparison of the 1H-15N TROSY spectra of a stoichiometric complex of 180 μM [U-15N] 334D-peptide and unlabeled ​p38α-2P, in the absence (black) and presence (red) of 5 mM ATP analog. Several resonances, including that of the phosphoacceptor residue, Thr334, lost intensity when the ATP analog was added, reflecting the increase in the transverse relaxation rate. This would be explained by the increase in the effective local rotational correlation times due to the binding of Thr334 to the active site of ​p38α-2P and/or the generation of chemical exchange with relatively long (msec – μsec) time scale associating with the binding. This suggested that an extensive interaction is formed by the loading of the ATP analog on ​p38α-2P. (b) Spectral perturbations induced upon the addition of the T334A mutants of 334D-peptide to ​p38α-2P. Comparison of the regions of the 1H-13C HMQC spectra of 40 μM [ILV-methyl] ​p38α-2P that contain the resonance of either Ile229 or Ile259, which is located around the P+1 site, in the absence (black) and presence (red) of an equimolar amount of the T334A mutant of the 334D-peptide. The panels on the left and right are the spectra in the absence of presence of ​ATP (not analog; 5 mM). The addition of the unlabeled peptide induced CSPs in the P+1 site resonances only in the presence of ​ATP. The addition of the peptide induced CSPs to the P+1 site resonances of ​p38α-2P in the presence of ​ATP. Spectra were recorded at 10 °C, with an 800 MHz 1H resonance frequency spectrometer. (c, d) Comparison of the ​Ile-δ1 regions of 1H-13C HMQC spectra of 40 μM [ILV-methyl] ​p38α-2P, in the absence (black) and presence (red) of a 10-fold molar excess of the 334-peptide without (c) and with (d) a saturating amount of the ATP analog (5 mM). Regions containing the resonances of Ile229 and Ile259, which are located near the P+1 site, are enlarged. In the presence of the ATP analog, the resonance from Ile229 disappeared upon the addition of the 334-peptide. (e) Spectral perturbations induced upon the addition of the T334A mutants 334-peptide to ​p38α-2P. The experimental conditions are the same as (b), except that the titrated peptide was a 10-fold molar excess of the unlabeled 334-peptide with the T334A mutation. (f, g) Extensive interaction between the 334-peptide and ​p38α-2P induced by the ATP analog. The methyl region of the one dimensional (1D) proton NMR spectrum of 100 μM unlabeled 334-peptide in the free form (f, blue) was compared to that with 50 μM ​p38α-2P in the absence (g, black) and presence of the ATP analog (4 mM; g, red). The signals from the 334-peptide were not significantly affected by the presence of ​p38α-2P itself, while the broad signals derived from ​p38α-2P are also observed (g, black). However, considerable broadening of the 334-peptide signals were observed upon the addition of 4 mM ATP analog (g, red), indicating that the transverse relaxation rates of the methyl proton resonances of the 334-peptide increased. This would result from the increased effective rotational correlation time of the 334-peptide upon the binding to ​p38α-2P, as well as the relaxation enhancement due to the exchange between free and bound states with distinct chemical shifts. Thus, the 334-peptide interacts with ​p38α-2P, which is loaded with the ATP analog.

  12. Allosteric effects induced by the docking interaction to p38α-2P.
    Supplementary Fig. 6: Allosteric effects induced by the docking interaction to ​p38α-2P.

    (a) Comparison of the ILV-methyl regions of the 1H-13C HMQC spectra of ​p38α-2P in the absence (black) and presence (magenta) of a 1.1 molar equivalent of the D-peptide. Spectra were recorded in the presence of 5 mM ATP analog. Regions containing the resonances of Val89, located in the ATP-binding site, and Ile259, located in the P+1 site, are shown in orange and cyan squares, respectively. (b) Mapping of the methyl groups that showed CSPs larger than 0.02 ppm, upon the addition of the D-peptide in the presence of 5 mM ATP analog, shown in red, while those that are not assigned only in the presence of the ATP analog are shown in black. The docking sequence of ​MK2 is shown as purple ribbon. The structure figure was generated from the X-ray structure of ​p38α in complex with the ​MK2 docking-peptide (PDB code: 2OKR29).

  13. Enhancements of catalytic steps of p38α by the docking interaction.
    Supplementary Fig. 7: Enhancements of catalytic steps of ​p38α by the docking interaction.

    (a) Determination of the affinity of ​p38α-2P to the ATP analog. Ile84-δ1 signals in the 1H-13C HMQC spectra of 40 μM [ILV-methyl] ​p38α-2P, without (upper) and with (lower) a 1.1 molar excess of the D-peptide. Both samples contained 200 μM of the ATP analog. High field (upper right) and low field resonances (lower left) are from the ATP analog unbound and bound states, respectively. 1D slices in the 1H dimension are overlaid on the 2D signals. The intensity of the resonance from the ATP analog-bound state becomes stronger in the presence of the D-peptide. (b) Determination of the affinity of ​p38α-2P to ​ATP. Left; Expanded regions of the spectra for the probe resonances. The upper and lower rows correspond to the conditions without and with a 1.1 molar equivalent of the D-peptide, respectively. From left to right, the conditions represent 5 mM ATP analog, the mixture of 2.5 mM ATP analog and 2.5 mM ​ATP, and 5 mM ​ATP, respectively. Right; The intensities of the probe resonances with the different concentrations of ​ATP and the ATP analog. The intensities of the resonances in each condition were normalized to the reference spectra for the ATP-bound form. The upper and lower panels are the data in the absence and presence of the D-peptide, respectively. The intensities for the ATP and ATP analog bound resonances were plotted as the orange squares and green diamonds, respectively, and were used to determine the affinity of ​ATP, as described in the Supplemental Materials and Methods. The error bars correspond to the level of noise in each spectrum. (c) Titration of the 334-peptide to ​p38α-2P. Overlays of the ​Ile-δ1 regions of 1H-13C HMQC spectra of 40 μM [ILV-methyl] ​p38α-2P with different concentrations of the 334-peptide, in the absence (left) and presence (right) of a 1.1 molar equivalent of the D-peptide. Black, orange, and red spectra correspond to 0, 120, and 400 μM concentrations of the 334-peptide, respectively. Only three spectra of the five titration points are shown for visual clarity. The resonances of Ile259 near the P+1 site are indicated by the cyan rectangles, and are enlarged in the insets. The dose-dependent CSPs of the Ile259 methyl resonance were used to determining the affinity of ATP-loaded ​p38α-2P for the 334-peptide. (d) Phosphorylation of the 334-peptide by ​p38α-2P. Left; Expanded region of the phosphorylated Thr334 resonance in the 1H-15N HSQC spectra of 100 μM [U-15N] 334-peptide, before (black) and after (red) phosphorylation by ​p38α-2P. Only after reaction, the resonance of the phosphorylated Thr334 is observed, with the low-field chemical shifts characteristic of phosphorylated residues64. Right; Time-dependent increase of the phosphorylated 334-peptide produced by ​p38α-2P, in the absence (black) and presence (magenta) of the D-peptide. The concentration of the product was estimated from the intensity of the phospho-Thr334 resonance. The detailed experimental conditions are provided in the Online Methods.

  14. Determination of the affinities of the docking fragments for p38α-2P.
    Supplementary Fig. 8: Determination of the affinities of the docking fragments for ​p38α-2P.

    (a-d) NMR titration experiments of the docking fragment of ​MEF2A to ​p38α-2P, in the absence and presence of the ATP analog. (a, b) Comparison of the 1H-13C HMQC spectra (​Ile-δ1 region) of 20 μM [ILV-methyl] ​p38α-2P with different concentrations of the ​MEF2A docking fragment in the absence (a) and presence (b) of 4 mM ATP analog. The arrows trace the chemical shift changes of Ile116, which exhibited the largest CSP values. (c, d) CSPs induced in the methyl resonances of Val83, Ile116, Val117, Val127, Ile131, Ile134, and Val158, which are located in and near the docking site, by titration of the ​MEF2A docking fragment. The dissociation constants are indicated in the plots with standard deviation estimated from fitting errors. (e, f) ITC experiments of the D-peptide to ​p38α-2P in the absence (e) and presence (f) of the ATP analog. The dissociation constants are indicated in the plots. The detailed experimental conditions are provided in the Online Methods.

Accession codes

Primary accessions

Biological Magnetic Resonance Data Bank

Referenced accessions

Biological Magnetic Resonance Data Bank

References

  1. Raman, M., Chen, W. & Cobb, M.H. Differential regulation and properties of MAPKs. Oncogene 26, 31003112 (2007).
  2. Avruch, J. MAP kinase pathways: the first twenty years. Biochim. Biophys. Acta 1773, 11501160 (2007).
  3. Cargnello, M. & Roux, P.P. Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases. Microbiol. Mol. Biol. Rev. 75, 5083 (2011).
  4. Plotnikov, A., Zehorai, E., Procaccia, S. & Seger, R. The MAPK cascades: signaling components, nuclear roles and mechanisms of nuclear translocation. Biochim. Biophys. Acta 1813, 16191633 (2011).
  5. Marshall, C.J. Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activation. Cell 80, 179185 (1995).
  6. Cuadrado, A. & Nebreda, A.R. Mechanisms and functions of p38 MAPK signalling. Biochem. J. 429, 403417 (2010).
  7. Schindler, J.F., Monahan, J.B. & Smith, W.G. p38 pathway kinases as anti-inflammatory drug targets. J. Dent. Res. 86, 800811 (2007).
  8. Dodeller, F. & Schulze-Koops, H. The p38 mitogen-activated protein kinase signaling cascade in CD4 T cells. Arthritis Res. Ther. 8, 205 (2006).
  9. Johnson, G.L. & Lapadat, R. Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science 298, 19111912 (2002).
  10. Anderson, N.G., Maller, J.L., Tonks, N.K. & Sturgill, T.W. Requirement for integration of signals from two distinct phosphorylation pathways for activation of MAP kinase. Nature 343, 651653 (1990).
  11. Zhang, Y.Y., Mei, Z.Q., Wu, J.W. & Wang, Z.X. Enzymatic activity and substrate specificity of mitogen-activated protein kinase p38α in different phosphorylation states. J. Biol. Chem. 283, 2659126601 (2008).
  12. Keyse, S.M. Protein phosphatases and the regulation of mitogen-activated protein kinase signalling. Curr. Opin. Cell Biol. 12, 186192 (2000).
  13. Goldsmith, E.J., Cobb, M.H. & Chang, C.I. Structure of MAPKs. Methods Mol. Biol. 250, 127144 (2004).
  14. Wang, Z. et al. The structure of mitogen-activated protein kinase p38 at 2.1-A resolution. Proc. Natl. Acad. Sci. USA 94, 23272332 (1997).
  15. Wilson, K.P. et al. Crystal structure of p38 mitogen-activated protein kinase. J. Biol. Chem. 271, 2769627700 (1996).
  16. Zhang, F., Strand, A., Robbins, D., Cobb, M. & Goldsmith, E. Atomic structure of the MAP kinase ERK2 at 2.3 A resolution. Nature 367, 704711 (1994).
  17. Bellon, S., Fitzgibbon, M.J., Fox, T., Hsiao, H.M. & Wilson, K.P. The structure of phosphorylated p38γ is monomeric and reveals a conserved activation-loop conformation. Structure 7, 10571065 (1999).
  18. Canagarajah, B.J., Khokhlatchev, A., Cobb, M.H. & Goldsmith, E.J. Activation mechanism of the MAP kinase ERK2 by dual phosphorylation. Cell 90, 859869 (1997).
  19. Zhang, Y.Y., Wu, J.W. & Wang, Z.X. Mitogen-activated protein kinase (MAPK) phosphatase 3-mediated cross-talk between MAPKs ERK2 and p38α. J. Biol. Chem. 286, 1615016162 (2011).
  20. Weston, C.R., Lambright, D.G. & Davis, R.J. Signal transduction. MAP kinase signaling specificity. Science 296, 23452347 (2002).
  21. Ubersax, J.A. & Ferrell, J.E. Mechanisms of specificity in protein phosphorylation. Nat. Rev. Mol. Cell Biol. 8, 530541 (2007).
  22. Tanoue, T. & Nishida, E. Docking interactions in the mitogen-activated protein kinase cascades. Pharmacol. Ther. 93, 193202 (2002).
  23. Tanoue, T., Adachi, M., Moriguchi, T. & Nishida, E. A conserved docking motif in MAP kinases common to substrates, activators and regulators. Nat. Cell Biol. 2, 110116 (2000).
  24. Gupta, S., Campbell, D., Derijard, B. & Davis, R.J. Transcription factor ATF2 regulation by the JNK signal transduction pathway. Science 267, 389393 (1995).
  25. Kallunki, T. et al. JNK2 contains a specificity-determining region responsible for efficient c-Jun binding and phosphorylation. Genes Dev. 8, 29963007 (1994).
  26. Yang, S.-H., Galanis, A. & Sharrocks, A.D. Targeting of p38 mitogen-activated protein kinases to MEF2 transcription factors. Mol. Cell. Biol. 19, 40284038 (1999).
  27. Lukas, S.M. et al. Catalysis and function of the p38αMK2a signaling complex. Biochemistry 43, 99509960 (2004).
  28. Chang, C.I., Xu, B.E., Akella, R., Cobb, M.H. & Goldsmith, E.J. Crystal structures of MAP kinase p38 complexed to the docking sites on its nuclear substrate MEF2A and activator MKK3b. Mol. Cell 9, 12411249 (2002).
  29. ter Haar, E., Prabhakar, P., Prabakhar, P., Liu, X. & Lepre, C. Crystal structure of the p38α-MAPKAP kinase 2 heterodimer. J. Biol. Chem. 282, 97339739 (2007).
  30. Tanoue, T., Maeda, R., Adachi, M. & Nishida, E. Identification of a docking groove on ERK and p38 MAP kinases that regulates the specificity of docking interactions. EMBO J. 20, 466479 (2001).
  31. Reményi, A., Good, M.C., Bhattacharyya, R.P. & Lim, W.A. The role of docking interactions in mediating signaling input, output, and discrimination in the yeast MAPK network. Mol. Cell 20, 951962 (2005).
  32. Songyang, Z. et al. A structural basis for substrate specificities of protein Ser/Thr kinases: primary sequence preference of casein kinases I and II, NIMA, phosphorylase kinase, calmodulin-dependent kinase II, CDK5, and Erk1. Mol. Cell. Biol. 16, 64866493 (1996).
  33. White, A., Pargellis, C.A., Studts, J.M., Werneburg, B.G. & Farmer, B.T. Molecular basis of MAPK-activated protein kinase 2:p38 assembly. Proc. Natl. Acad. Sci. USA 104, 63536358 (2007).
  34. Szafranska, A.E. & Dalby, K.N. Kinetic mechanism for p38 MAP kinase α. FEBS J. 272, 46314645 (2005).
  35. Akella, R., Min, X., Wu, Q., Gardner, K.H. & Goldsmith, E.J. The third conformation of p38α MAP kinase observed in phosphorylated p38α and in solution. Structure 18, 15711578 (2010).
  36. Jacobs, D., Glossip, D., Xing, H., Muslin, A.J. & Kornfeld, K. Multiple docking sites on substrate proteins form a modular system that mediates recognition by ERK MAP kinase. Genes Dev. 13, 163175 (1999).
  37. Lee, S. et al. A model of a MAPKsubstrate complex in an active conformation: a computational and experimental approach. PLoS ONE 6, e18594 (2011).
  38. Masterson, L.R., Mascioni, A., Traaseth, N.J., Taylor, S.S. & Veglia, G. Allosteric cooperativity in protein kinase A. Proc. Natl. Acad. Sci. USA 105, 506511 (2008).
  39. Zimmerman, S.B. & Trach, S.O. Estimation of macromolecule concentrations and excluded volume effects for the cytoplasm of Escherichia coli. J. Mol. Biol. 222, 599620 (1991).
  40. Lugo, T.G., Pendergast, A.M., Muller, A.J. & Witte, O.N. Tyrosine kinase activity and transformation potency of bcr-abl oncogene products. Science 247, 10791082 (1990).
  41. Heisterkamp, N. et al. Acute leukaemia in bcr/abl transgenic mice. Nature 344, 251253 (1990).
  42. Dennis, P.B. et al. Mammalian TOR: a homeostatic ATP sensor. Science 294, 11021105 (2001).
  43. Knight, Z.A. & Shokat, K.M. Features of selective kinase inhibitors. Chem. Biol. 12, 621637 (2005).
  44. Steenbergen, C., Murphy, E., Watts, J.A. & London, R.E. Correlation between cytosolic free calcium, contracture, ATP, and irreversible ischemic injury in perfused rat heart. Circ. Res. 66, 135146 (1990).
  45. Schütt, F., Aretz, S., Auffarth, G.U. & Kopitz, J. Moderately reduced ATP levels promote oxidative stress and debilitate autophagic and phagocytic capacities in human RPE cells. Invest. Ophthalmol. Vis. Sci. 53, 53545361 (2012).
  46. Kumphune, S., Chattipakorn, S. & Chattipakorn, N. Role of p38 inhibition in cardiac ischemia/reperfusion injury. Eur. J. Clin. Pharmacol. 68, 513524 (2012).
  47. Nielsen, G. & Schwalbe, H. NMR spectroscopic investigations of the activated p38α mitogen-activated protein kinase. ChemBioChem 12, 25992607 (2011).
  48. Francis, D.M. et al. Structural basis of p38α regulation by hematopoietic tyrosine phosphatase. Nat. Chem. Biol. 7, 916924 (2011).
  49. Piserchio, A. et al. Docking interactions of hematopoietic tyrosine phosphatase with MAP kinases ERK2 and p38α. Biochemistry 51, 80478049 (2012).
  50. Arnold, K., Bordoli, L., Kopp, J. & Schwede, T. The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics 22, 195201 (2006).
  51. Wang, Z. et al. Structural basis of inhibitor selectivity in MAP kinases. Structure 6, 11171128 (1998).
  52. Gardner, K.H., Rosen, M.K. & Kay, L.E. Global folds of highly deuterated, methyl-protonated proteins by multidimensional NMR. Biochemistry 36, 13891401 (1997).
  53. Rosen, M.K. et al. Selective methyl group protonation of perdeuterated proteins. J. Mol. Biol. 263, 627636 (1996).
  54. Underwood, K.W. et al. Catalytically active MAP KAP kinase 2 structures in complex with staurosporine and ADP reveal differences with the autoinhibited enzyme. Structure 11, 627636 (2003).
  55. Kinoshita, E., Takahashi, M., Takeda, H., Shiro, M. & Koike, T. Recognition of phosphate monoester dianion by an alkoxide-bridged dinuclear zinc(II) complex. Dalton Trans. 21, 11891193 (2004).
  56. Vogtherr, M. et al. NMR backbone assignment of the mitogen-activated protein (MAP) kinase p38. J. Biomol. NMR 32, 175 (2005).
  57. Tugarinov, V., Hwang, P.M., Ollerenshaw, J.E. & Kay, L.E. Cross-correlated relaxation enhanced 1H-13C NMR spectroscopy of methyl groups in very high molecular weight proteins and protein complexes. J. Am. Chem. Soc. 125, 1042010428 (2003).
  58. Goto, N.K., Gardner, K.H., Mueller, G.A., Willis, R.C. & Kay, L.E. A robust and cost-effective method for the production of Val, Leu, Ile (δ1) methyl-protonated 15N-, 13C-, 2H-labeled proteins. J. Biomol. NMR 13, 369374 (1999).
  59. Amero, C. et al. Fast two-dimensional NMR spectroscopy of high molecular weight protein assemblies. J. Am. Chem. Soc. 131, 34483449 (2009).
  60. Maruyama, Y. et al. Human Gene and Protein Database (HGPD): a novel database presenting a large quantity of experiment-based results in human proteomics. Nucleic Acids Res. 37, D762D766 (2009).
  61. Maruyama, Y. et al. HGPD: Human Gene and Protein Database, 2012 update. Nucleic Acids Res. 40, D924D929 (2012).
  62. Schanda, P., Kupce, E. & Brutscher, B. SOFAST-HMQC experiments for recording two-dimensional heteronuclear correlation spectra of proteins within a few seconds. J. Biomol. NMR 33, 199211 (2005).
  63. Selenko, P. et al. In situ observation of protein phosphorylation by high-resolution NMR spectroscopy. Nat. Struct. Mol. Biol. 15, 321329 (2008).

Download references

Author information

Affiliations

  1. Research and Development Department, Japan Biological Informatics Consortium, Tokyo, Japan.

    • Yuji Tokunaga
  2. Graduate School of Pharmaceutical Sciences, University of Tokyo, Tokyo, Japan.

    • Yuji Tokunaga &
    • Ichio Shimada
  3. Biomedicinal Information Research Center, National Institute of Advanced Industrial Science and Technology, Tokyo, Japan.

    • Koh Takeuchi,
    • Hideo Takahashi &
    • Ichio Shimada
  4. Graduate School of Medical Life Science, Yokohama City University, Kanagawa, Japan.

    • Hideo Takahashi

Contributions

Y.T., K.T., H.T. and I.S. conceived the project. Y.T. performed the experiments. Y.T., K.T., H.T. and I.S. wrote the manuscript.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Supplementary information

Supplementary Figures

  1. Supplementary Figure 1: Characterization of ​p38α and the model substrates. (206 KB)

    (a) Phos-tagTM SDS-PAGE of ​p38α, which is dually phosphorylated by ​MKK6DD (lane 2). The sample was then dephosphorylated by either ​HePTP, a phosphatase that is specific to phospho-Tyr182 (lanes 3 and 4), or ​PPM1A, a phosphatase that is specific to phospho-Thr180 (lanes 7 and 8). In lanes 5, 6, 9, and 10, the ​p38α-2P was treated with both of the phosphtases in different orders. The mobility of ​p38α-2P treated with ​HePTP increased (​p38α-1PT), indicating the dephosphorylation of Tyr182 (lanes 3 and 4). Thr180, however, remained phosphorylated, as the mobility is still slower than unphosphorylated ​p38α. Further treatment of ​p38α-1PT with ​PPM1A led to the complete regression to unphosphorylated ​p38α (lanes 5 and 6). This indicated that ​p38α was successfully dually phosphorylated by ​MKK6DD. Although ​p38α-2P first treated with ​PPM1A exhibited almost the same mobility as ​p38α-2P (​p38α-1PY; lanes 7 and 8), the subsequent treatment with ​HePTP resulted in the same mobility as unphosphorylated ​p38α. It seems that the mobilities of ​p38α-1PY and ​p38α-2P are indiscernible. The details of the reaction conditions are described in the Online Methods. (b-i) Characterization of the phosphorylation states of ​p38α based on the spectral pattern of the Thr-γ2 region of the 1H-13C HSQC spectra. Panels (b) to (e) show the spectra of ​p38α, ​p38α-2P, ​p38α-1PT, and ​p38α-1PY, respectively. The ​p38α-1PT and ​p38α-1PY proteins were prepared by treating ​p38α-2P with ​HePTP and ​PPM1A, respectively. Methyl resonances with distinctive chemical shifts in the respective phosphorylation states are highlighted by the blue rectangles. Panels (f) to (i) are the spectra of ​p38α phosphorylated with 1/70 mol. eq. of ​MKK6DD at 14 °C for the indicated times. A 20 hour reaction rendered most of the ​p38α dually phosphorylated. (j) Phos-tagTM SDS-PAGE of ​MK2 (47–400) phosphorylated by ​p38α-2P. Lane 1 is the reference which is not treated with ​p38α-2P, and lanes 2 to 7 show the timecourse of the phosphorylation of 5 μM ​MK2 (47–400) by 0.5 μM ​p38α-2P in the presence of 2 mM ​ATP at 25 °C. The time-dependent, discrete decrease in the mobility is indicative of the multisite phosphorylation of ​MK2 (47–400) by ​p38α-2P. (k) The retention volumes in the SEC analyses of ​p38α-2P in the absence and presence of 1 mM ​ATP. Samples were prepared as 240 μL solution of 25 μM ​p38α and were analyzed by chromatography on a Superdex 200 GL 10 300 column connected to AKTA Explorer 10S (GE Healthcare) at 4 °C. The retention volumes and the error bars are the averages and the standard deviations of five repeated runs. The larger error bar in the presence of ​ATP is due to the rough baseline caused by the high background absorbance at 280 nm by ​ATP. Nevertheless, the retention volume in the presence of ​ATP is larger than that in the absence of ​ATP, suggesting that the conformation of ​p38α-2P is more compact (i.e. “closed”) when ​ATP is loaded. (l) Competition experiment between the 334D-peptide and ​MK2 (47-400). Western blot of ​p38α-2P, detected with 1:1000 dilution of the anti-p38 monoclonal antibody. The N-terminally GB1-fused 334D-peptide was immobilized on IgG Sepharose beads, and then ​p38α-2P was further immobilized on them (input; lane 6). After washing the beads, ​p38α-2P was eluted with the longer ​MK2 fragment (47–400) (0.3, 0.6, and 1.2 μM; lanes 2–4). The intensity of the ​p38α-2P band increased in proportion to the ​MK2 concentration, which indicates that the longer ​MK2 fragment competes with the GB1-334D-peptide for the same site on ​p38α-2P. The dissociation of ​p38α-2P from the beads by washing is negligible, as the band is very weak when the elution buffer without the longer ​MK2 fragment is used (lane 1). The non-specific binding of ​p38α-2P on the beads is also negligible, as shown in the experiment using GB1 instead of the GB1-334D-peptide (lane 5). The detailed experimental conditions are provided in the Online Methods. (m) Phosphorylations of the 334D-peptide and the 334-peptide by ​p38α-2P were monitored by time-dependent increase of the phospho-334D-peptide (red circles) and the phospho-334-peptide (cyan squares). The reactions were tracked by successive measurements of 1H-15N SOFAST-HMQC spectra. The concentration of the phosphorylated product was estimated from the intensity of the phospho-Thr334 resonance for each peptide. The peptides (100 μM) were phosphorylated by 20 nM of ​p38α-2P, in the presence of 2 mM ​ATP at 25 °C. The error bars correspond to the level of noise in each spectrum.

  2. Supplementary Figure 2: Effect of phosphorylation and ​ATP titration on ​p38α. (390 KB)

    (a) Overlay of the annotated ​Leu/​Val methyl 1H-13C SOFAST-HMQC spectra of ​p38α-2P in the absence (black) and presence (red) of 5 mM ATP analog. Large chemical shift changes are indicated by blue arrows. (b) Left, center: Overlay of the ​Ile(δ1) and ​Leu/​Val methyl 1H-13C SOFAST-HMQC spectra of 20 μM [ILV-methyl] ​p38α-2P in the presence (red) and absence (black) of 5 mM ​ATP. Large chemical shift changes in the ​Ile(δ1) region are indicated by blue arrows. Spectra were measured with an 800 MHz 1H resonance frequency spectrometer at 10 °C. Right: Methyl sites that exhibited chemical shift changes larger than the linewidth or were not assigned only in the ATP-bound state are indicated as red spheres on the crystal structure (PDB code: 1A9U51). The ATP-binding site is highlighted in the orange oval. (c) Overlay of the annotated ​Leu/​Val methyl 1H-13C SOFAST-HMQC spectra of ​p38α-2P (blue) and unphosphorylated ​p38α (black) in the apo state. (d) Comparison of the ​Ile(δ1) and ​Leu/​Val methyl regions of 1H-13C HMQC spectra of 80 μM [ILVM-methyl] unphosphorylated ​p38α, in the presence (green) and absence (red) of 4 mM ​ATP. Spectra were measured at 800 MHz (1H frequency) at 25 °C. For the methyl resonances of ​Leu and ​Val residues, only those dispersed in the spectra are labeled, for clarification. (e) ​Met methyl regions of 1H-13C HMQC spectra of unphosphorylated ​p38α titrated with various concentration of ​ATP. The resonance of Met109, which was traced during the titration to determine the affinity of unphosphorylated ​p38α for ​ATP, is expanded as an inset, in which all of the titration points are overlaid. The dissociation constant of unphosphorylated ​p38α for ​ATP was estimated to be larger than 15 mM, based on the CSPs of the Met109 methyl 1H and 13C resonances. Since the chemical shift changes did not reach a plateau at the highest ​ATP concentration (15 mM), a unique dissociation constant was not obtained.

  3. Supplementary Figure 3: Changes in 1H-15N TROSY spectra of ​p38α upon dual phosphorylation. (305 KB)

    Comparison of 1H-15N TROSY spectra of unphosphorylated ​p38α (​p38α-0P; black) and ​p38α-2P (red) in the absence of ​ATP. The regions containing the resonances of Met179, Thr180, Gly181, Tyr182, and Val183, which are located in the activation loop, are expanded to indicate that these resonances disappeared upon dual phosphorylation. In the lower left panel, the signal from ​p38α-2P, which overlaps with the Met179 signal from ​p38α-0P, does not belong to Met179 and is presumably from the unassigned resonance (n.a.) below.

  4. Supplementary Figure 4: Effects of substrate binding on ​p38α NMR spectra. (325 KB)

    (a) Overlay of the annotated ​Leu/​Val methyl 1H-13C SOFAST-HMQC spectra of ​p38α-2P without the ATP analog, in the absence (black) and presence (red) of an equimolar amount of the 334D-peptide. Large chemical shift changes are indicated by blue arrows. (b) The same as (a), except that ​p38α-2P is in complex with the ATP analog.

  5. Supplementary Figure 5: Interaction between the substrate phosphoacceptor site and ​p38α-2P upon ​ATP loading. (359 KB)

    (a) Extensive interaction between the 334D-peptide and ​p38α-2P upon the ATP analog loading. Comparison of the 1H-15N TROSY spectra of a stoichiometric complex of 180 μM [U-15N] 334D-peptide and unlabeled ​p38α-2P, in the absence (black) and presence (red) of 5 mM ATP analog. Several resonances, including that of the phosphoacceptor residue, Thr334, lost intensity when the ATP analog was added, reflecting the increase in the transverse relaxation rate. This would be explained by the increase in the effective local rotational correlation times due to the binding of Thr334 to the active site of ​p38α-2P and/or the generation of chemical exchange with relatively long (msec – μsec) time scale associating with the binding. This suggested that an extensive interaction is formed by the loading of the ATP analog on ​p38α-2P. (b) Spectral perturbations induced upon the addition of the T334A mutants of 334D-peptide to ​p38α-2P. Comparison of the regions of the 1H-13C HMQC spectra of 40 μM [ILV-methyl] ​p38α-2P that contain the resonance of either Ile229 or Ile259, which is located around the P+1 site, in the absence (black) and presence (red) of an equimolar amount of the T334A mutant of the 334D-peptide. The panels on the left and right are the spectra in the absence of presence of ​ATP (not analog; 5 mM). The addition of the unlabeled peptide induced CSPs in the P+1 site resonances only in the presence of ​ATP. The addition of the peptide induced CSPs to the P+1 site resonances of ​p38α-2P in the presence of ​ATP. Spectra were recorded at 10 °C, with an 800 MHz 1H resonance frequency spectrometer. (c, d) Comparison of the ​Ile-δ1 regions of 1H-13C HMQC spectra of 40 μM [ILV-methyl] ​p38α-2P, in the absence (black) and presence (red) of a 10-fold molar excess of the 334-peptide without (c) and with (d) a saturating amount of the ATP analog (5 mM). Regions containing the resonances of Ile229 and Ile259, which are located near the P+1 site, are enlarged. In the presence of the ATP analog, the resonance from Ile229 disappeared upon the addition of the 334-peptide. (e) Spectral perturbations induced upon the addition of the T334A mutants 334-peptide to ​p38α-2P. The experimental conditions are the same as (b), except that the titrated peptide was a 10-fold molar excess of the unlabeled 334-peptide with the T334A mutation. (f, g) Extensive interaction between the 334-peptide and ​p38α-2P induced by the ATP analog. The methyl region of the one dimensional (1D) proton NMR spectrum of 100 μM unlabeled 334-peptide in the free form (f, blue) was compared to that with 50 μM ​p38α-2P in the absence (g, black) and presence of the ATP analog (4 mM; g, red). The signals from the 334-peptide were not significantly affected by the presence of ​p38α-2P itself, while the broad signals derived from ​p38α-2P are also observed (g, black). However, considerable broadening of the 334-peptide signals were observed upon the addition of 4 mM ATP analog (g, red), indicating that the transverse relaxation rates of the methyl proton resonances of the 334-peptide increased. This would result from the increased effective rotational correlation time of the 334-peptide upon the binding to ​p38α-2P, as well as the relaxation enhancement due to the exchange between free and bound states with distinct chemical shifts. Thus, the 334-peptide interacts with ​p38α-2P, which is loaded with the ATP analog.

  6. Supplementary Figure 6: Allosteric effects induced by the docking interaction to ​p38α-2P. (326 KB)

    (a) Comparison of the ILV-methyl regions of the 1H-13C HMQC spectra of ​p38α-2P in the absence (black) and presence (magenta) of a 1.1 molar equivalent of the D-peptide. Spectra were recorded in the presence of 5 mM ATP analog. Regions containing the resonances of Val89, located in the ATP-binding site, and Ile259, located in the P+1 site, are shown in orange and cyan squares, respectively. (b) Mapping of the methyl groups that showed CSPs larger than 0.02 ppm, upon the addition of the D-peptide in the presence of 5 mM ATP analog, shown in red, while those that are not assigned only in the presence of the ATP analog are shown in black. The docking sequence of ​MK2 is shown as purple ribbon. The structure figure was generated from the X-ray structure of ​p38α in complex with the ​MK2 docking-peptide (PDB code: 2OKR29).

  7. Supplementary Figure 7: Enhancements of catalytic steps of ​p38α by the docking interaction. (280 KB)

    (a) Determination of the affinity of ​p38α-2P to the ATP analog. Ile84-δ1 signals in the 1H-13C HMQC spectra of 40 μM [ILV-methyl] ​p38α-2P, without (upper) and with (lower) a 1.1 molar excess of the D-peptide. Both samples contained 200 μM of the ATP analog. High field (upper right) and low field resonances (lower left) are from the ATP analog unbound and bound states, respectively. 1D slices in the 1H dimension are overlaid on the 2D signals. The intensity of the resonance from the ATP analog-bound state becomes stronger in the presence of the D-peptide. (b) Determination of the affinity of ​p38α-2P to ​ATP. Left; Expanded regions of the spectra for the probe resonances. The upper and lower rows correspond to the conditions without and with a 1.1 molar equivalent of the D-peptide, respectively. From left to right, the conditions represent 5 mM ATP analog, the mixture of 2.5 mM ATP analog and 2.5 mM ​ATP, and 5 mM ​ATP, respectively. Right; The intensities of the probe resonances with the different concentrations of ​ATP and the ATP analog. The intensities of the resonances in each condition were normalized to the reference spectra for the ATP-bound form. The upper and lower panels are the data in the absence and presence of the D-peptide, respectively. The intensities for the ATP and ATP analog bound resonances were plotted as the orange squares and green diamonds, respectively, and were used to determine the affinity of ​ATP, as described in the Supplemental Materials and Methods. The error bars correspond to the level of noise in each spectrum. (c) Titration of the 334-peptide to ​p38α-2P. Overlays of the ​Ile-δ1 regions of 1H-13C HMQC spectra of 40 μM [ILV-methyl] ​p38α-2P with different concentrations of the 334-peptide, in the absence (left) and presence (right) of a 1.1 molar equivalent of the D-peptide. Black, orange, and red spectra correspond to 0, 120, and 400 μM concentrations of the 334-peptide, respectively. Only three spectra of the five titration points are shown for visual clarity. The resonances of Ile259 near the P+1 site are indicated by the cyan rectangles, and are enlarged in the insets. The dose-dependent CSPs of the Ile259 methyl resonance were used to determining the affinity of ATP-loaded ​p38α-2P for the 334-peptide. (d) Phosphorylation of the 334-peptide by ​p38α-2P. Left; Expanded region of the phosphorylated Thr334 resonance in the 1H-15N HSQC spectra of 100 μM [U-15N] 334-peptide, before (black) and after (red) phosphorylation by ​p38α-2P. Only after reaction, the resonance of the phosphorylated Thr334 is observed, with the low-field chemical shifts characteristic of phosphorylated residues64. Right; Time-dependent increase of the phosphorylated 334-peptide produced by ​p38α-2P, in the absence (black) and presence (magenta) of the D-peptide. The concentration of the product was estimated from the intensity of the phospho-Thr334 resonance. The detailed experimental conditions are provided in the Online Methods.

  8. Supplementary Figure 8: Determination of the affinities of the docking fragments for ​p38α-2P. (230 KB)

    (a-d) NMR titration experiments of the docking fragment of ​MEF2A to ​p38α-2P, in the absence and presence of the ATP analog. (a, b) Comparison of the 1H-13C HMQC spectra (​Ile-δ1 region) of 20 μM [ILV-methyl] ​p38α-2P with different concentrations of the ​MEF2A docking fragment in the absence (a) and presence (b) of 4 mM ATP analog. The arrows trace the chemical shift changes of Ile116, which exhibited the largest CSP values. (c, d) CSPs induced in the methyl resonances of Val83, Ile116, Val117, Val127, Ile131, Ile134, and Val158, which are located in and near the docking site, by titration of the ​MEF2A docking fragment. The dissociation constants are indicated in the plots with standard deviation estimated from fitting errors. (e, f) ITC experiments of the D-peptide to ​p38α-2P in the absence (e) and presence (f) of the ATP analog. The dissociation constants are indicated in the plots. The detailed experimental conditions are provided in the Online Methods.

PDF files

  1. Supplementary Text and Figures (2,238 KB)

    Supplementary Figures 1–8 and Supplementary Table 1

Additional data