Pseudophosphatases as Regulators of MAPK Signaling

Abstract

Mitogen-activated protein kinase (MAPK) signaling pathways are highly conserved regulators of eukaryotic cell function. These enzymes regulate many biological processes, including the cell cycle, apoptosis, differentiation, protein biosynthesis, and oncogenesis; therefore, tight control of the activity of MAPK is critical. Kinases and phosphatases are well established as MAPK activators and inhibitors, respectively. Kinases phosphorylate MAPKs, initiating and controlling the amplitude of the activation. In contrast, MAPK phosphatases (MKPs) dephosphorylate MAPKs, downregulating and controlling the duration of the signal. In addition, within the past decade, pseudoenzymes of these two families, pseudokinases and pseudophosphatases, have emerged as bona fide signaling regulators. This review discusses the role of pseudophosphatases in MAPK signaling, highlighting the function of phosphoserine/threonine/tyrosine-interacting protein (STYX) and TAK1-binding protein (TAB 1) in regulating MAPKs. Finally, a new paradigm is considered for this well-studied cellular pathway, and signal transduction pathways in general.

Keywords: MAPK phosphatase (MKP); MAPK phosphoserine/threonine/tyrosine-binding protein (MK-STYX); TAK1-binding protein (TAB 1); dual-specificity phosphatases (DUSPs); extracellular signal-regulated kinase (ERK); kinase; mitogen-activated protein kinase (MAPK); phosphoserine/threonine/tyrosine-interacting protein (STYX); protein tyrosine phosphatases (PTPs); pseudophosphatase.

Figure 1

Roles of pseudophosphatases as signaling molecules. There are several different mechanisms that pseudophosphatases utilize to regulate signaling pathways. They may serve as an integrator and integrate multiple signaling pathways (top); a molecular anchor by binding a target protein and restricting its subcellular localization (right); a competitor for substrate binding (bottom); and as a modulator of enzymes such as an active phosphatase by either promoting or inhibiting the activity of the phosphatase (left) (adapted image) [20,21,25].

Figure 2

MKPs regulate MAPKs. (A) The domains of a canonical MKP. MKPs have two domains, a C-terminal catalytic DUSP domain that dephosphorylates MAPKs and an N-terminal CH2 domain that binds the MAPK-binding domain. The CH2 domain recognizes and docks the specific MAPK substrate, stabilizing it for dephosphorylation. (B) The DUSP domain of MKPs dephosphorylates T and Y residues in the MAPK activation loop (pTXpY), downregulating MAPK activation. The CH2 domain of the MKP binds the MAPK, inducing a conformational change, which activates and increases the catalytic activity of the MKP DUSP domain. This promotes the DUSP active site to dephosphorylate the MAPK residues, thereby inactivating the MAPK.

Figure 3

Bioinformatics structural analysis of pseudophosphatase MK-STYX. (A) The atypical domains of MK-STYX. MK-STYX has a C-terminal DUSP domain and an N-terminal CH2 domain, similar to its active homologs. A mutation in the active site of the DUSP domain from HCX₅R to FSX₅R renders MK-STYX catalytically inactive. In addition, the presumed kinase-interacting motif (KIM) within the CH2 domain of MK-STYX lacks consecutive arginine residues, required for MAPK binding. Therefore, MK-STYX may not bind MAPKs. (B) The KIM analysis of MK-STYX and active homologs, MKP-1 and MKP-3. The KIM regulates substrate specificity and substrate docking in CH2 domain of classical MKPs. The KIM requires 2 to 3 consecutive arginine residues to create a consecutive positively charged region that interacts with the negatively charged aspartic acid residues on the target MAPK substrate for docking of the MAPK. Unlike its active homologs, MKP-1 and MKP-3, the KIM consensus motif of MK-STYX contains only a single arginine. This may explain why MK-STYX does not bind and regulate MAPKs in the same manner as typical MKPs. The sequence logo was made using WebLogo 3.7.4 [59] and has a 2.0 bit scale (positive residues are shown in blue, neutral residues are shown in black, no negative residues are present). (C) Computational mutagenesis of the KIM of MK-STYX affects the size and shape of a predicted binding pocket. To probe for possible explanations behind why MK-STYX lacks the ability to regulate MAPK signaling, a predictive model of the macromolecular structure of MK-STYX was generated by Iterative Threading ASSEmbly Refinement (I-TASSER) [60]. The best I-TASSER model (validated through MolProbity [61] with a score of 3.03) was refined using DeepRefiner [62]. The best quality refined model (predicted global quality score of 0.142 and MolProbity score of 2.14) was mutated using the mutagenesis function in the PyMOL Molecular Graphics System 4.6.0 [63] to create the V53R mutant, which restores the consecutive arginine residues in the KIM. The structures of both wild-type (WT) MK-STYX and the V53R mutant were submitted to Pocket Cavity Search Application (POCASA) [64] to probe the surface of the protein and predict possible binding sites (1.0 Å grid size and 2 Å probe radius). When the predicted pockets were compared between the WT MK-STYX and the V53R mutant, the V53R mutant displayed a smaller and differently shaped predicted binding pocket in the area of the KIM (rank 3 out of 13 predicted pockets, decrease in volume from 98 to 96, and decrease in volume depth value of 271 to 261). The area of change (shown by black arrow) was in the immediate area of the mutated residue (WT residue shown in yellow and V53R mutation in red). In addition, the WT model was submitted to Missense3D [65], which also predicted that the V53R mutation altered a cavity and leads to the contraction of cavity volume (predicted volume contraction of 89.424 ų). For further validation, the WT model was submitted again, to predict the effect of an S246C mutation in the signature active site motif; there was no predicted structural damage (data not shown). This was expected because most pseudophosphatases maintain their three-dimensional fold. This may indicate that while MK-STYX does not bind MAPKs at this site; this pocket may allow MK-STYX to bind a novel set of binding partners. Techniques used to produce the data for this study were learned through the Malate Dehydrogenase CUREs Community Workshop on Virtual CUREs and UREs (https://mdh-cures-community.squarespace.com/virtual-cures-and-ures (accessed on 27 June 2021)), supported by a grant from the National Science Foundation. NSF-1726932 EHR-IUSE, Principal Investigator Ellis Bell [66].

Figure 4

Pseudophosphatase STYX regulates ERK1/2 activity. (A) Schematic diagram of the pseudophosphatase STYX. STYX has a DUSP domain, which is catalytically inactive due to a C120G mutation in the active site motif (HCX₅R). (B) Despite lacking catalytic activity and a CH2 domain, STYX competes with DUSP-4/MKP-2 to regulate the nucleocytoplasmic shuttling of ERK1/2. STYX binds ERK and serves as a nuclear anchor, inhibiting the nuclear export of ERK.