Anchoring TRP to the INAD macromolecular complex requires the last 14 residues in its carboxyl terminus

Drosophila transient-receptor-potential (TRP) is a Ca²⁺ channel responsible for the light-dependent depolarization of photoreceptors. TRP is anchored to a macromolecular complex by tethering to inactivation-no-afterpotential D (INAD). We previously reported that INAD associated with the carboxyl tail of TRP via its third post-synaptic density protein 95, discs-large, zonular occludens-1 domain. In this paper, we further explored the molecular basis of the INAD interaction and demonstrated the requirement of the last 14 residues of TRP, with the critical contribution of Gly1262, Val1266, Trp1274, and Leu1275. We also revealed by pull-down assays that the last 14 residues of TRP comprised the minimal sequence that competes with the endogenous TRP from fly extracts, leading to the co-purification of a partial INAD complex containing INAD, no-receptor-potential A, and eye-protein kinase C (PKC). Eye-PKC is critical for the negative regulation of the visual signaling and was shown to phosphorylate TRP in vivo . To uncover the substrates of eye-PKC in the INAD complex, we designed a complex-dependent eye-PKC assay, which utilized endogenous INAD complexes isolated from flies. We demonstrate that activated eye-PKC phosphorylates INAD, TRP but not no-receptor-potential A. Moreover, phosphorylation of TRP is dependent on the presence of both eye-PKC and INAD. Together, these findings indicate that stable kinase-containing protein complexes may be isolated by pull-down assays, and used in this modified kinase assay to investigate phosphorylation of the proteins in the complex. We conclude that TRP associates with INAD via its last 14 residues to facilitate its regulation by eye-PKC that fine-tunes the visual signaling.     

TRP and the PDZ protein, INAD, form the core complex required for retention of the signalplex in Drosophila photoreceptor cells

The light response in Drosophila photoreceptor cells is mediated by a series of proteins that assemble into a macromolecular complex referred to as the signalplex. The central player in the signalplex is inactivation no afterpotential D (INAD), a protein consisting of a tandem array of five PDZ domains. At least seven proteins bind INAD, including the transient receptor potential (TRP) channel, which depends on INAD for localization to the phototransducing organelle, the rhabdomere. However, the determinants required for localization of INAD are not known. In this work, we showed that INAD was required for retention rather than targeting of TRP to the rhabdomeres. In addition, we demonstrated that TRP bound to INAD through the COOH terminus, and this interaction was required for localization of INAD. Other proteins that depend on INAD for localization, phospholipase C and protein kinase C, also mislocalized. However, elimination of any other member of the signalplex had no impact on the spatial distribution of INAD. A direct interaction between TRP and INAD did not appear to have a role in the photoresponse independent of localization of multiple signaling components. Rather, the primary function of the TRP/ INAD complex is to form the core unit required for localization of the signalplex to the rhabdomeres.     

The second PDZ domain of INAD is a type I domain involved in binding to eye protein kinase C. Mutational analysis and naturally occurring variants

INAD is a scaffolding protein containing five PSD95/dlg/zonular occludens-1 (PDZ) domains that tether NORPA (phospholipase Cbeta(4)), the TRP calcium channel, and eye-PKC in Drosophila photoreceptors. We previously showed that eye-PKC interacted with the second PDZ domain (PDZ2) of INAD. Sequence comparison with a prototypical type I PDZ domain predicts that PDZ2 is the best candidate among the five PDZ domains to recognize eye-PKC that contains a type I PDZ ligand, Ile-Thr-Ile-Ile, at its carboxyl terminus. Replacement of Ile(-3) in eye-PKC with charged residues resulted in a drastic reduction of the PDZ2 interaction. Substitution of a conserved His with Arg at the second alpha-helix of PDZ2 led to a reduced binding; however, a Leu replacement resulted in an enhanced eye-PKC association. We isolated and sequenced the InaD gene. The coding sequence of InaD contains nine exons spanning 3 kilobases. Translation of coding sequences from three wild-type alleles revealed three SNPs affecting residues, 282, 319, and 333 of INAD. These polymorphisms are localized in PDZ2. Interestingly, we found two of three PDZ2 variants displayed a greater affinity for eye-PKC. In summary, we evaluated the molecular basis of the eye-PKC and PDZ2 association by mutational analysis and concluded that PDZ2 of INAD is a type I domain important for the eye-PKC interaction.     

Regulation of dopamine D-receptor activation in vivo by protein phosphatase 2B (calcineurin)

Mice lacking dopamine D2 receptors exhibit a significantly decreased agonist-promoted forebrain neocortical D1 receptor activation that occurs without changes in D1 receptor expression levels. This raises the possibility that, in brains of D2 mutants, a substantial portion of D1 receptors are uncoupled from their G protein, a phenomenon known as receptor desensitization. To test this, we examined D1-agonist-stimulated [35S]GTPgammaS binding (in the presence and absence of protein phosphatase inhibitors) and cAMP production (in the presence and absence of pertussis toxin) in forebrain neocortical tissues of wild-type mice and D2-receptor mutants. These studies revealed a decreased agonist-stimulated G-protein activation in D2 mutants. Moreover, whereas protein phosphatase 1/2A (PP1/2A) and 2B (PP2B) inhibitors decrease [35S]GTPgammaS binding in a concentration-dependent manner in wild type, they have either no (PP2B) or only partial (PP1/2A) effects in D2 mutants. Furthermore, for D2 mutants, immunoprecipitation experiments revealed increased basal and D1-agonist-stimulated phosphorylation of D1-receptor proteins at serine residues. Finally, D1 immunoprecipitates of both wild type and D2 mutants also contain protein kinase A (PKA) and PP2B immunoreactivities. In D2 mutants, however, the catalytic activity of the immunoprecipitated PP2B is abolished. These data indicate that neocortical D1 receptors are physically linked to PKA and PP2B and that the increased phosphorylation of D1 receptors in brains of D2 mutants is due to defective dephosphorylation of the receptor rather than increased kinase-mediated phosphorylation.     

Phosphorylation of hUPF1 induces formation of mRNA surveillance complexes containing hSMG-5 and hSMG-7

Eukaryotic mRNAs containing premature termination codons (PTCs) are degraded by a process known as nonsense-mediated mRNA decay (NMD). NMD has been suggested to require the recognition of PTC by an mRNA surveillance complex containing UPF1/SMG-2. In multicellular organisms, UPF1/SMG-2 is a phosphoprotein, and its phosphorylation contributes to NMD. Here we show that phosphorylated hUPF1, the human ortholog of UPF1/SMG-2, forms a complex with human orthologs of the C. elegans NMD proteins SMG-5 and SMG-7. The complex also associates with protein phosphatase 2A (PP2A), resulting in dephosphorylation of hUPF1. Overexpression of hSMG-5 mutants that retain interaction with P-hUPF1 but which cannot induce its dephosphorylation impair NMD, suggesting that NMD requires P-hUPF1 dephosphorylation. We also show that P-hUPF1 forms distinct complexes containing different isoforms of hUPF3A. We propose that sequential phosphorylation and dephosphorylation of hUPF1 by hSMG-1 and PP2A, respectively, contribute to the remodeling of the mRNA surveillance complex.     

Regulation of Metabotropic Glutamate Receptor 7 (mGluR7) Internalization and Surface Expression by Ser/Thr Protein Phosphatase 1

The metabotropic glutamate receptor type 7 (mGluR7) is the predominant group III mGluR in the presynaptic active zone, where it serves as an autoreceptor to inhibit neurotransmitter release. Our previous studies show that PKC phosphorylation of mGluR7 on Ser-862 is a key mechanism controlling constitutive and activity-dependent surface expression of mGluR7 by regulating a competitive interaction of calmodulin and protein interacting with C kinase (PICK1). As receptor phosphorylation and dephosphorylation are tightly coordinated through the precise action of protein kinases and phosphatases, dephosphorylation by phosphatases is likely to play an active role in governing the activity-dependent or agonist-induced changes in mGluR7 receptor surface expression. In the present study, we find that the serine/threonine protein phosphatase 1 (PP1) has a crucial role in the constitutive and agonist-induced dephosphorylation of Ser-862 on mGluR7. Treatment of neurons with PP1 inhibitors leads to a robust increase in Ser-862 phosphorylation and increased surface expression of mGluR7. In addition, Ser-862 phosphorylation of both mGluR7a and mGluR7b is a target of PP1. Interestingly, agonist-induced dephosphorylation of mGluR7 is regulated by PP1, whereas NMDA-mediated activity-induced dephosphorylation is not, illustrating there are multiple signaling pathways that affect receptor phosphorylation and trafficking. Importantly, PP1γ1 regulates agonist-dependent Ser-862 dephosphorylation and surface expression of mGluR7.     

Interaction and feedback regulation between STK15/BTAK/Aurora-A kinase and protein phosphatase 1 through mitotic cell division cycle

STK15 is an Aurora/Ipl-1 related serine/threonine kinase that is associated with centrosomes and induces aneuploidy when overexpressed in mammalian cells. It is well known that phosphorylation and dephosphorylation of kinases are important for regulation of their activity. But mechanisms by which STK15 activity is regulated have not been elucidated. We report that STK15 contains two functional binding sites for protein phosphatase type 1 (PP1), and the binding of these proteins is cell cycle-regulated peaking at mitosis. Activated STK15 at mitosis phosphorylates PP1 and inhibits PP1 activity in vitro. In vivo, PP1 activity co-immunoprecipitated with STK15 is also reduced. These data indicate that STK15 inhibits PP1 activity during mitosis. Also, PP1 is shown to dephosphorylate active STK15 and abolish its activity in vitro. Furthermore, we show that non-binding mutants of STK15 for PP1 are superphosphorylated, but their kinase activities are markedly reduced. Cells transfected with these non-binding mutants manifest aberrant chromosome alignment during mitosis. Our results suggest that a feedback regulation through phosphorylation/dephosphorylation events between STK15 kinase and PP1 phosphatase operates through the cell cycle. Deregulation of this balance may contribute to anomalous segregation of chromosomes during mitotic progression of cancer cells.     

Dephosphorylation specificities of protein phosphatase for cardiac troponin I, troponin T, and sites within troponin T

Protein dephosphorylation by protein phosphatase 1 (PP1), acting in concert with protein kinase C (PKC) and protein kinase A (PKA), is a pivotal regulatory mechanism of protein phosphorylation. Isolated rat cardiac myofibrils phosphorylated by PKC/PKA and dephosphorylated by PP1 were used in determining dephosphorylation specificities, Ca(2+)-stimulated Mg(2+)ATPase activities, and Ca(2+) sensitivities. In reconstituted troponin (Tn) complex, PP1 displayed distinct substrate specificity in dephosphorylation of TnT preferentially to TnI, in vitro. In situ phosphorylation of cardiomyocytes with calyculin A, a protein phosphatase inhibitor, resulted in an increase in the phosphorylation stiochiometry of TnT (0.3 to 0.5 (67%)), TnI (2.6 to 3.6 (38%)), and MLC2 (0.4 to 1.7 (325%)). These results further confirmed that though MLC2 is the preferred target substrate for protein phosphatase in the thick filament, the Tn complex (TnI and TnT) from thin filament and C-protein in the thick filament are also protein phosphatase substrates. Our in vitro dephosphorylation experiments revealed that while PP1 differentially dephosphorylated within TnT at multiple sites, TnI was uniformly dephosphorylated. Phosphopeptide maps from the in vitro experiments show that TnT phosphopeptides at spots 4A and 4B are much more resistant to PP1 dephosphorylation than other TnT phosphopeptides. Mg(2+)ATPase assays of myofibrils phosphorylated by PKC/PKA and dephosphorylated by PP1 delineated that while PKC and PKA phosphorylation decreased the Ca(2+)-stimulated Mg(2+)ATPase activities, dephosphorylation antagonistically restored it. PKC and PKA phosphorylation decreased Ca(2+) sensitivity to 3.6 microM and 5.0 microM respectively. However, dephosphorylation restored the Mg(2+)ATPase activity of PKC (99%) and PKA (95%), along with the Ca(2+) sensitivities (3.3 microM and 3.0 microM, respectively).     

A vitamin D receptor-Ser/Thr phosphatase-p70 S6 kinase complex and modulation of its enzymatic activities by the ligand

We provide evidence of a cross-talk between nuclear receptor and Ser/Thr protein phosphatases and show that vitamin D receptor (VDR) interacts with the catalytic subunit of protein phosphatases, PP1c and PP2Ac, and induces their enzymatic activity in a ligand-dependent manner. PP1c specifically interacts with VDR but not retinoic acid receptor alpha and retinoid X receptor alpha in yeast. Although VDR-PP1c and VDR-PP2Ac interaction is ligand-independent in vivo, 1alpha,25-dihydroxy-vitamin D(3) induces VDR-associated phosphatase activity. Further, VDR modulation of PP1c/PP2Ac activity results in a rapid and specific dephosphorylation and inactivation of their substrate, p70 S6 kinase (p70(S6k)). Finally, we demonstrate that the endogenous VDR, PP1c or PP2Ac, and p70(S6k) are present in a ternary complex in vivo, and the interaction of p70(S6k) with the VDR-PP complex is modulated by the phosphorylation state of the kinase. Since p70(S6k) is essential for G(1)-S transition, our results provide a molecular basis of 1alpha,25-dihydroxyvitamin D(3)-induced G(1) block in colon cancer cells.     

S100 proteins modulate protein phosphatase 5 function: a link between CA2+ signal transduction and protein dephosphorylation

PP5 is a unique member of serine/threonine phosphatases comprising a regulatory tetratricopeptide repeat (TPR) domain and functions in signaling pathways that control many cellular responses. We reported previously that Ca(2+)/S100 proteins directly associate with several TPR-containing proteins and lead to dissociate the interactions of TPR proteins with their client proteins. Here, we identified protein phosphatase 5 (PP5) as a novel target of S100 proteins. In vitro binding studies demonstrated that S100A1, S100A2, S100A6, and S100B proteins specifically interact with PP5-TPR and inhibited the PP5-Hsp90 interaction. In addition, the S100 proteins activate PP5 by using a synthetic phosphopeptide and a physiological protein substrate, Tau. Overexpression of S100A1 in COS-7 cells induced dephosphorylation of Tau. However, S100A1 and permanently active S100P inhibited the apoptosis signal-regulating kinase 1 (ASK1) and PP5 interaction, resulting the inhibition of dephosphorylation of phospho-ASK1 by PP5. The association of the S100 proteins with PP5 provides a Ca(2+)-dependent regulatory mechanism for the phosphorylation status of intracellular proteins through the regulation of PP5 enzymatic activity or PP5-client protein interaction.     

Dephosphorylation of clustered phosphoserine residues in human Grb14 by protein phosphatase 1 and its effect on insulin receptor complex formation

The physical interaction of the human growth factor receptor‐bound protein 14 (hGrb14) and the insulin receptor (IR) represses insulin signaling. With respect to the recruiting mechanism of hGrb14 to IR respond to insulin stimulus, our previous reports have suggested that phosphorylation of Ser³⁵⁸, Ser³⁶², and Ser³⁶⁶ in hGrb14 by glycogen synthase kinase‐3 repressed hGrb14–IR complex formation. In this study, we investigated phosphatase‐mediated dephosphorylation of the hGrb14 phosphoserine residues. An in vitro phosphatase assay with hGrb14‐derived synthetic phosphopeptides suggested that protein phosphatase 1 (PP1) is involved in the dephosphorylation of Ser³⁵⁸ and Ser³⁶². Furthermore, coimmunoprecipitation experiments suggested that insulin‐induced hGrb14–IR complex formation was repressed by the substitution of Ser³⁵⁸ or Ser³⁶² with glutamic acid. These findings suggested that phosphate groups on Ser³⁵⁸ and Ser³⁶² in hGrb14 are dephosphorylated by PP1, and the dephosphorylation facilitates hGrb14–IR complex formation.     

PP1 phosphatase-binding motif in Reg1 protein of Saccharomyces cerevisiae is required for interaction with both the PP1 phosphatase Glc7 and the Snf1 protein kinase

In Saccharomyces cerevisiae, Snf1 kinase, the ortholog of the mammalian AMP-activated protein kinase, is activated by an increase in the phosphorylation of the conserved threonine residue in its activation loop. The phosphorylation status of this key site is determined by changes in the rate of dephosphorylation catalyzed by the yeast PP1 phosphatase Glc7 in a complex with the Reg1 protein. Reg1 and many PP1 phosphatase regulatory subunits utilize some variation of the conserved RVxF motif for interaction with PP1. In the Snf1 pathway, the exact role of the Reg1 protein is uncertain since it binds to both the Glc7 phosphatase and to Snf1, the Glc7 substrate. In this study we sought to clarify the role of Reg1 by separating the Snf1- and Glc7-binding functions. We generated a series of Reg1 proteins, some with deletions of conserved domains and one with two amino acid changes in the RVxF motif. The ability of Reg1 to bind Snf1 and Glc7 required the same domains of Reg1. Further, the RVxF motif that is essential for Reg1 binding to Glc7 is also required for binding to Snf1. Our data suggest that the regulation of Snf1 dephosphorylation is imparted through a dynamic competition between the Glc7 phosphatase and the Snf1 kinase for binding to the PP1 regulatory subunit Reg1.     

Protein phosphatase 2A is associated with class C L-type calcium channels (Cav1.2) and antagonizes channel phosphorylation by cAMP-dependent protein kinase

Phosphorylation by cAMP-dependent protein kinase (PKA) regulates a vast number of cellular functions. An important target for PKA in brain and heart is the class C L-type Ca(2+) channel (Ca(v)1.2). PKA phosphorylates serine 1928 in the central, pore-forming alpha(1C) subunit of this channel. Regulation of channel activity by PKA requires a proper balance between phosphorylation and dephosphorylation. For fast and specific signaling, PKA is recruited to this channel by an protein kinase A anchor protein (Davare, M. A., Dong, F., Rubin, C. S., and Hell, J. W. (1999) J. Biol. Chem. 274, 30280-30287). A phosphatase may be associated with the channel to effectively balance serine 1928 phosphorylation by channel-bound PKA. Dephosphorylation of this site is mediated by a serine/threonine phosphatase that is inhibited by okadaic acid and microcystin. We show that immunoprecipitation of the channel complex from rat brain results in coprecipitation of PP2A. Stoichiometric analysis indicates that about 80% of the channel complexes contain PP2A. PP2A directly and stably binds to the C-terminal 557 amino acids of alpha(1C). This interaction does not depend on serine 1928 phosphorylation and is not altered by PP2A catalytic site inhibitors. These results indicate that the PP2A-alpha(1C) interaction constitutively recruits PP2A to the channel complex rather than being a transient substrate-catalytic site interaction. Functional assays with the immunoisolated class C channel complex showed that channel-associated PP2A effectively reverses serine 1928 phosphorylation by endogenous PKA. Our findings demonstrate that both PKA and PP2A are integral components of the class C L-type Ca(2+) channel that determine the phosphorylation level of serine 1928 and thereby channel activity.     

Protein kinase C (PKC) isoforms in Drosophila

Six protein kinase C (PKC) genes are present in Drosophila, comprising two classical PKCs (PKC53E and eye-PKC), two novel PKCs (PKC98E and PKCdelta), an atypical PKC (DaPKC), and a PKC-related kinase. Loss of function alleles affecting DaPKC and eye-PKC are available and their mutant phenotypes have been characterized. DaPKC is essential for early embryonic development because it regulates cell polarity and asymmetric cell division. Eye-PKC plays a role in the regulation of visual signaling, a G-protein coupled phospholipase Cbeta-mediated cascade. Both eye-PKC and DaPKC are differentially localized through tethering to multimolecular complexes. DaPKC interacts with partitioning-defective 3 (Par-3) and Par-6 proteins, which contain PDZ (PSD95, DLG, ZO-1) domains. Similarly, eye-PKC is anchored to a PDZ domain containing scaffolding protein INAD. Characterization of these two PKCs in Drosophila revealed a universal mechanism by which PKC is tethered to specific protein complexes for participation in distinct signal transduction processes.     

Activation of the Drosophila TRP and TRPL channels requires both Ca2+ and protein dephosphorylation

The Transient Receptor Potential (TRP) proteins constitute a large and diverse family of channel proteins, which is conserved through evolution. TRP channel proteins have critical functions in many tissues and cell types, but their gating mechanism is an enigma. In the present study patch-clamp whole-cell recordings was applied to measure the TRP- and TRP-like (TRPL)-dependent currents in isolated Drosophila ommatidia. Also, voltage responses to light and to metabolic stress were recorded from the eye in vivo. We report new insight into the gating of the Drosophila light-sensitive TRP and TRPL channels, by which both Ca2+ and protein dephosphorylation are required for channel activation. ATP depletion or inhibition of protein kinase C activated the TRP channels, while photo-release of caged ATP or application of phorbol ester antagonized channels openings in the dark. Furthermore, Mg(2+)-dependent stable phosphorylation event by ATPgammaS or protein phosphatase inhibition by calyculin A abolished activation of the TRP and TRPL channels. While a high reduction of cellular Ca2+ abolished channel activation, subsequent application of Ca2+ combined with ATP depletion induced a robust dark current that was reminiscent of light responses. The results suggest that the combined action of Ca2+ and protein dephosphorylation activate the TRP and TRPL channels, while protein phosphorylation by PKC antagonized channels openings.     

Dephosphorylation of Carma1 by PP2A negatively regulates T-cell activation

The Carma1-Bcl10-Malt1 (CBM) complex bridges T-cell receptor (TCR) signalling to the canonical IκB kinase (IKK)/NF-κB pathway. NF-κB activation is triggered by PKCθ-dependent phosphorylation of Carma1 after TCR/CD28 co-stimulation. PKCθ-phosphorylated Carma1 was suggested to function as a molecular scaffold that recruits preassembled Bcl10-Malt1 complexes to the membrane. We have identified the serine-threonine protein phosphatase PP2A regulatory subunit Aα (PPP2R1A) as a novel interaction partner of Carma1. PPP2R1A is associated with Carma1 in resting as well as activated T cells in the context of the active CBM complex. By siRNA-mediated knockdown and in vitro dephosphorylation, we demonstrate that PP2A removes PKCθ-dependent phosphorylation of Ser645 in Carma1, and show that maintenance of this phosphorylation is correlated with increased T-cell activation. As a result of PP2A inactivation, we find that enhanced Carma1 S645 phosphorylation augments CBM complex formation, NF-κB activation and IL-2 or IFN-γ production after stimulation of Jurkat T cells or murine Th1 cells. Thus, our data define PP2A-mediated dephosphorylation of Carma1 as a critical step to limit T-cell activation and effector cytokine production.     

Multiple interactions within the AKAP220 signaling complex contribute to protein phosphatase 1 regulation

The phosphorylation status of cellular proteins is controlled by the opposing actions of protein kinases and phosphatases. Compartmentalization of these enzymes is critical for spatial and temporal control of these phosphorylation/dephosphorylation events. We previously reported that a 220-kDa A-kinase anchoring protein (AKAP220) coordinates the location of the cAMP-dependent protein kinase (PKA) and the type 1 protein phosphatase catalytic subunit (PP1c) (Schillace, R. V., and Scott, J. D. (1999) Curr. Biol. 9, 321-324). We now demonstrate that an AKAP220 fragment is a competitive inhibitor of PP1c activity (K(i) = 2.9 +/- 0.7 micrometer). Mapping studies and activity measurements indicate that several protein-protein interactions act synergistically to inhibit PP1. A consensus targeting motif, between residues 1195 and 1198 (Lys-Val-Gln-Phe), binds but does not affect enzyme activity, whereas determinants between residues 1711 and 1901 inhibit the phosphatase. Analysis of truncated PP1c and chimeric PP1/2A catalytic subunits suggests that AKAP220 inhibits the phosphatase in a manner distinct from all known PP1 inhibitors and toxins. Intermolecular interactions within the AKAP220 signaling complex further contribute to PP1 inhibition as addition of the PKA regulatory subunit (RII) enhances phosphatase inhibition. These experiments indicate that regulation of PP1 activity by AKAP220 involves a complex network of intra- and intermolecular interactions.     

Targeting of PKA, PKC and protein phosphatases to cellular microdomains

The intracellular responses to many distinct extracellular signals involve the direction of broad-based protein kinases and protein phosphatases to catalyse quite specific protein phosphorylation/dephosphorylation events. It is now clear that such specificity is often achieved through subcellular targeting of distinct pools of kinase or phosphatase towards particular substrates at specific subcellular locations. Given the dynamic nature of protein phosphorylation reactions, coordinated control of both kinase and phosphatases is often required and complexes formed by common scaffold or targeting proteins exist to direct both kinase and phosphatase to the same subcellular location. In many cases more than one kinase or phosphatase is required and binding proteins which target more than one kinase or phosphatase have now been identified. This review summarizes recent findings relating to the concept of targeting PKA, PKC and the major serine/threonine phosphatases, PP1, PP2A and PP2B, through the formation of multi-enzyme signalling complexes.