inactivation no afterpotential D

InaD and signal transduction

The subcellular compartmentalization of signaling molecules helps to ensure the selective activation of different signal-transduction cascades within a single cell. Although there are many examples of compartmentalized signaling molecules, there are few examples of entire signaling cascades being organized as distinct signaling complexes. In Drosophila photoreceptors, the InaD protein, which consists of five PDZ domains, functions as a multivalent adaptor that brings together several components of the phototransduction cascade into a macromolecular complex. Single-photon responses have been studied in several photoreceptor mutant backgrounds. These show that the InaD macromolecular complex is the unit of signaling that underlies elementary responses. The localized activity of this signaling unit promotes reliable single-photon responses as well as rapid activation and feedback regulation. Genetic and electrophysiological tools were used to illustrate how the assembly of signaling molecules into a transduction complex limits signal amplification in vivo (Scott, 1998).

The rapid activation and feedback regulation of many G protein signaling cascades raises the possibility that the critical signaling proteins may be tightly coupled. Previous studies have shown that the PDZ domain containing protein InaD, which functions in Drosophila vision, coordinates a signaling complex by binding directly to the light-sensitive ion channel, TRP, and to phospholipase C (PLC). The InaD signaling complex also includes rhodopsin, protein kinase C (PKC), and Calmodulin, though it is not known whether these proteins bind to InaD. This study shows that rhodopsin, calmodulin, and PKC associate with the signaling complex by direct binding to InaD. A second ion channel, TRPL, also binds to InaD. Thus, most of the proteins involved directly in phototransduction appear to bind to InaD. Furthermore, InaD forms homopolymers and the homomultimerization occurs through two PDZ domains. Thus, it is proposed that the InaD supramolecular complex is a higher order signaling web consisting of an extended network of InaD molecules through which a G protein-coupled cascade is tethered (Xu, 1998a).

Homomultimerization of InaD

The finding that PDZ3 and PDZ4 bind rhodopsin, TRPL and PKC indicates that a single InaD molecule would not have the capacity to nucleate the entire visual transduction signaling complex unless InaD functions as a homomultimeric protein. To address this hypothesis, full-length InaD fused with MYC or FLAG epitope tags are co-expressed in 293T cells and it was found that InaD-FLAG coimmunoprecipitates with InaD-MYC. Further evidence that InaD homomultimerizes was obtained by demonstrating that 35S-InaD binds to a GST-InaD fusion immobilized on a glutathione column. The InaD homomultimerization occurs through PDZ domains (either PDZ3 or PDZ4). Furthermore, InaD PDZ3 and PDZ4 can form either homomeric or heteromeric interactions. The PDZ-PDZ interaction appears to be specific to PDZ3 and PDZ4 since neither PDZ1, PDZ2, nor PDZ5 binds to any portion of InaD, including PDZ3 or PDZ4 (Xu, 1998a).

The observation that homomeric interactions occur through either PDZ3 or PDZ4 raises the possibility that InaD may form a homopolymer rather than just a dimer. To address this possibility, a segment of InaD including just PDZ3-PDZ4 was translated in vitro and the products were fractionated by sucrose gradient sedimentation. Although a proportion of PDZ3-4 fractionates near the predicted molecular weight of the dimer (52 kD), a significant amount sediments as a much larger protein of ~200 kD. PDZ1-PDZ2 loaded onto the same gradient sediments with a single peak near its predicted monomer molecular weight of ~39 kD. Thus, PDZ1-2 does not homomultimerize or interact with PDZ3-PDZ4. The data that a proportion of PDZ3-PDZ4 sediments as a protein greater than 200 kD suggests that InaD may be capable of forming homopolymers with a subunit composition of greater than 8. In contrast to PDZ1-PDZ2, which fractionates with a single peak, four small peaks are detected with PDZ3-PDZ4 that roughly correspond to the predicted sizes of molecules with 1, 2, 4, and 6 subunits. Since the PDZ1-PDZ2 monomer and marker proteins distribute over many fractions, the PDZ3-PDZ4 peaks may be small due to a similar broad distribution of the PDZ3-PDZ4 monomer, dimer, and higher order forms (Xu, 1998a).

The findings that InaD can form homomultimers through PDZ3 and PDZ4 raises the question as to whether homomultimerization precludes InaD-target interaction or vice versa. To investigate whether homomultimerization and PDZ-target interactions can occur simultaneously, advantage was taken of the finding that PDZ3 alone is sufficient to promote homotypic interactions, whereas PDZ3L is required for binding to the opsin, TRPL, or PKC. Therefore, tests were run to determine whether PDZ3 coimmunoprecipitates with the targets after coexpressing PDZ3 and PDZ3L with either TRPL or PKC. TRPL or PKC coimmunoprecipitate with PDZ3 in the presence (but not in the absence) of PDZ3L. These results indicate that PDZ3 and PDZ3L formed a ternary complex with the target proteins and suggest that the InaD PDZ-PDZ and PDZ-target interactions are mediated via different interfaces. Consistent with this latter proposal, an NH2-terminal truncation that removes the first and second putative beta barrel from PDZ3L (PDZ3LdeltaN) disrupts interaction with PKC; however, homomeric binding still occurs. Furthermore, the COOH-terminal extension in PDZ3L is required for binding to PKC but not for the PDZ-PDZ association. The extra COOH-terminal residues in PDZ3L are not sufficient for binding to PKC since PDZ3LdeltaN does not bind PKC (Xu, 1998a).

Interaction of InaD with the major rhodopsin encoded by the ninaE locus

A suggestion that the major rhodopsin encoded by the ninaE locus may interact with InaD is that it coimmunoprecipitates with the TRP channel from wild-type but not InaDP215 mutant fly heads (Chevesich, 1997). To test whether the rhodopsin interacts with InaD in vivo, a coimmunoprecipitation experiment was performed using extracts from Drosophila heads. InaD coimmunoprecipitates with rhodopsin but not in a control reaction carried out with nonimmune serum. In addition, the proteins were co-expressed in vitro using a mammalian tissue culture system, 293T cells, and it was found that ninaE opsin and InaD coimmunoprecipitate. Evidence that the association between the opsin and InaD is direct is that in vitro-translated opsin binds to GST-InaD immobilized on a glutathione-Sepharose column but not to the GST control (Xu, 1998a).

To map the regions in InaD mediating the interactions with the opsin, TRPL, and PKC, a series of InaD constructs were generated and coexpressed with each target protein in 293T cells. Either PDZ3 or PDZ4 is sufficient to interact with the opsin, TRPL, and PKC. However, binding of each target to PDZ3 requires an extra 28 amino acids COOH-terminal to PDZ3 (PDZ3L). A requirement for additional residues for target binding to a PDZ domain is not unique since a similar COOH-terminal extension is necessary for the binding of target peptides to the PDZ domain in neuronal nitric acid synthase. The association of the targets with PDZ3L requires PDZ3 rather than just the extra 28 COOH-terminal amino acids since truncation of the NH2-terminal portion of PDZ3L-obliterates binding. Binding of the opsin, TRPL, and PKC to PDZ3L and PDZ4 appears to be specific since none of the eight other proteins or protein fragments tested bind to PDZ3L or PDZ4. These included TRPC3, Shaker B, Calmodulin, and the PLC encoded by the no receptor potential A (norpA) locus. Furthermore, consistent with a recent report that PLC binds to a GST-PDZ1 fusion protein (van Huizen, 1998), it has been found that the COOH-terminal 123 residues of PLC expressed in 293T cells coimmunoprecipitate with either PDZ1 or PDZ1-PDZ2. These data suggest that the lack of interaction between these latter two PDZ domains and either the opsin, TRPL, or PKC is not due toimproper folding of PDZ1 or PDZ1-PDZ2. Studies in other labs, using bacterial fusion proteins, indicate that PLC binds to either PDZ1 and PDZ5 (van Huizen, 1998) or PDZ5 only (Tsunoda, 1997). Although the PLC expressed in 293T cells does not bind to PDZ5, PLC interacts with PDZ5 expressed in E. coli. The lack of interaction of PLC with PDZ5 in 293T cells may be due to interference by post-translational modifications of PDZ5, endogeous proteins that bind to either PLC or PDZ5 (precluding the PLC/PDZ5 interaction), or misfolding of PDZ5 (Xu, 1998a).

Interaction of InaD with NorpA (an eye specific phospholipase C)

The Transient receptor potential protein (Trp) is a putative capacitative Ca2+ entry channel present in fly photoreceptors, which use the inositol 1,4,5-trisphosphate (InsP3) signaling pathway for phototransduction. By immunoprecipitation studies, Trp is found associated into a multiprotein complex with the norpA-encoded phospholipase C, an eye-specific protein kinase C (InaC) and with the InaD protein (InaD). InaD is a putative substrate of InaC and contains two PDZ repeats, putative protein-protein interaction domains. These proteins are present in the photoreceptor membrane at about equimolar ratios. The Trp homolog can be isolated together with NorpA, InaC and InaD from blowfly (Calliphora) photoreceptors. Compared to Drosophila Trp, the Calliphora Trp homolog displays 77% amino acid identity. The highest sequence conservation is found in the region that contains the putative transmembrane domains S1-S6 (91% amino acid identity). As investigated by immunogold labeling with specific antibodies directed against Trp and InaD, the Trp signaling complex is located in the microvillar membranes of the photoreceptor cells. The spatial distribution of the signaling complex argues against a direct conformational coupling of Trp to an InsP3 receptor supposed to be present in the membrane of internal photoreceptor Ca2+ stores. It is suggested that the organization of signal transducing proteins into a multiprotein complex provides the structural basis for an efficient and fast activation and regulation of Ca2+ entry through the Trp channel (Huber, 1996b).

Photoreceptors which use a phospholipase C-mediated signal transduction cascade harbor a signaling complex in which the phospholipase Cbeta (PLCbeta), the light-activated Ca2+ channel TRP, and an eye-specific protein kinase C (ePKC) are clustered by the PDZ domain protein InaD. The function of ePKC was investigated by cloning the Calliphora homolog of Drosophila ePKC, by precipitating the TRP signaling complex with anti-ePKC antibodies, and by performing phosphorylation assays in isolated signaling complexes and in intact photoreceptor cells. The deduced amino acid sequence of Calliphora ePKC comprises 685 amino acids and displays 80.4% sequence identity with Drosophila ePKC. Immunoprecipitations with anti-ePKC antibodies leads to the coprecipitation of PLCbeta, TRP, InaD and ePKC but not of rhodopsin. Phorbolester- and Ca2+-dependent protein phosphorylation reveals that, apart from the PDZ domain protein InaD, the Ca2+ channel TRP is a substrate of ePKC. TRP becomes phosphorylated in isolated signaling complexes. TRP phosphorylation in intact photoreceptor cells requires the presence of extracellular Ca2+ in micromolar concentrations. It is proposed that ePKC-mediated phosphorylation of TRP is part of a negative feedback loop that regulates Ca2+ influx through the TRP channel (Huber, 1998).

Drosophila InaD, which contains five tandem protein interaction PDZ domains, plays an important role in the G protein-coupled visual signal transduction. Mutations in inaD alleles display mislocalizations of signaling molecules of phototransduction that include the essential effector, phospholipase C-beta (PLC-beta), also known as NORPA. The molecular and biochemical details of this functional link are unknown. InaD directly binds to NORPA via two terminally positioned PDZ1 and PDZ5 domains. PDZ1 binds to the C-terminus of NORPA, while PDZ5 binds to an internal region overlapping with the G box-homology region (a putative G protein-interacting site). Altered NORPA proteins lacking binding sites display normal basal PLC activity but can no longer associate with InaD in vivo. These truncations cause significant reduction of NORPA protein expression in rhabdomeres and severe defects in phototransduction. Thus, the two terminal PDZ domains of InaD, through intermolecular and/or intramolecular interactions, are brought into proximity in vivo. Such domain organization allows for the multivalent InaD-NORPA interactions, which are essential for G protein-coupled phototransduction (van Huizen, 1998).

Visual transduction in Drosophila is a G protein-coupled phospholipase C-mediated process that leads to depolarization via activation of the transient receptor potential (TRP) calcium channel. Inactivation-no-afterpotential D (InaD) is an adaptor protein containing PDZ domains known to interact with TRP. Immunoprecipitation studies indicate that InaD also binds to eye-specific protein kinase C (INAC) and the phospholipase C, no-receptor-potential A (NORPA). By overlay assay and site-directed mutagenesis the essential elements of the NORPA-InaD association have been defined and three critical residues in the C-terminal tail of NORPA, required for the interaction, have been identified. These residues, Phe-Cys-Ala, constitute a novel binding motif distinct from the sequences recognized by the PDZ domain in InaD. To evaluate the functional significance of the InaD-NORPA association in vivo, transgenic flies were derived expressing a modified NORPA that lacks the InaD interaction: NORPAC1094S. The transgenic animals display a unique electroretinogram phenotype characterized by slow activation and prolonged deactivation. Double mutant analysis suggests a possible inaccessibility of eye-specific protein kinase C to NORPAC1094S, undermining the observed defective deactivation. Similarly, delayed activation may result from NORPAC1094S being unable to localize in close proximity to the TRP channel. It is concluded that InaD acts as a scaffold protein that facilitates NORPA-TRP interactions required for gating of the TRP channel in photoreceptor cells (Shieh, 1997).

Because PLC is activated by a G protein, the inactivation of the Galpha-GTP by the GTPase reaction may be relevant to response termination. Previous in vitro studies of mammalian PLC-beta and GAP activity used a purified recombinant M 1 muscarinic receptor reconstituted into phospholipid vesicles with G_q/11 and PLC-beta1. The addition of PLC-beta1 to the reconstituted system increases the rate at which G_q hydrolyses GTP by three orders of magnitude. Phototransduction by vertebrate photoreceptors depends on a specific GAP activity. Accordingly, genetic elimination of regulators of G-protein signaling (RGS) proteins reduces and slows down GAP activity and leads to slow termination of responses to light. Thus, it is possible that in Drosophila, the effect of the association of PLC and INAD on response termination is related to the GAP activity of PLC (Cook, 2000 and references therein).

Both vertebrate photoreceptors are able to discern single photons by a reliable and reproducible production of a unitary response (bump) upon absorption of each photon, whereas the macroscopic response to light is a superposition of multiple bumps. Generation of unitary responses is a characteristic feature of many signal transduction cascades. There is a profound difference between the mechanism of bump production in vertebrate and invertebrate species. In vertebrate phototransduction, which is characterized by signal amplification at the early stages of the cascade, the shape of the bump is, at least partially, determined at the stages of rhodopsin and G-protein action. Accordingly, in transgenic mice lacking RGS 9 and showing reduced GAP activity, the declining phase of the bump is markedly slowed down. In invertebrates, however, there is little, if any, amplification at early stages of phototransduction and the shape of the bump is determined downstream of PLC activation. The interaction of GAP activity, PLC and PLC-INAD may thus have a key role in the mechanism that ensures production of a single bump following absorption of a single photon (Cook, 2000 and references therein).

In the work reported here, Drosophila phototransduction was used as a model system to study the physiological implications of formation of a signaling complex and PLC-induced GAP activity. The functional role of the association of PLC with the INAD signaling complex was investigated. This association is required for a high rate of GTPase activity induced by the large concentration of PLC. The dual role of PLC, the target of the G protein, as an activator of the cascade and as an essential negative regulator of the G protein, ensures that every activated G protein produces a bump. Reduced levels of PLC result in accumulation of active G protein, leading to slow response termination and hence to impaired temporal and intensity resolution (Cook, 2000).

The continuous response in the PLC-deficient mutants, long after the cessation of light, suggests that reduction in PLC levels results in light-induced accumulation of a long-lived signaling component that can activate the phototransduction cascade in the dark. The most likely candidate for this component is the active form of the G_q protein (G_qalpha-GTP), since it operates directly upstream of PLC, has a long life-time and precedes bump formation. Following the previous experiments showing PLC-dependent GAP activity in a reconstituted mammalian system in vitro, a test was performed to see whether the mechanism underlying the defects in response termination is a reduced PLC-dependent GAP activity. To this end, both the light-dependent binding of the non-hydrolysable GTP analogue GTP-gamma-S, which represents the amounts of G protein available for activation by light, and GTPase activity, which reflects the amount of light-activated G protein that is turned off, were measured. The rate of turn-off reflects the level of GAP activity. Using these biochemical assays the relationship between the electrophysiological phenotype of the above mutants, their PLC levels and their GAP activity were measured. The GTP-gamma-S binding assay shows that the amounts of G protein available for activation by light are similar in the wild type and in all the PLC-defective and inaD mutants. Although G-protein activation is similar, the light-dependent GTPase activity is graded and is strongly dependent on the PLC levels in these mutants (Cook, 2000).

Since several different G proteins are present in each cell, specificity of the GTPase to the G_qalpha that operates in phototransduction was demonstrated by assaying the Galphaq1 mutant. This mutant expresses only ~1% of the normal amount of G_qalpha that is required for generating the response to light. If the GTPase that was measure reflects the activity of the G_qalpha that operates in phototransduction, it would be expected that GTP-gamma-S binding and GTPase activity will be much reduced in this mutant. As expected, in the Galphaq 1 mutant highly reduced light-activated GTP-gamma-S binding and GTPase activity is observed. The reduction in GTPase activity of the Galphaq 1 mutant does not arise from reduced levels of PLC since the mutant has normal amounts (Cook, 2000).

To demonstrate in vitro that the PLC-induced GAP activity is specifically required for inactivation of Gqalpha, a biochemical complementation approach was undertaken. GTPase activity was reconstituted by fusing head membranes of the null norpAP24 mutant with those of the Galphaq 1 hypomorphic mutant using polyethyleneglycol (PEG). The GTPase activity obtained after fusing the membranes was significantly higher than the activity observed without the complementing membranes, whereas no effect was found without PEG. Thus, the membrane fusion supplement containing PLC and G_qalpha was sufficient to reconstruct GTPase activity in mutants that otherwise show only background activity. The rescue of GTPase activity is not expected to reach wild-type level since a significant amount of membrane will not fuse and only some of each membrane type will fuse with the other type in the optimal orientation and distance. Therefore, the rescue by complementation is significant, and indicates that in Drosophila photoreceptors eye-specific PLC (NORPA) is the component required for the GAP activity of Gqalpha.

To further establish a relationship between GAP activity and amount of PLC, GTPase activity in the PLC-INAD interacting and non-interacting mutants was plotted as a function of their PLC levels. GTPase activity correlates linearly with the PLC levels. The correlation is evident in non-interacting mutants as well as in the interacting mutant norpAP57 and the transgenic Drosophila T 6, which expresses reduced amounts of normal NORPA12. The linear correlation, which applies to all mutants, suggests that PLC-INAD interaction is not required for induction of GAP activity, as was revealed in the electrophysiological experiments. Rather, it is the amount of PLC that is required for GAP activity. This conclusion is consistent with earlier biochemical data showing that the carboxy-terminal region of PLC-beta1 is sufficient to induce GAP activity in a reconstituted vesicle preparation from mammalian cells (Cook, 2000).

The ability of transduction cascades to resolve signals reliably in time depends on the turn-off of each activated component when the stimulus ceases. Negative feedback by the activated effectors fulfils this requirement and the shortest possible regulatory cycle would be the direct turnoff of the active signaling component by its target. Here it has been shown that in Drosophila phototransduction, PLC, the known target of the active G protein, induces GTPase activity and thereby inactivates the G_qalpha-GTP. This GAP activity is critical for termination of responses to light. Previous studies have shown that grouping of signaling proteins into a complex by INAD is essential for localization and for maintaining sufficiently high levels of the protein (Cook, 2000).

The results presented here suggest that all these properties are important for the ability to produce enough GAP activity to facilitate response termination and allow the target-dependent inactivation process. Therefore, the grouping of key transduction components into multiprotein structures is functionally important not only in achieving speed and efficiency of signaling by reducing diffusion distances, but also for maintaining optimal stoichiometric ratios of the participating components. When PLC levels in the signaling membranes are low relative to the amount of the active G protein, light induces production of G_qalpha-GTP at a higher rate than it is inactivated by PLC. The G_qalpha-GTP that has accumulated during illumination continues to produce bumps in the dark until all active G_qalpha-GTP molecules are hydrolysed by the scarce PLC. Hence, flies are not able to resolve light stimuli and become virtually blind at low levels of PLC. Assembly of transduction components into signaling complexes is therefore important for many facets of the physiological response, as revealed by the crucial role of PLC association with the signaling complex (Cook, 2000).

In Drosophila, phototransduction is mediated by Gq-activation of phospholipase C and is a well studied model system for understanding the kinetics of signal initiation, propagation and termination controlled by G proteins. The proper intracellular targeting and spatial arrangement of most proteins involved in fly phototransduction requires the multi-domain scaffolding protein InaD, composed almost entirely of five PDZ domains, which independently bind various proteins including NorpA, the relevant phospholipase C-ß isozyme. The crystal structure of the N-terminal PDZ domain of InaD bound to a peptide corresponding to the C-terminus of NorpA has been determined to 1.8 Å resolution. The structure highlights an intermolecular disulfide bond necessary for high affinity interaction as determined by both in vitro and in vivo studies. Since other proteins also possess similar, cysteine-containing consensus sequences for binding PDZ domains, this disulfide-mediated 'dock-and-lock' interaction of PDZ domains with their ligands may be a relatively ubiquitous mode of coordinating signaling pathways (Kimple, 2001).

As shown by gel exclusion chromatography, NorpA forms a stable homodimer through its C-terminal domain, supporting the recent discovery that other PLC-ßs, namely mammalian PLC-ß1 and -ß2 and turkey PLC-ß, form stable dimers. As is shown in the present work, a stable CT_Dm dimer can be covalently linked to two PDZ1 molecules through intermolecular disulfide bond formation. Evidence of dimerization of NorpA and other PLC-ß isoforms, along with disulfide-mediated interaction of NorpA and InaD, necessitates a revision of the current model of Drosophila phototransduction. In this revised model, a NorpA homodimer can covalently bind two InaD molecules, each of which can homodimerize through PDZ3 and PDZ4. This increases the number and strength of connections between phototransduction components by linking several rhodopsins, eyePKCs and TRP isozymes through NorpA dimers. As in the former model, PDZ1 can also bind to the myosin NinaC, linking the entire membrane-bound signalplex to the actin cytoskeleton. Binding of PDZ1 to NinaC, which has the C-terminal sequence VDI-COO^–, requires at least the C-terminal 21 residues of NinaC, suggesting that PDZ1 may interact with NinaC in a different mode than it does with NorpA (Kimple, 2001).

The spatial localization and connectivity of relevant proteins would help explain why the Drosophila phototransduction pathway is one of the fastest G protein-coupled signaling cascades known, with activation and termination occurring within tens of milliseconds. Another factor contributing to the rapid cycling of the Drosophila phototransduction cascade is that NorpA functions not only as an effector for G_q signaling, but also as a GTPase-activating protein (GAP) for GTP-bound Galpha_q. Drosophila norpA mutations suggest that the isosteric Cys1094Ser point mutation effectively abrogates all functional PDZ1 binding to NorpA, mimicking the effects of completely deleting the PDZ1 binding site (NorpA DeltaCterm 25). The combination of well characterized norpA mutants together with structural and biochemical analyses strongly suggest that disulfide-bond formation between InaD and the penultimate residue of NorpA is critical for proper phototransduction in Drosophila. By holding NorpA in a tight, membrane-bound complex, a covalent association between InaD and NorpA aids in the rapid regulation of G_q by NorpA (Kimple, 2001).

Interaction of InaD with Calmodulin

continued: inactivation no afterpotential D Regulation: protein interactions part 2/2

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.