The regulation of kinases by scaffolding proteins greatly contributes to the fidelity of signal transduction. In the present study, we explored an interaction between the ubiquitous enzyme PKC (protein kinase C) and the scaffolding protein AKAP7 (A-kinase-anchoring protein 7). Using protein biochemistry and surface plasmon resonance approaches, we demonstrate that both AKAP7γ and AKAP7α are capable of high-affinity interactions with multiple isoenzymes of PKC. Furthermore, this interaction is achieved via multi-site binding on both proteins. FRET (fluorescence resonance energy transfer) analysis using a PKC activity reporter suggests that anchoring of the kinase within AKAP7 complexes enhances the phosphorylation of substrate proteins. Finally, we determined using FRAP (fluorescence recovery after photobleaching) and virtual modelling that AKAP7 restricts the mobility of PKC within cells by tethering it to subcellular compartments. Collectively, the results of the present study suggests that AKAP7 could play an integral role in dictating PKC localization and function in tissues where the two proteins are co-expressed.
Protein phosphorylation is an archetypal signal to trigger changes in cellular physiology and function. These critical events are mediated by a cellular complement of an estimated 500 protein kinases . The existence of a vastly greater number of phosphorylatable targets, however, necessitates redundancy in kinase action and highlights the importance of their contextual regulation . Although many kinase and substrate relationships have been identified, the molecular framework that confers their specificity is often a mystery.
An example of this occurs with PKC (protein kinase C), a ubiquitous lipid-dependent kinase. There are nine isoenzymes of PKC characterized in mammals, and collectively they mediate many aspects of cellular physiology through reversible phosphorylation of numerous substrate proteins . Yet despite the abundance of substrate proteins available, PKC must act on its intended target with precision. Deregulation of the enzyme is necessary and sufficient for the progression of several pathologies such as tumorigenesis and cardiac hypertrophy, making it the target of several ongoing clinical trials and the crux of many future studies [4,5]. In light of this, a better understanding of its regulatory underpinnings is essential. The isoenzymes of PKC are classified into three groups on the basis of their activation determinants . The activation of the atypical class of PKC (ι and ξ) is poorly understood, but genetic manipulations have defined hallmark features of the conventional and novel classes. Both the conventional isoenzymes (α, β and γ) and the novel isoenzymes (∊, η, θ and δ) possess a C2 domain in the N-terminal regulatory region. This domain binds to DAG (diacylglycerol) and phospholipids necessary for activation of the enzyme. The two classes differ in the C1 region, which is responsible for binding intracellular Ca2+ in the conventional PKCs, making this class of isoenzymes uniquely sensitive to intracellular Ca2+ release . These differences in activation could serve to broadly define the activity of PKC in a specific region of the cell. For example, the higher affinity of the novel PKCs for DAG is thought to keep them in close proximity to membranes where DAG is generated . It is of note, however, that there is a high degree of homology in the catalytic domain of PKC . Under aberrant conditions this similarity probably underlies the promiscuity of the activated enzyme, fuelling frequent controversies over which isoenzymes are important for which cellular processes [4,10]. However, differences in activation cannot explain how PKC achieves specificity for an individual substrate. A large body of work has revealed that PKC is often regulated by interactions with scaffolding proteins, such as AKAPs (A-kinase-anchoring proteins), which can confer this requisite specificity .
AKAPs have a defined role in directing enzyme activity towards specific substrates, and are most widely studied for their role in the compartmentalization of intracellular cAMP signalling. This is achieved by tethering the PKA (cAMP-dependent protein kinase) holoenzyme in a protein complex with its effectors at discrete locations within the cell. The association of tertiary signalling molecules, such as adenylate cyclase and phosphodiesterase, facilitates the local synthesis and degradation of cAMP [12,13]. These cellular microenvironments facilitate the acceleration, amplification and precision of signal transduction .
Our laboratory is interested in AKAP7 (AKAP15/18), an alternatively spliced scaffolding protein with four characterized isoforms enriched in many tissues, including the brain, heart and kidney . The low-molecular-mass AKAP7 isoforms (α and β) were found to directly associate with cell membranes via N-terminal lipid modifications. AKAP7α was also shown to bind to the L-type calcium channel in muscle tissues via a leucine zipper motif, a necessity for the agonist-evoked phosphorylation of the channel . Larger isoforms of AKAP7 (γ and δ), although initially thought to be cytosolic, now appear to have some cell-type-specific localizations that are dictated by protein–protein interactions . In the heart, these isoforms are known to bind the Ca2+ regulatory protein phospholamban and are thereby targeted to the sarcoplasmic reticulum . By promoting phosphorylation of phospholamban by PKA, AKAP7 plays an integral role in Ca2+ cycling within the cardiac myocyte. Studies in other cell types have implicated it in insulin secretion and aquaporin-2 shuttling [16,18,19].
Given its multifaceted importance in these various tissues, a more thorough understanding of this signalling complex is needed. Moreover, the importance of AKAP7 probably extends far beyond PKA and cAMP. For example, other AKAPs are known to associate with phosphatases and cAMP-independent kinases such as ERK (extracellular-signal-regulated kinase), or PKC . Although a great deal is known about the role of AKAP7 in the propagation of cAMP signals, prior to our present study, its role in regulating cAMP-independent second messenger systems was unknown. Given the demonstrated importance of AKAP7 in several systems, and the urgent need to identify novel regulators of PKC, in the present study we investigated an interaction between these proteins. Our results demonstrate a high-affinity interaction between AKAP7 and PKC, with multiple sites of contact on both proteins. Interestingly, these interactions are neither isoenzyme-specific, nor do they require an active kinase. The findings of the present study suggest that the primary function of the PKC–AKAP7 interaction is to tether the kinase to subcellular compartments and enhance the phosphorylation of intracomplex substrates. In support of this, FRET (fluorescence resonance energy transfer) analysis using an AKAP7-anchored CKAR (PKC activity reporter) revealed that AKAP7 amplified phosphorylation by PKC. Furthermore, FRAP (fluorescence recovery after photobleaching) experiments paired with virtual modelling demonstrate a reduced mobility and diffusion rate of intracellular PKC in the presence of AKAP7. Collectively, the results of the present study suggest that AKAP7 could play an integral role in dictating PKC localization and function in tissues where the two proteins are co-expressed, such as the heart, brain, pancreas and kidney.
The following primary antibodies were used for immunoblotting: mouse monoclonal anti-GFP (green fluorescent protein; Santa Cruz Biotechnology, 1:500 dilution), mouse monoclonal FLAG M2 (Sigma, 1:1000 dilution), goat anti-AKAP18 (AKAP7) (from Dr John Scott, University of Washington, Seattle, WA, U.S.A., 1:500 dilution), polyclonal anti-AKAP7 (Sigma, 1:1000 dilution) and mouse monoclonal anti-phospholamban (Abcam, 1:500 dilution). Immunoprecipitations were carried out using the following antibodies: mouse monoclonal FLAG M2 (Sigma; 3 μg), goat polyclonal anti-AKAP18 (AKAP7) (5 μl of whole serum) and monoclonal anti-phospholamban (Abcam, 3 μg).
The following constructs were obtained from Addgene: FLAG–PKC∊ (plasmid number 10795), PKCα–FLAG (plasmid number 10805), PKCα(K/N)–FLAG (plasmid number 10806), CKAR (plasmid number 10806) and PKCβII–YFP (yellow fluorescent protein) (plasmid number 14866). GFP–AKAP7γ and AKAP7α-pcDNA was obtained from Dr John Scott. AKAP7γ mapping constructs were cloned with EcoR1/HindIII restriction sites using PCR, and sub-cloned into the PET32 vector. AKAP7γ–CKAR was flanked with HindIII restriction sites and fused to the N-terminus of CKAR.
HEK (human embryonic kidney)-293 cells were transfected at 50–70% confluency in 60 mm-diameter plates using the calcium phosphate method with 6 μg of each plasmid DNA. Cells were harvested and lysed 24 h after transfection in 0.5 ml of HSE buffer [20 mM Hepes (pH 7.4), 150 mM NaCl, 5 mM EDTA, 1% Triton X-100 and protease inhibitors (leupeptin, pepstatin, benzamidine and AEBSF)]. Supernatants were incubated overnight at 4°C with the indicated antibody and 15 μl of prewashed Protein A– or G–agarose. Following extensive washing, captured proteins were solubilized in 2× sample buffer and analysed by immunoblot.
PKC activity assays
AKAP7 immunoprecipitates were isolated from transfected HEK-293 cells as described above. Following the final wash, immunoprecipitates were subjected to PKC activity assays using a kit, according to the manufacturer's protocol (Millipore). For purified protein experiments, 1 μg of purified AKAP7 fragments were pre-charged overnight on beads, followed by a 4 h incubation with 50 ng of purified PKC protein.
For inhibition assays, 20 ng of activated PKCα (Calbiochem) was incubated with the indicated concentrations of purified recombinant AKAP7γ or AKAP7α. The PKC activity relative to an AKAP-free control condition was determined in a PKC activity assay as described above.
Mapping via competition
AKAP7α or AKAP7γ-(1–250) (1 μg) were pre-charged on beads overnight. The next day, 5 μg of the indicated competing fragment and PKC were incubated for 4 h at 4°C. Immediately following this incubation period, the PKC activity was determined as described above.
SPR (surface plasmon resonance)
SPR analysis was performed using a BIAcore T100. PKCα purified protein was covalently immobilized using NHS (N-hydroxysuccinimide) and EDC [N-ethyl-N′-(3-dimethylaminopropyl)carbodi-imide] (BIAcore amine coupling kit) to the surface of a sensor chip (BIAcore type CM5). The amount of ligand bound [in RUs (response units)] was 300 RU. Protein analytes (AKAPα and AKAP7γ) were diluted at increasing concentrations (12.50–200 nM) in HBS buffer [10 mM Hepes (pH 7.4), 150 mM NaCl and 0.005% Surfactant P20] and were injected over the sensor surface at a flow rate of 30 μl/min for 300 s. Post-injection phase, dissociation was monitored in HBS buffer for 300 s at the same flow rate. The surface was regenerated between injections using 10 mM NaCl at a flow rate of 50 μl/min for 30 s. Sensorgrams were all processed by BIAcore T100 evaluation software.
Vero cells were seeded at 50% confluency on to glass coverslips (Warner Instruments) and transfected with 1 μg of plasmid DNA using the Lipofectamine™ PLUS transfection system (Life Technologies). Cells were maintained at room temperature (26°C) in imaging buffer [172 mM NaCl, 2.4 mM KCl, 10 mM Hepes, 4 mM CaCl2, 4 mM MgCl2 and 10 mM glucose (pH 7.2)] for the duration of the imaging, which began following a 5 min equilibration period on the microscope. All images were collected using a Zeiss Pascal confocal microscope and a 40×/1.2 NA (numerical aperture) objective. Excitation of CFP (cyan fluorescent protein) was carried out using a 440 nM laser (Toptica Photonics). A HQ535/50M and HQ480/40M emission filter with a 510DCLP dichroic were used (Chroma Technology). Channel intensities were quantified using ImageJ software (NIH). Graphing and analysis were performed in Microsoft Excel. Individual traces were standardized against their baseline values to put them on a scale of 1. Data are composite traces from multiple cells, which have been normalized against unstimulated cells imaged under identical conditions.
Transient transfections of Vero cells with PKCβII–YFP, pcDNA and AKAP7α-pcDNA were carried out as described above. Approximately 18 h post-transfection, cells were transferred into HBSS (Hanks balanced salt solution; +Ca2+) and imaged on a Zeiss Confocor 3 microscope. A circular region of PKCβII–YFP fluorescence was bleached using 514 nm and 488 nm lasers (Lasos) at 100% power for 25 iterations. A total of 75 images, ten pre-bleach and 65 post-bleach, were collected from each cell over a period of 146 s using a 40× 1.2 NA objective. Images were analysed using the Virtual FRAP software contained within VCell5.1 (http://www.nrcam.uchc.edu), which uses parameter optimization tools to determine best fits of spatial simulations to experimental data. Data were fitted to a model using a single diffusing component to determine the mobile fraction and effective diffusion coefficient. Experimental images were normalized to an averaged pre-bleach image and were exported directly from VCell software and assembled in Microsoft Powerpoint.
RESULTS AND DISCUSSION
Direct cellular interaction between AKAP7 and PKC
Previous work reporting an interaction between PKC and AKAP7 performed in Xenopus led us to become interested in understanding the relationship between these two proteins in human systems . In order to examine this interaction in detail we expressed FLAG–PKCα in HEK-293 cells in the presence or absence of the large isoform GFP–AKAP7γ. Immunoprecipitates from these cells isolated with an AKAP7 antibody contained PKCα only in cells expressing AKAP7 (Figure 1A). In a reciprocal experiment, cells were transfected with GFP–AKAP7γ in the presence or absence of PKC–FLAG. In agreement with our previous result, AKAP7γ was only found in immunoprecipitates from cells containing PKCα (Figure 1B). These experiments demonstrate a specific interaction between PKCα and AKAP7γ that is dependent upon co-expression of the two proteins. Given the conservation of AKAP7α within AKAP7γ (Figure 1C), we predicted that the smaller isoforms would also bind the enzyme. To test this, we repeated these experiments using AKAP7α and found that it too associated with PKC (Figure 1D). Since the interaction of PKC with scaffolding proteins is thought to be a major determinant of PKC activity, we looked at whether the PKC associated with AKAP7 retained its catalytic activity. To test this, we immunoprecipitated AKAP7γ complexes from co-transfected cells and performed PKC activity assays, which measure the incorporation of 32P into a PKC substrate peptide. We found the PKC activity within AKAP7 complexes to be 9-fold higher than background captured with an IgG-matched control antibody (Figure 1E). To confirm that PKC was the kinase underlying this result, we treated the AKAP7 immunoprecipitates with the PKC inhibitor BIS1 (bisindolylmaleimide I). Under these control conditions, we did not detect any PKC activity. Thus we conclude that the PKC associated with the AKAP7 complex is active. However, activity is not a prerequisite for association since a kinase-dead mutant of PKC can also bind AKAP7 (results not shown). Since these proteins are co-expressed in many different tissues, we were interested to know whether this interaction required an intermediary protein. To test this, we attempted to co-purify recombinant streptavidin-tagged AKAP7γ (AKAP–strep) and PKC–His in vitro using streptavadin–agarose beads. We observed that AKAP7–strep, but not strep-control, was able to bind directly to the kinase in our in vitro pull-down assay (Figure 1F). We also verified direct binding to the smaller isoform in pull-down assays using S-tagged AKAP7α (the S-tag sequence is KETAAAKFERQHMDS). We found PKCα in S-agarose pull downs charged with AKAP7α, but not with S-agarose control (Figure 1G). These results demonstrate a direct interaction between PKC and both isoforms of AKAP7.
PKC and AKAP7 isoforms directly interact in cells
Previous studies of the PKC-binding scaffolds AKAP12 (Gravin) and AKAP5 (AKAP79) have revealed an interesting paradox: although AKAP anchoring ultimately leads to greater substrate phosphorylation, their direct interaction actually inhibits the kinase until a cellular trigger initiates its release . To assess whether this inhibitory effect was also true of AKAP7, we incubated activated PKC with increasing amounts (2–10 μM) of either AKAP7γ or AKAP7α (Figure 2). On account of the fact that we observed no significant changes in PKC activity upon treatment with these high concentrations of AKAP7, we conclude that scaffolding by AKAP7 is not inhibitory. In this regard, AKAP7 is unique from the other PKC-binding AKAPs.
AKAP7 does not inhibit PKC activity
AKAP7 is a broad specificity PKC-binding protein
Given the structural differences among PKC enzymes, we hypothesized that AKAP7 might be selective for particular PKC isoenzymes. We tested this idea by measuring the kinase activity of various recombinant PKC isoenzymes captured by AKAP7γ–S-agarose in vitro. Our results from this series of experiments indicate significant activity from representative isoenzymes from all classes of PKC: the conventional class, PKCα and PKCβ (Figure 3A), the novel class, PKCδ and PKC∊ (Figure 3B), and the atypical isoenzyme PKCζ (Figure 3C). Importantly, this activity was not observed with an S-tagged control protein (AKAP79ΔPKC). Thus we conclude that, like AKAP12 and AKAP5, AKAP7 binds directly to multiple isoenzymes of PKC, but is unique in that it is not inhibitory to the enzyme. This broad specificity was also observed with AKAP7α (results not shown).
AKAP7 is a broad specificity PKC scaffolding protein
PKC binds the N-terminal and C-terminal regions of AKAP7γ
To examine the kinetic strength of the interaction between AKAP7 and PKC we elected to use SPR. We immobilized PKCα on a CM5 sensor chip to capture purified full-length AKAP7γ at various concentrations ranging from 12.5 to 400 nM. We confirmed a direct high-affinity interaction between the two proteins that, strikingly, was best fit using a heterogeneous analyte curve. This strongly suggests that there are two binding domains for PKC on AKAP7γ, with affinities of 29 nM and 14 nM (Figure 4A). To deduce the location of these binding domains, we generated a series of truncation mutants of S-tagged AKAP7γ and measured their associated PKC activity in pull-down experiments, as in Figure 3. We observed comparable amounts of PKC activity between AKAP7γ-(FL), AKAP7γ-(1–100), AKAP7γ-(1–150), AKAP7γ-(1–200) and AKAP7γ-(1–250), suggesting that one or both PKC-binding domains were contained within the first 100 amino acids of the AKAP (Figure 4B). To identify the location of the second domain, we cloned an AKAP7 mutant containing only the C-terminal half of the protein, AKAP7γ-(150–326). We found similar amounts of PKC activity in both halves of the protein, suggesting that one PKC-binding domain is in the C-terminal half of the protein, and the second lies in the N-terminal half (Figure 4C). To confirm this, we repeated our SPR experiment using these two fragments and found that the N-terminal half, AKAP7-(1–150), had a Kd of 13.5 nM that is best fit by a one-site binding curve (Figure 4D). Furthermore, the C-terminal half, AKAP7γ-(150–326), had a Kd of 27 nM that is also fit by a one-site model (Figure 4E). Importantly, the binding kinetics for each of these two fragments correlates with the two-site fit of the full-length protein determined in Figure 4(A). On the basis of our results from this series of experiments, we conclude that there are two high-affinity PKC interaction sites on AKAP7γ. This contrasts with AKAP7α, which our SPR analysis predicted to have a single site of interaction with PKC. Yet, despite this difference, the smaller isoform still has a strong affinity (23 nM) for the kinase (Figure 4F).
AKAP7γ and PKC bind with high affinity at multiple locations
To more narrowly define these regions of AKAP7α and AKAP7γ that interact with PKC we cloned several truncated AKAP7 mutants (Figure 5A, top panel) and tested their ability to compete for PKC binding. We found that excess AKAP7γ-(1–150) or AKAP7γ-(1–100) was sufficient to compete PKC away from AKAP7γ-(1–250), whereas AKAP7γ-(1–66) and AKAP7α were not (Figure 5A, bottom panel). These findings suggest that 34 amino acids (residues 66–100) of AKAP7γ are critical for the scaffolding of the enzyme, and this binding motif is not conserved within the smaller isoform. We attempted to use this same strategy to deduce the binding site of PKC on the conserved region between AKAP7γ and AKAP7α (Figure 5B, top panel). However, AKAP7α-(1–30), AKAP7α-(1–50) nor AKAP7α-(44–81) could robustly compete PKC away from this region (Figure 5B, bottom panel). This could indicate that the binding of PKC to AKAP7α requires several domains of the scaffold, perhaps to stabilize a vital secondary or tertiary structure that is lost in the deletion fragments.
PKC binds two unique domains in AKAP7
AKAP7γ binds regulatory and catalytic domains of PKC
Our early observation that multiple PKC isoenzymes interact with AKAP7 led us to hypothesize that the binding domain was contained within a conserved domain of PKC, most probably the highly homologous catalytic region. This is the region where AKAP5 (AKAP79/150), another PKC scaffold, is known to associate . Because PKC is a trypsin-labile enzyme, cleavage with this protease liberates the catalytic domain of the enzyme from the N-terminal regulatory regions . Therefore, to test our hypothesis, we attempted to pull down trypsinized PKC(cat)in vitro using S-tagged AKAP7γ. We observed robust PKC activity from the catalytic domain captured in these pull downs, confirming an interaction between AKAP7γ and the catalytic domain of PKC (Figure 6A). This was also observed in analogous experiments using AKAP7α (results not shown). However, in light of the fact that there are two sites of contact on AKAP7γ, it was plausible that each site could associate with a different region of PKC. To test this, we repeated this assay using our AKAP7γ-(1–150), and AKAP7γ-(150–326) mutants. We found that binding to the catalytic core was unique to the C-terminal fragment of AKAP7γ, since we detected comparable activity between this fragment and the full-length protein, but no activity in pull downs using the N-terminal fragment (Figure 6A).
AKAP7 binds the regulatory and catalytic domains of PKC
In light of these results, we endeavoured to find the location of the second binding site within the regulatory region of PKC that interacts with the N-terminal half of AKAP7γ-(1–150). The regulatory region comprises a C1 (lipid-binding) and C2 (Ca2+-binding) domain. We obtained GST (glutathione transferase)-tagged C1 and C2 domains of PKCβII and attempted to pull down purified AKAP7γ. Western blot analysis identified AKAP7γ in the C1–GST, but not C2–GST, pull downs (Figure 6B). To confirm that it was the N-terminal binding site mediating this interaction, we attempted to pull down both the AKAP7γ-(1–150) mutant and the AKAP7γ-(150–326) mutant with C1–GST. As expected, only the N-terminal fragment could be found in the pull downs (Figure 6C). These mapping data demonstrate two binding domains on PKC for AKAP7γ, one in the catalytic domain of the enzyme for the C-terminus of AKAP7γ and the second in the C1 regulatory region for the N-terminus of AKAP7γ (Figure 6D).
Anchoring amplifies the phosphorylation of substrates
The regulation of PKC by AKAP scaffolding proteins is known to be an important determinant of the activity of the enzyme. For example, the inhibition of PKCα by AKAP12 is thought to be an important component of the cytoskeletal and morphological changes that occur during the contact inhibition process . In contrast, the enhancement of PKC activity by AKAP5 is known to enhance the phosphorylation and sensitization of neuronal M-type KCNQ channels . Since AKAP7 does not inhibit PKC, we hypothesized that its function is to direct and enhance kinase activity towards substrates within its signalling complex, akin to AKAP5. We pursued this hypothesis using confocal FRET microscopy and a genetic PKC activity reporter (CKAR). This construct has been extensively characterized as a specific reporter of PKC activity that undergoes FRET in a dephosphorylated state, but loses this ability upon its phosphorylation by PKC . We created a signalling complex by fusing CKAR to AKAP7γ. This strategy allowed us to compare phosphorylation of a freely diffusible enzyme, PKC, and substrate, CKAR, to those sequestered within a pre-formed AKAP complex. We validated our fusion construct by Western blotting extracts from transfected cells. Using an antibody directed against GFP we identified a protein of ~120 kDa, which is consistent with the predicted combined molecular mass of CKAR (75 kDa) and AKAP7γ (44 kDa) (Figure 7A). Probing these same extracts with an antibody recognizing AKAP7, we found a protein of identical molecular mass in the cells expressing the AKAP7γ–CKAR fusion protein (Figure 7B). Both of these constructs exhibit a cellular diffuse localization, consistent with previous data on both CKAR and AKAP7γ (Figure 7C) [15,27]. Following maximal stimulation of these two constructs with PDBu (phorbol 12,12-dibutyrate), we observed a 3.86±0.64% change in FRET using the free CKAR reporter over the course of 15 min. Strikingly, the AKAP7γ–CKAR reporter exhibited a 2-fold greater change in FRET magnitude (8.37±0.57%) over an equivalent observation period (Figure 7D). This result demonstrates that, within the cellular milieu, anchored substrates have a distinct phosphorylation advantage over those limited by diffusive kinetics. Data from a recent study using this same experimental approach with AKAP5 additionally demonstrated a decrease in the t1/2max of anchored CKAR phosphorylation compared with free CKAR . Interestingly, this acceleration is not reflected in our AKAP7γ–CKAR data. This discrepancy could potentially be explained by differential targeting between the two AKAPs, although it is perhaps more reflective of a ‘lag time’ for second messenger diffusion to these cytosolic species prior to activation of the enzyme.
AKAP7γ enhances substrate phosphorylation
AKAP7 sequesters PKC in cellular compartments
The spatial restriction of kinases is also thought to be a critical regulatory mechanism. Our experiments using untargeted FRET sensors demonstrated that substrates co-localized with PKC in a signalling complex are preferentially phosphorylated. In light of this, the subcellular tethering of PKC by an anchoring protein is likely to alter the phosphorylation profile of the enzyme by changing the identities of the available substrates. PKC undergoes a step-wise maturation process that yields a catalytically competent enzyme with a cytosolic localization . Under stimulated conditions, the production of DAG, phospholipids and Ca2+ shuttles PKC to membranes where it becomes active. Consistent with this dogma, in a FRAP analysis we observed a rapid redistribution of PKCβII–YFP into a photobleached region of interest, suggesting that this inactive cytosolic species of PKC is highly diffusible within the cell. We created a simulation using VFRAP (virtual FRAP) software (Figure 8A) to fit these data, and determined that 89±2% of the PKCβII–YFP within these cells is highly mobile (Figures 8B and 8C). Furthermore, this species of PKCβII redistributes with an effective diffusion coefficient of 2.54±0.41 μm2·s−1 (Figure 8D). We hypothesized that AKAP7α, a membrane-targeted isoform of AKAP7, would sequester PKC and restrict its mobility within the cell. Indeed, in cells co-transfected with AKAP7α, FRAP analysis revealed the mobile fraction of PKC to be reduced by 14±3% (Figure 8C). Furthermore, the effective diffusion coefficient was reduced nearly 3-fold to 0.91±0.22 μm2·s−1 (Figure 8D). This finding supports the idea that AKAP7 traps a pool of PKC within a subcellular compartment, obliging the phosphorylation of the local substrates in that vicinity. Experiments using AKAP7γ were inconclusive owing to the fact that approximately half of the cells analysed could be fit with a two-component recovery curve. This two-component fit is probably due to diffusion of both the AKAP7γ–PKC complexes and the free enzyme in the cytosol, although the fact that we could not uniformly fit the data confounded our analysis.
AKAP7 anchors PKC within cells
The characterization of AKAP7 as a broad specificity PKC-binding protein could have important implications for the regulation of the enzyme in many tissues. Unlike several other PKC scaffolds, such as AKAP12, caveolin-1 and PICOT, which inhibit the kinase, AKAP7 tethers an active enzyme. This anchoring to AKAP7 enhances the phosphorylation of intracomplex substrates [25,30,31]. Furthermore, AKAP7α appears to be important for the subcellular targeting of the enzyme since anchored PKC exhibits reduced mobility within the cell. Collectively, these findings support a model wherein AKAP7 tethers PKC within a local signalling complex, where it is poised to phosphorylate its specific downstream effectors.
cyan fluorescent protein
protein kinase C activity reporter
fluorescence recovery after photobleaching
fluorescence resonance energy transfer
green fluorescent protein
human embryonic kidney
cAMP-dependent protein kinase
protein kinase C
surface plasmon resonance
yellow fluorescent protein
John Redden designed experiments, performed the FRET and FRAP experiments, generated reagents, analysed data, and wrote the paper. Andrew Le performed the BIAcore experiments and analysed data. Arpita Singh, Kyle Federkiewicz and Samantha Smith provided technical assistance, generated reagents and helped perform experiments. Kimberly Dodge-Kafka conceived the study, designed experiments, performed the pull-down experiments and PKC activity assays, analysed data, and assisted in writing the paper.
We thank Dr Ann Cowan for help with the use of the VCell and VFRAP software, and thank Dr Laurinda Jaffe for use of the Zeiss Pascal confocal microscope.
This work was supported by the National Institutes of Health [grant number HL82705 (to K.L.D.-K.)] and the State of Connecticut Department of Public Health [grant number 2011-0142 (to K.L.D.-K.)], and an American Heart Association (AHA) Predoctoral Fellowship [grant number 11PRE7830027 (to J.M.R.)]. The Virtual Cell is supported by the National Institutes of Health [grant number P41RR013186] from the National Center For Research Resources.