Propofol Activates and Allosterically Modulates Recombinant Protein Kinase Cϵ

The recombinant PKCϵ protein was expressed using a baculovirus vector and purified from Sf  9 insect cells that normally do not express any detectable endogenous PKC activity.12,13Therefore, the recombinant PKCϵ protein used in these studies was free of contamination by any other PKC isoforms. Immunoblot analysis was carried out on recombinant PKCϵ as previously described for PKC isoforms.14Protein concentration was assessed using the Bradford method.15Equal amounts of protein (50 μg) were electrophoresed on 12% sodium dodecyl sulfate-polyacrylamide gels and transferred to nitrocellulose membranes. Nonspecific binding was blocked with Tris-buffered saline solution (0.1% [v/v] Tween-20 in 20 mm Tris base, 137 mm NaCl adjusted to pH 7.6 with HCl, containing 3% [w/v] bovine serum albumin) for 1 h at room temperature. Polyclonal antibodies against PKCϵ (holoenzyme) and the autophosphorylation site at serine 729 (PKCϵpS⁷²⁹) were diluted 1:1000 in Tris-buffered saline solution containing 1% bovine serum albumin for immunoblotting (2 h). After washing in Tris-buffered saline solution 3 times (10 min each), filters were incubated for 1 h at room temperature with horseradish peroxidase–linked secondary antibody (1:5,000 dilution in Tris-buffered saline solution containing 1% bovine serum albumin). Filters were again washed, and bound antibody was detected by the enhanced chemiluminescence method. The density of the individual bands was analyzed using National Institutes of Health Image software (National Institutes of Health, Washington D.C.).

PKCϵ was immunopurified from phosphatase-treated recombinant PKCϵ (pt-PKCϵ) using an Amino-Link Plus Immobilization Kit (Pierce Biotechnology Inc., Rockford, IL) per manufacturer’s instructions. Briefly, 0.5–1.5 ml of cell lysate, diluted 1:1 in sample buffer (0.1 M phosphate, 0.15 M NaCl, pH 7.2), was applied to an Amino-Link Plus- PKCϵ affinity column, washed, and equilibrated at room temperature. The column was then incubated at room temperature for 60 min or overnight at 4°C, washed, and eluted with 8–10 ml of ImmunoPure immunoglobulin G elution buffer; 0.5-ml elution fractions were collected and monitored by absorbance at 280 nm.

The activity of recombinant PKCϵ before and after treatment with propofol was measured using a colorimetric PKC activity assay kit (Stressgen Bioreagents, Victoria, British Columbia, Canada) per manufacturer’s instructions. Briefly, a readily PKC phosphorylated substrate (cyclic adenosine monophosphate response element binding protein) was precoated onto the wells of a PKC substrate microtiter plate provided in the kit. The recombinant protein was added to the wells, and activation of the enzyme was initiated by addition of adenosine triphosphate in the presence of phosphatidylserine (200 μg/ml). After 90 min of incubation at 30°C, the reaction was terminated by emptying the contents of each well. A phospho-specific substrate antibody was then added to each well followed by a peroxidase-conjugated antirabbit immunoglobulin secondary antibody. After incubation (30 min, 23°C) and four washes, tetramethylbenzidine substrate was added to develop the reaction. The developing reaction was terminated after 45 min with acid stop solution (2 N HCl). The intensity of the color was measured on a microplate reader at 450 nm and the relative kinase activity (compared to untreated, baseline controls) of the samples was calculated from the absorbance measurements.

The PKCϵ C1B subdomain was synthesized and purified by high-performance liquid chromatography by the Molecular Biotechnology Core facility located in the Lerner Research Institute at Cleveland Clinic. Fluorescence measurements were performed on a Photon Technology International (Birmingham, NJ) spectrofluorometer equipped with temperature and stirring control for cuvette (1.5-ml quartz)–based studies. Recombinant pt-PKCϵ, C1B subdomain, and sapintoxin D (SAPD, at varying molar ratios) were mixed in a buffer (50 mm Tris, pH 7.2) at a final volume of 1 ml with gentle stirring (20 min; 23°C). The molecular interactions between SAPD (0.2 μm) and the C1B subdomain (2 μm) were assessed by evaluating shifts in the SAPD fluorescence emission spectra. Deconvolution of the spectra obtained under these conditions (10:1 molar ratio of protein to SAPD) was used to delineate differences in the peaks for the protein-bound SAPD (405 nm) from the aqueous SAPD (437 nm).16To assess allosteric modulation of SAPD binding to the C1B subdomain by propofol, the molar ratio of protein to SAPD was reduced from 10:1 to 2:1 (2 μm C1B subdomain; 1 μm SAPD). For both protocols, emission spectra were acquired from 375 to 575 nm at an excitation wavelength of 355 nm. The C1B subdomain contains 1 tryptophan (Trp 264), affording intrinsic fluorescence to the protein. To assess the effects of propofol or phorbol myristic acid (PMA) on the fluorescence emission spectra of the C1B subdomain, 2-μl aliquots of propofol or PMA (1 μm stock solutions) were titrated against the C1B subdomain (1 μm). Spectra were recorded after incubation (30 min) with slow stirring. Emission spectra were acquired from 310 nm to 480 nm at an excitation wavelength of 290 nm.

Changes in pt-PKCϵ autophosphorylation, activity, or relative fluorescence were compared to baseline values (normalized to 100%) using one-way analysis of variance with repeated measures and the Bonferroni post hoc  test. Differences were considered statistically significant at P < 0.05. All results are expressed as means ± SD. Statistical analysis was conducted using NCSS software (Kaysville, UT).

Western blot analysis of recombinant PKCϵ before and after treatment (37°C, 30 min) with PMA (1 μm), propofol (0.1–10 μm), or intralipid (intralipid; 0.1–10 μm) was performed using an antibody (PKCϵpS⁷²⁹) recognizing the autophosphorylation site (serine 729) on PKCϵ. Because of significant autophosphorylation and constituitive activity of the holoenzyme under baseline conditions (see fig. 1), all subsequent protocols were performed after the recombinant PKCϵ was subjected to λ phosphatase (200 units, 30 min, 30°C) to remove the phosphate from serine 729, which is required for catalytic activity of the enzyme. Therefore, λ phosphatase treatment significantly reduces baseline activity of the recombinant protein in the absence of any putative activators, allowing for a greater difference in our ability to assess an increase in catalytic activity by the interventions when compared to baseline. PKCϵ was then immunopurified by affinity chromatography and referred to as pt-PKCϵ throughout the text. Baseline autophosphorylation in the absence of any intervention was considered the control value and was normalized to 100%. Summarized data are expressed as the ratio of total PKCϵ to pPKCϵ (percent of baseline control).

Activity of the immunopurified pt-PKCϵ (50 ng) was assessed by colorimetric determination of cyclic adenosine monophosphate response element binding protein (CREB) phosphorylation. Propofol, intralipid, or the PKC activators dioctanoylglycerol (dioctanoyl-glycerol; 50 μm) or PMA (0.1 μm) were added alone or in combination to pt-PKCϵ and CREB (90 min). pt-PKCϵ activity was assessed on a microplate reader by assessing absorbance of the colored derivative at 450 nm. Baseline pt-PKCϵ activity in the absence of any intervention was considered the control value and was normalized to 100%. Summarized data are expressed as a percent of baseline control.

The fluorescence emission spectra of SAPD (0.2 μm) was assessed in 50 mm Tris buffer (pH 7.4) before and after addition of a 10-fold excess of PKCϵ C1B subdomain (2 μm; 30 min). SAPD was excited at 355 nm, and emission spectra were collected from 375 nm to 575 nm. Control experiments were also performed to assess the effect of denaturing by boiling (90°C, 20 min) the PKCϵ C1B subdomain on SAPD fluorescence emission spectra.

We monitored the intrinsic fluorescence of the PKCϵ C1B subdomain during titration with propofol, intralipid, or PMA as an indicator of direct binding to the protein. Fluorescence emission spectra of the PKCϵ C1B subdomain (1 μm) were assessed in 50 mm Tris buffer (pH 7.4) before and after successive additions of PMA (0.1–1 μm), propofol (0.1–1.0 μm), or intralipid (1 μm). Excitation was 290 nm, and emission spectra were collected from 310 to 480 nm.

Experiments were performed as described under Protocol 3, except the molar ratio of SAPD to PKCϵ C1B subdomain was reduced from 10:1 to 2:1. Fluorescence emission spectra for SAPD alone (1 μm) with and without propofol (1 μm) and/or the PKCϵ C1B subdomain (2 μm) were recorded as described in Protocol 3. Control experiments were also performed to assess the effect of intralipid on the fluorescence emission spectra for SAPD alone and for SAPD in the presence of PKCϵ C1B subdomain.

Recombinant PKCϵ expressed and purified from insect Sf  9 cells was obtained from Panvera (Carlsbad, CA). Monoclonal antibodies for PKCϵ and PKCϵpS⁷²⁹were purchased from Cell Signaling (Beverly, MA). PMA and dioctanoyl-glycerol were purchased from Sigma Chemical Co. (St. Louis, MO). SAPD was obtained from Alexis (San Diego, CA). The PKCϵ C1B subdomain was synthesized and purified by high-performance liquid chromatography by the Molecular Biotechnology Core facility (Lerner Research Institute at Cleveland Clinic, Cleveland Ohio). Propofol and the intralipid vehicle were obtained from Cleveland Clinic Pharmacy.

Western blot analysis of recombinant PKCϵ before and after treatment (37°C, 30 min) with intralipid, PMA, or propofol was performed using anti-PKCϵpS⁷²⁹. In the absence of any intervention, recombinant PKCϵ exhibited some degree of autophosphorylation (fig. 1A). PMA and propofol, but not intralipid, increased the amount of detectable recombinant PKCϵpS⁷²⁹compared to baseline. Because of the high degree of baseline auto-phosphorylation of recombinant PKCϵ, we incubated recombinant PKCϵ with λ phosphatase (200 units, 30 min, 30°C), and the pt- PKCϵ was then immunopurified using affinity chromatography. The pt-PKCϵ exhibited minimal autophosphorylation (fig. 1B). intralipid, PMA, and propofol markedly increased autophosphorylation of pt-PKCϵ. The summarized data for the effects of intralipid, PMA, and propofol on autophosphorylation of PKCϵ or pt-PKCϵ are presented in panels C and D, respectively.

We assessed the extent to which propofol, intralipid, and classic PKC activators increase activity of pt-PKCϵ. Activation of pt-PKCϵ was assessed, measuring the extent to which the enzyme phosphorylates CREB. Propofol, intralipid, or the PKC activators dioctanoyl-glycerol or PMA were added alone and in combination with pt-PKCϵ into the wells containing the substrate, CREB. Propofol caused a dose-dependent increase in phosphorylation of CREB (fig. 2A). intralipid alone also caused modest phosphorylation of CREB at high concentrations. PMA and dioctanoyl-glycerol increased phosphorylation of CREB to a similar extent (fig. 2B). Pretreatment with dioctanoyl-glycerol or PMA potentiated propofol-stimulated (1 μm) pt-PKCϵ–dependent phosphorylation of CREB.

In 50 mm Tris buffer (pH 7.4), SAPD (0.2 μm) has an emission maximum (λmax) of 437 nm (fig. 3, trace A). Addition of the PKCϵ C1B subdomain (2 μm) to SAPD (fig. 3, trace B) caused a shift in the emission spectra that could be deconvoluted into two components with a λmax of 405 nm and 437 nm representing bound and free SAPD, respectively (fig. 3, traces C and D). Boiling (90°C; 30 min) the C1B:SAPD solution resulted in an emission spectra with a single λmax centered at 437 nm (data not shown). Similarly, a single λmax of 437 nm was observed when the C1B subdomain was boiled (90°C, 20 min) before mixing with SAPD (see inset, fig. 3).

The PKCϵ C1B subdomain contains a single tryptophan (Trp-264), which showed a λmax of 348 nm (fig. 4A, left). The intrinsic fluorescence of PKCϵ C1B subdomain (1 μm) was quenched by successive addition of 0.1 μm propofol (fig. 4A, left, lower traces) in a concentration–dependent manner. Summarized data for the concentration dependent effects of propofol (0.1–1.0 μm) and the effect of intralipid (1 μm) are depicted in the right panel of figure 4A.

Similarly, the intrinsic fluorescence of PKCϵ C1B subdomain (1 μm) was quenched by successive addition of 0.1 μm PMA (fig. 4B, left) in a concentration–dependent manner. Summarized data for the concentration dependent effects of PMA (0.1–1.0 μm) are depicted in the right panel of figure 4B.

As shown in figure 3, the emission spectra of SAPD in the presence of a 10-fold excess of protein consists of overlapping free and bound contributions with a λmax of 437 and 405, respectively. However, when we performed experiments in which the C1B (2 μm) to SAPD (1 μm) molar ratio was reduced from 10:1 to 2:1, the composite peaks were more symmetrical and exhibited a λmax of about 437 nm for the free and 425 nm for the bound (fig. 5, traces A and C, respectively). Propofol did not have any effect on the SAPD emission spectra in the absence of the PKCϵ C1B subdomain (fig. 5, trace B). However, in the presence of the PKCϵ C1B subdomain, addition of propofol (1 μm) caused more SAPD to bind, resulting in an even further leftward shift that had a λmax of about 415 nm (fig. 5, trace D). In contrast, intralipid (1 μm) had no effect on the SAPD λmax alone or in the presence of the PKCϵ C1B subdomain (see inset, fig. 5). Summarized data for the effects of propofol (1 μm), intralipid (1 μm), and PKCϵ C1B subdomain on SAPD emission spectra are depicted in figure 5B.

This is the first study to directly assess the effects of propofol on molecular mechanisms regulating activation of purified, recombinant PKCϵ. Previous studies using purified rat brain PKC under various assay conditions have demonstrated that propofol and halothane can stimulate PKC activity17; however, the PKC isoforms affected and molecular mechanisms of activation have not been identified. These studies also did not address the potential for anesthetic-induced, allosteric modulation of PKC activity by phorbol esters and/or other lipid activators such as dioctanoyl-glycerol or free fatty acids.17A recent study from our laboratory demonstrated that propofol activates PKCα, PKCδ, PKCϵ, and PKCζ, resulting in their translocation to distinct intracellular sites in cardiomyocytes.14In addition, previous studies from our laboratory using pharmacological inhibitors of PKC have demonstrated propofol-induced, PKC-dependent alterations in cardiomyocyte intracellular free Ca²⁺concentration and shortening,18α adrenoreceptor,19and β adrenoreceptor20signal transduction, phosphorylation of contractile proteins,21Na⁺-H⁺activity, and intracellular pH and myofilament Ca²⁺sensitivity.22The key finding of the current study is that propofol activates and allosterically modulates activation of purified, recombinant PKCϵvia  a molecular interaction with PKCϵ at or near the phorbol ester/diacylglycerol binding domain (C1A and/or C1B subdomain). A schematic diagram depicting domains of the PKCϵ holoenzyme and proposed molecular interaction between propofol and the PKCϵ C1B subdomain is shown in figure 6.

A key step in the activation process of PKC isoforms is the binding of endogenous mediators (diacylglycerol, free fatty acids) at a tandem repeat of cysteine-rich zinc finger motifs (C1A and C1B) contained in the regulatory domain.6Binding of diacylglycerol causes activation by the enzyme by releasing the pseudosubstrate domain from the catalytic site and allowing for autophosphorylation of the enzyme on serine 729 in the C terminal hydrophobic motif.23We identified that the recombinant PKCϵ used in these studies exhibits a significant degree of autophosphorylation at serine 729 that is markedly reduced by treatment with λ phosphatase. Moreover, the extent to which PMA and propofol stimulate autophosphorylation of recombinant PKCϵ is significantly enhanced after dephosphorylation of serine 729 with λ phosphatase. The intralipid at concentrations 10 times higher than propofol was also capable of stimulating some autophosphorylation. This is likely due to the presence of fatty acids (soybean oil) and phospholipids (lecithin), which create the lipid emulsion facilitating solubility of propofol. Fatty acids are known to activate some isoforms of PKC, and phospholipids serve as cofactors for activation.24These data provide molecular evidence for an interaction between propofol and recombinant PKCϵ and support our previous pharmacological studies demonstrating propofol-induced, PKC-dependent alterations in cardiomyocyte signal transduction and function.14,20–22,25 

Autophosphorylation of PKCϵ at serine 729 is an important step in the activation of PKCϵ.23To confirm that the propofol-induced increase in autophosphorylation at serine 729 correlated with catalytic activity of the enzyme, we directly measured pt-PKCϵ activity by assessing the ability of propofol to stimulate PKCϵ-dependent phosphorylation CREB. Although time course studies were not performed, we have identified that propofol stimulates a dose-dependent increase in catalytic activity of recombinant pt-PKCϵ, providing a direct correlation between propofol-induced autophosphorylation at serine 729 and catalytic activity of the enzyme. The intralipid vehicle also stimulates some catalytic activity at high concentrations consistent with the modest autophosphorylation observed at serine 729. Moreover, propofol potentiated PMA and dioctanoyl-glycerol-induced recombinant pt-PKCϵ activity, suggesting that propofol can allosterically modulate catalytic activity induced by other known classic activators of most PKC isoforms, including PKCϵ. Taken together with the autophosphorylation data, these results indicate that propofol increases pt-PKCϵ activity via  an intermolecular interaction with the recombinant protein. Moreover, propofol potentiates pt-PKCϵ activity after pretreatment with the classic activators, dioctanoyl-glycerol or PMA.

In an effort to establish the efficacy of our experimental methodology with regards to accurately measuring changes in fluorescence emission spectra, we first assessed the effect of PKCϵ C1B subdomain on fluorescence emission spectra of SAPD. Using SAPD, we determined that the λmax for the emission spectrum of this fluorescent phorbol ester was quenched and exhibited a left shift upon incubation with the PKCϵ C1B subdomain protein and that could be deconvoluted into two peaks representing bound and free SAPD. The data in figure 3demonstrate an intermolecular interaction between SAPD and the PKCϵ C1B subdomain, confirming the sensitivity of the experimental approach to detect changes in the fluorescence emission spectra of SAPD.

We also took advantage of the tryptophan (Trp 264) that affords fluorescence to the C1B subdomain protein and examined the extent to which propofol or PMA (nonfluorescent) modify the C1B subdomain fluorescence emission spectra. Our findings indicate that propofol and PMA exert similar quenching effects on the emission spectra of the C1B subdomain. These data support a intermolecular interaction of propofol with the C1B subdomain similar to that observed with PMA and suggest that this is the likely mechanism to explain propofol-induced activation of PKCϵ. These findings are also consistent with a study suggesting the interaction of general anesthetic alcohols with the C1B subdomain of PKCδ,16which has greater than 60% conserved homology with the PKCϵ C1B subdomain. We propose a molecular interaction between the hydrophobic isopropyl groups of propofol with hydrophobic residues on the C1B subdomain near Tyr 250 or Trp 264. The ribbon diagram for the C1B subdomain (fig. 6) illustrates a potential binding pocket for propofol in close proximity to the Tyr 250 and Trp 264 residues where interactions between the hydrophobic side chains could occur, resulting in activation of PKCϵ or allosteric modulation of diacylglycerol binding.

We performed experiments to further assess the possibility that propofol allosterically modulates enzymatic activity by examining whether propofol alters SAPD binding to the PKCϵ C1B subdomain. Our data indicate that, when the enzyme to SAPD molar ratio was reduced from 10:1 to 2:1, addition of propofol caused a further leftward shift in the λmax for SAPD. In addition, propofol had no effect on SAPD fluorescence emission spectra in the absence of the PKCϵ C1B subdomain. These data suggest that propofol allosterically modulates SAPD binding to the PKCϵ C1B subdomain. Moreover, these data also suggest that propofol likely interacts at a discrete allosteric binding site on the PKCϵ C1B subdomain and synergistically enhances the binding of SAPD. This may explain our observations that propofol potentiates PMA- or diacylglycerol-induced activity of recombinant PKCϵ. Moreover, the fact that intralipid had no effect on the emission spectra suggests that the ability of the intralipid to activate recombinant PKCϵ or cause autophosphorylation at Ser 729 is likely due to a molecular interaction of the intralipid at yet another discrete binding site on the PKCϵ holoenzyme. It should be noted that the data in this study does not fully characterize whether the molecular interaction between propofol and PKCϵ is direct or allosterically mediated. Our results indicate a molecular interaction between propofol and PKCϵ, but additional studies incorporating time-resolved analysis of anisotropy would be needed to further delineate the precise nature of this interaction.

These results demonstrate that propofol interacts with the C1B subdomain of PKCϵ resulting in autophosphorylation and activation of the enzyme. Moreover, propofol enhances phorbol ester binding to the C1B subdomain, indicating that propofol can also allosterically modulate enzymatic activity.