Propofol induces excessive vasodilation of aortic rings by inhibiting protein kinase Cβ2 and θ in spontaneously hypertensive rats

Abstract

Background and Purpose

Exaggerated hypotension following administration of propofol is strongly predicted in patients with hypertension. Increased PKCs play a crucial role in regulating vascular tone. We studied whether propofol induces vasodilation by inhibiting increased PKC activity in spontaneously hypertensive rats (SHRs) and, if so, whether contractile Ca²⁺ sensitization pathways and filamentous–globular (F/G) actin dynamics were involved.

Experimental Approach

Rings of thoracic aorta, denuded of endothelium, from normotensive Wistar‐Kyoto (WKY) rats and SHR were prepared for functional studies. Expression and activity of PKCs in vascular smooth muscle (VSM) cells were determined by Western blot analysis and elisa respectively. Phosphorylation of the key proteins in PKC Ca²⁺ sensitization pathways was also examined. Actin polymerization was evaluated by differential centrifugation to probe G‐ and F‐actin content.

Key Results

Basal expression and activity of PKCβ2 and PKCθ were increased in aortic VSMs of SHR, compared with those from WKY rats. Vasorelaxation of SHR aortas by propofol was markedly attenuated by LY333531 (a specific PKCβ inhibitor) or the PKCθ pseudo‐substrate inhibitor. Furthermore, noradrenaline‐enhanced phosphorylation, and the translocation of PKCβ2 and PKCθ, was inhibited by propofol, with decreased actin polymerization and PKCβ2‐mediated Ca²⁺ sensitization pathway in SHR aortas.

Conclusion and Implications

Propofol suppressed increased PKCβ2 and PKCθ activity, which was partly responsible for exaggerated vasodilation in SHR. This suppression results in inhibition of actin polymerization, as well as that of the PKCβ2‐ but not PKCθ‐mediated, Ca²⁺ sensitization pathway. These data provide a novel explanation for the unwanted side effects of propofol.

Abbreviations

CPI‐17

PKC‐potentiated phosphatase inhibitor protein‐17 kDa

F

filamentous

G

globular

KRS

Krebs–Ringer solution

MARCKS

myristoylated alanine‐rich C kinase substrate

MLC

myosin light chain

MLCK

MLC kinase

MLCP

MLC phosphatase

MOI

multiplicity of infection

MYPT1

myosin phosphatase target subunit 1

PDBu

phorbol 12,13‐dibutyrate

PPS

PKCθ pseudo‐substrate inhibitor

SHR

spontaneously hypertensive rat

VSM

vascular smooth muscle

VSMC

VSM cell

WKY

Wistar‐Kyoto rat

Introduction

The prevalence of hypertension increases over the years in China. As a result, an increasing number of patients with hypertension are observed perioperatively under the care of anaesthetists. Hypertension is associated with increased sensitivity to the effects of drugs on cardiovascular system (Cortes et al., 1997; Pellegrino et al., 2016), leading to considerable haemodynamic instability with an increased risk of perioperative morbidity and mortality. Propofol, an intravenous general anaesthetic, is widely used by anaesthetists. However, in patients with hypertension, the use of propofol is a strongly associated with an exaggerated hypotensive response. This hypotension may be due to its direct vasodilation (Gragasin et al., 2013), myocardial depression (Yang et al., 2015) and decrease in sympathetic activity (Ebert, 2005). At vascular smooth muscle (VSM) levels, the direct relaxant action of propofol on the vasculature has been suggested to be mediated through the PKC‐regulated, contractile Ca²⁺ sensitization pathway (Kuriyama et al., 2012).

PKCs are generally classified into three groups: conventional (cPKCs: α, β1, β2 and γ), novel (nPKCs: δ, θ, ε and η) and atypical (λ and ζ). Substantial evidence has confirmed that PKCs play pivotal roles in VSM contraction. VSM tension induced by vasoconstrictors is closely related to the level of reversible myosin light chain (MLC) phosphorylation, which is determined by the balance between Ca²⁺/calmodulin‐dependent MLC kinase (MLCK) and MLC phosphatase (MLCP). At a given intracellular Ca²⁺ concentration, PKCs phosphorylate PKC‐potentiated phosphatase inhibitor protein‐17 kDa (CPI‐17) at Thr³⁸, which induces rapid inhibition of MLCP and increased MLC phosphorylation, causing VSM contraction (Kitazawa and Kitazawa, 2012; Khalil, 2013). Recent findings suggest that the contractile Ca²⁺‐sensitive mechanism of PKC plays critical roles in the enhanced vascular tone in hypertension (Salamanca and Khalil, 2005; Khalil, 2013). The involvement of PKC in hypertension is demonstrated by the observation that the increases in expression/activity of some PKC subtypes such as PKCα, PKCε and PKCδ may elicit increases in VSM growth and hypertrophic remodelling, with associated enhanced vasoconstriction and increased vascular resistance and BP (Khalil et al., 1992; Lim et al., 2014; Jackson et al., 2016).

Although inhibition of Ca²⁺ mobility by propofol has been demonstrated to contribute to the drug‐induced relaxation in VSM in hypertension (Kuriyama et al., 2012; Lawton et al., 2012), PKCs may have many effects on drug‐induced changes in vascular tension. Activation of PKCs in endothelium mediates thrombin and propofol‐enhanced NO bioavailability (Motley et al., 2007; Wang et al., 2010), whereas the activation of PKCs in VSM mediates agonist‐induced contraction (Kitazawa and Kitazawa, 2012; El‐Yazbi et al., 2015). In addition, recent findings suggest that PKC is also present in dynamic reorganization of the actin cytoskeleton in VSM. Actin polymerization plays an important role in VSM contraction, independent of the phosphorylation of myosin light chain (MLC) and of raised [Ca²⁺]_i (El‐Yazbi et al., 2015). Given the pleiotropic roles of PKCs in vascular contraction, we investigated the individual roles of PKCs in propofol‐induced relaxation in aorta VSM in spontaneously hypertensive rats (SHR) and the normotensive WKY strain.

Methods

Animals

All animal care and experimental protocols conformed to the Guide for the Care and Use of Laboratory Animals (eigth edition) published by the National Research Council (United States) and were approved by the Institutional Animal Care and Use Committee of Shanghai Jiaotong University. Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010; McGrath and Lilley, 2015). We have used the minimum possible number of animals to achieve statistical significance.

Adult male Wistar‐Kyoto rats (WKYs) and SHRs (Vital River Laboratory Animal Technology Co. Ltd. Division of Charles River Laboratory International Inc., Beijing, China) weighing 250–300 g were housed in group cages under controlled illumination (12:12 h light–dark cycle), humidity and temperature (22–26°C) and had free access to tap water and standard rat chow. SHR strain is the most widely studied animal model of human hypertension (Rapp, 2000). BP, at room temperature, was non‐invasively measured in conscious rats by a tail cuff and pulse transducer system (BP‐98A; Softron, Tokyo, Japan).

Organ bath experiments

SHR (12 weeks old, 250–300 g) and age‐matched WKY rats were anaesthetized by pentobarbital sodium. The thoracic aorta was then cleared of surrounding tissues and excised from the aortic arch to the diaphragm. The prepared aorta was stored in cold Krebs–Ringer solution (KRS) aerated with 95% oxygen (O₂) and 5% carbon dioxide (CO₂) to obtain a pH of 7.4. The constituents of KRS were as follows (in mM): 143 Na⁺, 4.6 K⁺, 126.4 Cl⁻, 2.5 Ca²⁺, 25.0 HCO₃ ⁻, 0.79 SO₄ ²⁻, 1.2 H₂PO⁴⁻, 5.5 glucose and 0.024 EDTA. From each vessel, adherent tissues were removed and a clean cylindrical ring was cut into several segments (approximately 3 mm each in length). For endothelium‐denuded rings, the endothelium was removed by gently rubbing the intimal surface of the ring with a pair of small forceps. Each ring was suspended on two L‐shaped hooks in 5 mL organ baths filled with normal KRS, which was oxygenated with 95% O₂ and 5% CO₂ and maintained at 37°C (pH 7.4). One of the two L‐shaped hooks was attached to a force–displacement transducer (Danish Myo Technology A/S, Aarhus, Denmark) to measure the isometric force. The rings were stretched to an optimal tension of 1.5 g and then allowed to equilibrate for 60 min. After the stabilization period, contractile responses to 60 mM KCl were used as control at the beginning and end of each experiment. After removing the KCl, a stable contraction was induced with phenylephrine (1 μM). The removal of the endothelium from the aorta rings were confirmed by a lack of ACh (1 μM)‐induced relaxation (<10% relaxation), while the endothelium‐intact rings showed >80% relaxation with ACh. The rings were thereafter rinsed in pre‐warmed KRS several times until the baseline tension was restored, and then each series of experiments was started. Detailed methods of organ bath experiments are available in the Supporting Information.

Cell culture

VSM cells (VSMCs) isolated from the rat aortas were cultured in DMEM, which contained Nutrient Mixture F‐12 (DMEM/F12; Gibco Life Technology, Gaithersburg, MD, USA) supplemented with 10% fetal bovine serum (Hyclone, Logan, Utah, USA) and penicillin/streptomycin. Cells were maintained in a humidified 95% air‐5% CO₂ incubator at 37°C. Cells in passages three to six were used in all experiments. The K562 cell line, purchased from American Type Culture Collection (ATCC, Rockville, MD, USA), was used as positive control for antibodies against PKCβ2 and PKCθ.

Whole cell lysates

VSMCs or aorta tissues were washed twice with ice‐cold PBS and lysed in western and immunoprecipitation cell lysis buffer (Beyotime Institute of Biotechnology, Haimen, China) with Protease Inhibitor Cocktail (Calbiochem, Schwalbach, Germany) and Phosphatase Inhibitor Cocktail 3 (Sigma‐Aldrich, St Louis, MO, USA). They were incubated for 30 min on ice and then centrifuged at 12 000 × g for 15 min at 4°C. The supernatants were collected. The protein concentration in all samples was determined by the Bicinchoninic Acid (BCA) Protein Assay kit (Beyotime Institute of Biotechnology, Haimen, China) using BSA as the standard. All samples were mixed with 5× SDS sample buffer and placed in a boiling water bath for 5 min. All samples were stored at −20°C for Western blotting analysis.

Subcellular lysates

Aorta tissues were subfractionated into cytosolic and membrane fractions by adapting the previously described methods (Wang et al., 2010). Briefly, aorta tissues were washed twice with ice‐cold PBS, ground in lysis buffer A [1 mM NaHCO₃, 5 mM MgCl₂·6H₂O, 50 mM Tris–HCl, 10 mM EGTA, 2 mM EDTA, 500 μM 4‐(2‐aminoethyl)‐benzenesulfonyl fluoride, 150 nM aprotinin, 1 μM leupeptin and 1 μM E‐46 protease inhibitor] at 4°C, and then homogenized by passing through a 26‐gauge needle five times. It was then incubated for 30 min on ice and ultracentrifuged at 100 000 × g for 1 h at 4°C using an ultracentrifuge (Beckman Coulter, Fullerton, CA, USA). The supernatant provided the cytosolic fraction. The pellet was resuspended in buffer B (buffer A with 1% Triton X‐100), homogenized by passing through a 26‐gauge needle five times, incubated for 30 min on ice and ultracentrifuged at 100 000 × g for 1 h at 4°C. The supernatant provided the membrane fraction. Protein concentration was determined by the BCA Protein Assay kit (Beyotime Institute of Biotechnology, Haimen, China) using BSA as the standard. All of the samples were mixed with 5× loading buffer, placed in a boiling water bath for 5 min and stored at −20°C for Western blotting analysis.

Western blot

Samples (50 μg per lane) were loaded and separated on an SDS‐PAGE and then transferred to a polyvinylidine difluoride membrane. The membrane was blocked with 5% nonfat milk and incubated with the primary antibodies against PKCα (1:1000; Abcam, Cambridge, UK), PKCβ1 (1:500; R&D Systems, Minneapolis, MN, USA), PKCβ2 (1:1000; Cell Signaling Technology, Danvers, MA, USA), phospho‐PKCβ2 (1:1000; Abcam, Cambridge, UK), PKCδ (1:1000; Cell Signaling Technology, Danvers, MA, USA), PKCθ (1:1000; Abcam, Cambridge, UK), phospho‐PKCθ (1:1000; Cell Signaling Technology, Danvers, MA, USA), PKCε (1:1000; Abcam, Cambridge, UK) and PKCζ (1:1000; Cell Signaling Technology, Danvers, MA, USA) overnight at 4°C, which was incubated with HRP‐conjugated secondary antibodies and developed by chemiluminescent HRP Substrate (Thermo Fisher Scientific, Waltham, MA, USA). Images were acquired using ImageQuant LAS 4000 mini (GE Healthcare Life Sciences, Piscataway, NJ, USA). The density of each band was analysed with imagej software. The results were expressed as ratios to control.

PKCβ2 and PKCθ kinase assay

PKCβ2 and PKCθ enzymic activity was assayed using a modification of an immunoprecipitation method that was previously described (Sutcliffe et al., 2009). Briefly, whole cell extracts were prepared and sonicated for 5 min. The sonicated samples were centrifuged at 14 000 × g for 5 min to remove debris, and the supernatant was collected. Samples were diluted with kinase assay dilution buffer and incubated overnight at 4°C with 5 μg of anti‐PKCβ2 or anti‐PKCθ (1:1000; Cell Signaling Technology, Danvers, MA, USA) and a goat anti‐rabbit IgG conjugated to agarose beads affinity isolated antibody (Sigma‐Aldrich, St Louis, MO, USA). Samples were then processed with the Kinase Wash Buffer (Enzo Life Sciences, Lausen, Switzerland). The beads were then resuspended in Kinase Assay Dilution Buffer (Enzo Life Sciences, Lausen, Switzerland). The samples were loaded in duplicate wells on the PKCβ2 and PKCθ kinase activity plates, and the assay was performed as per the manufacturer's guidelines (PKC kinase activity kit; Enzo Life Sciences, Lausen, Switzerland). The PKCβ2 and PKCθ kinase activity was measured at an absorbance of 450 nm using a microplate reader (BioTek, Winooski, VT, USA). PKC kinase activity was analysed by first subtracting the blank readings from the average of duplicate sample wells to correct for background absorbance. Then, the no‐antibody control well readings were subtracted from the corrected sample readings to give the relative kinase activity.

Silencing of PKCβ2 and PKCθ by shRNA and lentivirus infection

In order for PKCβ2 and PKCθ in SHR VSMCs to be specifically down‐regulated, shRNAs targeting PKCβ2 and PKCθ were designed and synthesized by GeneChem Co Ltd (Shanghai, China) based on their cDNA sequences (Gen‐Bank accession NM_001172305; NM_001276721). The sequence of PKCβ2 shRNA were as follows: 5′‐GAGCGGAAGGGTACAGATGAA‐3′ for shRNA 1, 5′‐CTACAGAAAGCCGGAGTGGAT‐3′ for shRNA 2 and 5′‐GTGGAACGAAACCTTCAGATT‐3′ for shRNA 3. The sequence of PKCθ shRNA were as follows: 5′‐CTGGGAAGAGCTTGAGAGAAA‐3′ for shRNA 1, 5′‐GTGGTATAAACCAGAAGCTAA‐3′ for shRNA 2 and 5′‐CAGATTCAAAGTCTACAACTA‐3′ for shRNA 3. The sequence 5′‐TTCTCCGAACGTGTCACGT‐3′ was used as a negative control. These oligonucleotides were then subcloned into a lentiviral vector (hU6‐MCS‐CMV‐EGFP; GeneChem, Shanghai, China), and the lentiviruses were produced in 293T cells. After determining the multiplicity of infection (MOI) using a standard procedure, the viruses were used to infect SHR VSMCs. Cells were subcultured at a density of 2 × 10⁵ cells per well into six‐well cell culture plates and infected for 72 h with PKCβ2 or PKCθ lentiviral vectors or a negative control lentiviral vector at the MOI according to the preliminary experiment. Thereafter, PKCβ2 and PKCθ knockdown was confirmed by Western blotting. The shRNAs with the best silencing effect were used for the subsequent experiments.

F‐/G‐actin assay

Filamentous (F)‐actin and globular (G)‐actin in smooth muscle cells were isolated by fractionation and differential centrifugation and quantified by electrophoresis according to the manufacturer's description (F‐actin/G‐actin in vivo assay kit; Cytoskeleton, Denver, CO, USA) (Chen et al., 2006; Turczynska et al., 2015). Briefly, rat aortic VSMCs were collected in lysis buffer provided with the kit, LAS02 containing ATP and protease inhibitor cocktail. F‐actin was pelleted by centrifugation at high speed (100 000 × g) using a Beckman ultracentrifuge at 37°C for 1 h. G‐actin was transferred to fresh test tubes. F‐actin pellet was dissolved in F‐actin depolymerizing buffer and lysed on ice for 1 h (pipetting every 15 min). Supernatant (G‐actin) and pellet (F‐actin) fractions were diluted 50 times and analysed by immunoblotting using anti‐actin rabbit polyclonal antibody. Anti‐rabbit HRP‐conjugated secondary antibody (1:5000; Huaan, Hangzhou, China) was used. Bands were visualized using ECL (Thermo Fisher Scientific, Waltham, MA, USA), and images were acquired using ImageQuant LAS 4000 mini (GE Healthcare Life Sciences, Piscataway, NJ, USA).

F‐actin staining assay

After treatments, VSMCs were fixed with 4% paraformaldehyde in PBS (pH 7.4) and permeabilized with 0.5% Triton X‐100. The F‐actin was labelled with FITC–phalloidin (1:100; Cytoskeleton, Denver, CO, USA) for 30 min, and the cell nuclei were stained with DNA‐binding fluorescence dye DAPI (Beyotime Institute of Biotechnology, Haimen, China) at room temperature. After labelling, cells were washed to remove excess label and examined using a fluorescent microscope (Olympus, Tokoyo, Japan).

Data and statistical analysis

The data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2015). All aortic rings and cultured cells used in this research were randomized. Experimental procedures or treatments and data analyses were carried out with blinding. Data were normalized to control group or baseline value. GraphPad Prism 5 software was used for all statistical analyses. The values are presented as the means ± SD and either refers to the number of experiments performed on rings from different rat aortas or the number of independent experiments with cultured VSMCs. One or two rings prepared from the same aorta were used in similar experiments performed in parallel. Technical replicates were used to ensure the reliability of single values. Consecutive cumulative concentration–response curves were constructed. The relaxant response to propofol was calculated as the % change in the tension induced by noradrenaline. The significance of the difference between curves was analysed by two‐way ANOVA followed by Bonferroni post tests. The results shown in the blots are representative of five independent experiments. The densitometry data were analysed using t‐test or one‐way ANOVA followed by Bonferroni post tests. A value of P < 0.05 was considered statistically significant.

Materials

Materials used in these experiments were supplied as follows;‐propofol (AstraZeneca Pharmaceuticals, Cheshire, UK); PPS (Santa Cruz Biotechnology, Santa Cruz, CA, USA); calphostin C, noradrenaline (Sigma‐Aldrich, St Louis, MO, USA); Gö6976 (Calbiochem, Schwalbach, Germany); LY333531, phorbol 12,13‐dibutyrate, ryanodine (Tocris Bioscience, Bristol, UK).

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Southan et al., 2016), and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 (Alexander et al., 2015).

Results

Parameters of WKY and SHR

The physiological parameters of WKY and SHR are shown in Supporting Information Table S1. All the observed parameters except for BP [systolic BP (SBP), diastolic BP (DBP) and mean arterial pressure (MAP)] value were similar between two strains with no significant difference. The BP (SBP, DBP and MAP) was significantly higher in SHR than in WKY.

Increased expression and activity of PKCβ2 and PKCθ in aortic VSMCs in SHR

We first examined the expression levels of individual PKC isoforms in cultured VSMCs as well as VSMs isolated from SHR and WKY aortas. Western blot showed that the PKCβ2 and PKCθ expression levels were significantly higher in aortic VSMCs from SHR than in those from WKY (Figure 1A). By contrast, the protein levels of PKCα, PKCδ, PKCε and PKCζ were comparable between WKY and SHR (Figure 1A). PKCβ1 was not found to be expressed in either SHR or WKY aorta smooth muscle (data not shown). To confirm the specificity of antibodies against PKCβ2 and θ, we used K562 cell lysates as the positive controls. The immunoblotting results indicated that the antibodies used in this study only recognize their specified isoform (Figure 1B). The basal increases in PKCβ2 and PKCθ levels were accompanied by corresponding augmentation of each isoform activity (Figure 1C), as assessed by elisa peptide assay, and the phosphorylation of myristoylated alanine‐rich C kinase substrate (MARCKS)(S152/156), a downstream target of PKC in general (Figure 1D). The augmented expression and activation of PKCβ2 and PKCθ were also observed in isolated SHR aortic ring tissue (Supporting Information Figure S1).

Figure 1.

Comparison of expression and activity of PKCβ2 and PKCθ in SHR and WKY VSMCs. (A) Western blotting showing the protein levels of PKCα, PKCβ2, PKCδ, PKCθ, PKCε and PKCζ in WKY and SHR VSMCs. (B) Positive control of antibodies against PKCβ2 and PKCθ. Western blotting showing the protein levels of PKCβ2 (left panel) and PKCθ (right panel) in WKY and SHR VSMCs and K562 cells (positive controls of PKCβ2 and PKCθ antibodies). (C) Kinase activity of PKCβ2 and PKCθ in WKY and SHR VSMCs. (D) Phosphorylation level of MARCKS, a substrate of PKCs, in WKY and SHR VSMCs. Relative values are normalized to WKY group. Data are presented as mean ± SD. n = 5. ^* P < 0.05, significantly different from WKY; t‐test.

The role of PKCβ2 and PKCθ in propofol‐induced endothelium‐independent vasodilation in SHR

We previously demonstrated that propofol induced relaxation of endothelium‐denuded, aortic rings from rats with normal BP, precontracted with phorbol 12,13‐dibutyrate (PDBu), a potent PKC activator (Wang et al., 2015). We then compared the effects of the drugs on aorta from SHR with those from WKY. PDBu elicited significantly greater contraction in SHR group than in WKY group (Supporting Information Figure S2A). The contraction to PDBu was significantly inhibited by propofol in both groups. However, the inhibition of contraction in aorta from SHR was greater than that from WKY (Supporting Information Figure S2B). Given that increases in the amount and activity of PKCs could promote VSM proliferation and contraction pathways, leading to persistent increases in BP (Rosen et al., 1999; Fan et al., 2014), we then investigated whether the propofol‐induced greater relaxation in denuded aortic rings of SHR involves PKCβ2 and PKCθ. Denuded aortic rings from both SHR and WKY were constricted with cumulative concentrations of noradrenaline (1 × 10⁻⁹ to 5 × 10⁻⁶ M). Based on our pilot data, the EC₅₀ of noradrenaline were 10⁻⁸ and 5 × 10⁻⁸ M for SHR and WKY respectively. These concentrations of noradrenaline were used in the following experiments.

Denuded aortic rings precontracted with noradrenaline were incubated with propofol (1–1000 μM). As shown in Figure 2A, the denuded rings from SHR showed a greater relaxation to propofol than those from WKY. Although the relaxation in response to a 1.8% dilution of a 10% fat emulsion, which corresponds to the lipid content of the 1000 μM propofol emulsion, was not negligible (23 ± 3% and 15 ± 3% maximal relaxation in SHR and WKY groups, respectively), it was significantly less than that obtained with the respective propofol emulsion.

Figure 2.

Role of PKCβ2 and PKCθ in propofol (prop)‐induced vasodilation of SHR aorta. (A) Comparison of the dose‐dependent relaxation of propofol on noradrenaline (NE)‐precontracted aortic rings of SHR and WKY rats. ^* P < 0.05, significantly different from WKY NE + prop; two‐way ANOVA followed by Bonferroni post hoc test. Relaxation by propofol of WKY (B) and SHR (C) aortic rings, with or without LY333531 (LY). ^* P < 0.05; ^# P < 0.05; ^& P < 0.05, significant effect of LY333531 at 0.1, 0.2 or 0.5 μM respectively; two‐way ANOVA followed by Bonferroni post hoc test. Relaxation by propofol of WKY (D) and SHR (E) aortic rings with or without PPS. ^* P < 0.05; ^# P < 0.05; ^& P < 0.05, significant effect of PPS at 1, 3 or 10 μM respectively; two‐way ANOVA followed by Bonferroni post hoc test. (F) Comparison of the effect of LY and other PKC inhibitors [calphostin C (CC) and Gö6976 (Gö)] on propofol ‐induced vasorelaxation of SHR aorta. ^* P < 0.05; ^# P < 0.05; ^& P < 0.05, significant effect of CC (1 μM), Gö (3 μM), LY at 0.2 μM respectively; two‐way ANOVA followed by Bonferroni post hoc test. The relaxant response to propofol was calculated as the % change in the tension induced by noradrenaline (NE). Data are presented as mean ± SD. n = 9 per group.

To investigate whether PKCβ2 and PKCθ are involved in propofol‐induced vasorelaxation in aortic rings, precontracted with noradrenaline, we pretreated denuded aortic rings from SHR and WKY with either the selective PKCβ inhibitor LY333531 at 0.1, 0.2 or 0.5 μM or the specific PKCθ pseudo‐substrate inhibitor (PPS) at 1, 3 or 10 μM, for 30 min. As shown in Supporting Information Figure S3, both inhibitors significantly but partly inhibited noradrenaline‐induced contraction of SHR aortic rings (Supporting Information Figure S3C, D) but not WKY (Supporting Information Figure S3A, B) in a dose‐dependent manner, with a maximal inhibitory effect of LY333531 at 0.5 μM and PPS at 10 μM. In SHR aortic rings, when restoring the amplitude of noradrenaline‐induced contraction to the control level by increasing the noradrenaline concentration, propofol‐induced relaxation was markedly reduced compared with that in the absence of each of the PKC inhibitors (Figure 2C, E), suggesting a significant role of PKCβ2 and PKCθ in the greater relaxation–induced by propofol. Pretreatment with LY333531 or PPS did not change the relaxation induced by propofol in WKY aorta (Figure 2B, D). elisa assay (Supporting Information Figure S4A–D) also indicated that propofol treatment significantly inhibited the activity of PKCβ2 and PKCθ in VSMCs of SHR but not WKY, which was consistent with the results obtained in aortic rings.

Notably, the maximal inhibition of noradrenaline contraction obtained in treatment with 0.5 μM of the selective PKCβ2 inhibitor LY333531 was more profound than that of PPS (at 10 μM; data not shown). The maximal inhibition of noradrenaline contraction refers to the maximal inhibitory effects of each PKC inhibitor on noradrenaline‐induced contraction. Consistently, the magnitude of propofol‐induced relaxation was decreased much more in the presence of LY333531 than in the presence of PPS (data not shown), suggesting that inhibition of PKCβ2 plays a dominant role in propofol‐induced exaggerated vasodilation in SHR aorta.

We have previously demonstrated that down‐regulation of nPKC isoforms (−δ, −θ and −ε) by prolonged incubation with another phorbol ester, PMA (Wang et al., 2015), mediates propofol‐induced vasodilation in normal rat aorta. Given that propofol‐induced relaxation was through inhibiting several PKC isoforms in rat aorta, we then investigated whether inhibition of individual PKC isoforms mediates different and unique effects on propofol relaxation in SHR aorta. Two other selective PKC inhibitors calphostin C and Gö6976 were used to compare with LY333531. Calphostin C is a natural chemical compound with a highly potent inhibitory effect on both classical PKCs and novel PKCs, while Gö6976 selectively inhibits the kinase domain of conventional rather than novel isoform PKCs. As shown in Figure 2F, Calphostin C and Gö6976 had similar inhibitory effects on noradrenaline‐induced contraction and subsequently propofol‐induced relaxation as LY333531, albeit with selective inhibition of cPKCs/nPKCs, suggesting that the Calphostin C, Gö6976 and LY333531 compounds suppress the noradrenaline contraction, as well as propofol relaxation, all by specifically inhibiting PKCβ2 downstream pathways in SHR aorta smooth muscle.

Effects of inhibition of PKCβ2 and PKCθ in absence of intracellular or extracellular Ca²⁺ on propofol‐induced relaxation

Aortic rings depleted of intracellular Ca²⁺ stores by ryanodine treatment were contracted with noradrenaline (5 × 10⁻⁸ M for WKY and 1 × 10⁻⁸ M for SHR) in the presence or absence of LY333531 (0.1, 0.2 and 0.5 μM) or PPS (1, 3 and 10 μM). Although depletion of intracellular Ca²⁺ markedly inhibited the amplitude of noradrenaline contraction in both SHR and WKY aortas (data not shown), pretreatment with LY333531 (0.2 or 0.5 μM) or PPS (3 or 10 μM) still significantly inhibited noradrenaline contraction in SHR aortas (Supporting Information Figure S5C, D) but not WKY (Supporting Information Figure S5A, B). Additionally, propofol‐induced relaxation was markedly inhibited by pretreatment with either of the specific PKC inhibitors in SHR rings (Figure 3C, D) but not in WKY rings (Figure 3A, B). Similar to depletion of intracellular Ca²⁺, removal of extracellular Ca²⁺ strongly inhibited the noradrenaline‐induced contraction as well as propofol relaxation in both SHR and WKY aortic rings (data not shown). Besides, both LY333531 and PPS still markedly reduced NE contraction as well as propofol relaxation in SHR aorta (Supporting Information Figure S5G, H and Figure 3G, H) but not in WKY aorta (Supporting Information Figure S5E, F and Figure 3E, F) in Ca²⁺‐free external solution. Together, these results suggest that sensitivity to inhibition of PKCβ2 and PKCθ for noradrenalin‐induced contraction and propofol‐induced relaxation were calcium‐independent events in SHR aorta.

Figure 3.

Comparison of propofol‐induced vasodilation in the presence or absence of LY333531 (LY) or PPS in WKY and SHR aorta after depletion of sarcoplasmic reticulum (SR) calcium or in calcium‐free external solution. Relaxation by propofol (prop) of WKY aorta, with or without LY (A) and with or without PPS (B), after depletion of SR calcium. ^* P < 0.05; ^# P < 0.05; ^& P < 0.05, with or without LY at 0.1, 0.2 or 0.5 μM respectively; two‐way ANOVA followed by Bonferroni post hoc test. Relaxation by propofol of SHR aorta, with or without LY (C) and PPS (D) after depletion of SR calcium ^* P < 0.05; ^# P < 0.05; ^& P < 0.05, significant effect of PPS at 1, 3 or 10 μM respectively; two‐way ANOVA followed by Bonferroni post hoc test. Relaxation by propofol of WKY aorta, with or without LY (E) and PPS (F) in calcium‐free external solution. ^* P < 0.05; ^# P < 0.05; ^& P < 0.05, significant effect of LY at 0.1, 0.2 or 0.5 μM respectively; two‐way ANOVA followed by Bonferroni post hoc test. Relaxation by propofol of SHR aorta, with or without LY (G) and PPS (H) in calcium‐free external solution. ^* P < 0.05; ^# P < 0.05; ^& P < 0.05, significant effect of PPS at 1, 3 or 10 μM respectively; two‐way ANOVA followed by Bonferroni post hoc test. The relaxant response to propofol was calculated as the % change in the tension induced by noradrenaline (NE). Data are presented as mean ± SD. n = 9 per group.

The effect of propofol on noradrenaline‐induced phosphorylation and translocation of PKCβ2 and PKCθ in SHR aortas

PKC translocation and phosphorylation have been shown to facilitate the tight binding of PKC to its target substrate in vitro and in vivo (Orr et al., 1992; Newton, 1995; Khalil, 2013), which induces phosphorylation of certain substrates, leading to activation of a cascade of protein kinases that enhance VSM contraction. To investigate the molecular mechanism responsible for propofol‐induced relaxation in SHR aortic smooth muscle, we then examined whether or not propofol affects phosphorylation and/or translocation of PKCβ2 and PKCθ by noradrenaline. SHR aortic strips were treated with noradrenaline (0.01 μM) for 15 min in the absence or presence of propofol (10 μM). As shown in Figure 4A, strips treated with noradrenaline alone showed an increased phosphorylation of both PKCβ2(S660) and PKCθ(T538) in SHR aortas. However, addition of propofol attenuated these phosphorylation events. Similar inhibitory effects were also observed for propofol in VSMCs isolated from SHR aortas (Supporting Information Figure S6). In addition, noradrenaline treatment induced the translocation of PKCβ2 and PKCθ from the cytosol to the surface membrane in SHR aorta, which was also attenuated in the presence of propofol (Figure 4B). Kinase activity assays shown in Figure 4E, F demonstrated that treatment with noradrenaline increased phosphorylation of MARCKS(S152/156), a specific PKC substrate, which was attenuated in cells co‐treated with propofol. Interestingly, propofol, LY333531 and PPS had similar effects in inhibiting the phosphorylation of MARCKS induced by noradrenaline (Figure 4C, D). To further confirm whether PKCβ2 and PKCθ were involved in propofol‐induced relaxation in SHR aortas, we used PKCβ2 and PKCθ shRNA to specifically down‐regulate PKCβ2 and PKCθ in VSMCs of SHR aortas. Supporting Information Figure S7 showed that down‐regulation of PKCβ2 or PKCθ with their respective shRNA significantly down‐regulated PKCβ2 or PKCθ in VSMCs from SHR. Silencing of PKCβ2 (Figure 4E) and PKCθ (Figure 4F) significantly inhibited the phosphorylation of MARCKS, which was markedly increased by noradrenaline in SHR VSMCs. However, neither propofol nor noradrenaline had any significant effect on the activity of both PKCβ2 (Supporting Information Figure S8A) and PKCθ (Supporting Information Figure S8B) in WKY aortas. Additionally, noradrenaline‐induced phosphorylation of MARCKS could only be inhibited by propofol but not by LY333531 or PPS in WKY VSMCs (Supporting Information Figure S8C).

Figure 4.

Effects of propofol (prop) on PKCβ2 and PKCθ activation in endothelium‐denuded aortic tissue and cells in SHR. (A) Effects of propofol on PKCβ2 and PKCθ phosphorylation level in SHR aorta. ^* P < 0.05, significantly different from control (con), ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. (B) Effects of propofol on PKCβ2 and PKCθ translocation in SHR aorta. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. (C) Effects of propofol and LY333531 (LY) on MARCKS phosphorylation level in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. (D) Effects of propofol and PPS on MARCKS phosphorylation level in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. (E) Effects of propofol or PKCβ2 silencing by shRNA on MARCKS phosphorylation level in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. (F) Effects of propofol and PKCθ silencing by shRNA on MARCKS phosphorylation level in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to control or vector group. Data are presented as mean ± SD. n = 5.

Phosphorylation of CPI‐17, MYPT and MLC in VSMCs from SHR aorta and the effect of noradrenaline, propofol and PKCβ2/θ inhibition

To investigate the contractile signalling pathways involved in PKCβ2‐ and PKCθ‐mediated contraction to noradrenaline, as well as propofol relaxation, we compared the phosphorylation levels of CPI‐17(T38), MYPT1(T853) and MLC(S20) in SHR aortas with those in WKY. As shown in Figure 5A, basal phosphorylation levels of CPI‐17(T38), MYPT1(T853) and MLC(S20) in SHR aortic VSMCs were significantly higher than those in WKY. Interestingly, the expression levels of total CPI‐17 in SHR were also higher, whereas total MLC and the regulatory subunit of myosin phosphatase (MYPT1) were not different between the groups. Noradrenaline (1 μM) treatment further enhanced the phosphorylation of CPI‐17(T38), MYPT1(T853) and MLC(S20) in cells from SHR, which was abolished by propofol (30 μM) or LY333531 (0.1 μM) (Figure 5B) but not PPS (10 μM) (Figure 6A). Notably, noradrenaline also slightly but significantly increased phosphorylation of CPI‐17(T38), MYPT1(T853) and MLC(S20) in cells from WKY. However, this phosphorylation event was only attenuated by propofol (Supporting Information Figure S9A, B). To define further the involvement of PKCβ2 and PKCθ in propofol‐induced vasodilation, we next used PKCβ2 shRNA and PKCθ shRNA to selectively down‐regulate the expression of PKCβ2 and PKCθ in SHR VSMCs. As shown in Figures 5C and 6B, similar to propofol, preincubation of cells with the indicated concentration of shRNA to PKCβ2 (Figure 5C) but not PKCθ (Figure 6B) resulted in inhibition of noradrenaline‐elicited phosphorylation of CPI‐17(T38), MYPT1(T853) and MLC(S20).

Figure 5.

Phosphorylation of CPI‐17, MYPT1 and MLC and the effect of noradrenaline (NE; 1 μM), propofol (prop; 30 μM), LY333531 (LY; 0.1 μM) or PKCβ2 silencing in SHR VSMCs. (A) Comparison of phosphorylation level of CPI‐17, MYPT1 and MLC in SHR and WKY VSMCs. ^* P < 0.05, significantly different from WKY, t‐test. Relative densities are normalized to WKY group. (B) Phosphorylation level of CPI‐17, MYPT1 and MLC in the presence of noradrenaline, propofol and LY in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to the control group. (C) Phosphorylation level of CPI‐17, MYPT1 and MLC after PKCβ2 silencing in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to vector group. Data are presented as mean ± SD. n = 5.

Figure 6.

Effect of noradrenaline (NE; 1 μM), propofol (prop; 30 μM), PPS (10 μM) and PKCθ silencing on phosphorylation of CPI‐17, MYPT1 and MLC in SHR VSMCs. (A) Phosphorylation level of CPI‐17, MYPT1 and MLC in the presence of noradrenaline, propofol and PPS in SHR VSMCs [^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to con group. (B) Phosphorylation level of CPI‐17, MYPT1 and MLC after PKCθ silencing in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to vector group. Data are presented as mean ± SD. n = 5.

The change in actin polymerization in VSMCs from SHR aorta and the effect of noradrenaline, propofol and PKCβ2/θ inhibition

Cipolla et al. (2002) have demonstrated the involvement of actin polymerization in myogenic contractions and, conversely, that depolymerization of F‐actin with cytochalasin D causes dVSM relaxation. Prolonged vasoconstriction of arteries, which causes hypertension, is involved in VSM actin polymerization (Staiculescu et al., 2013). To explain the finding that inhibition of PKCθ did not change noradrenaline‐induced increases in the phosphorylation of CPI‐17(T38), MYPT1(T853) and MLC(S20) in SHR aortic smooth muscle, although it still inhibited noradrenaline‐induced contraction as well as propofol‐induced relaxation (Figure 2E and 3D, H and Supporting Information S3B and S5D, H), we performed the polymerization of F‐actin assay. F‐actin to G‐actin ratio quantification indicated that the basal F‐actin/G‐actin ratio in VSMCs from SHR was significantly higher than that from WKY (data not shown). As shown in Figure 7A, B, treatment with noradrenaline (1 μM) increased the ratio of F‐actin to G‐actin, which was attenuated by either LY333531 (0.1 μM) (Figure 7A) or PPS (10 μM) (Figure 7B) to similar extent in VSMCs from SHR but not WKY. To further confirm the effects of inhibition of PKCβ2 and PKCθ on noradrenaline ‐induced F‐actin polymerization, we performed F‐actin staining assay. As shown in Figure 7C, D, noradrenaline increased the cytoskeleton reorganization in VSMCs from both SHR and WKY aortas, which was also abolished by treatment with LY333531 (Figure 7C) and PPS (Figure 7D) in SHR VSMCs but not WKY. Such noradrenaline‐stimulated polymerization of F‐actin was also inhibited by propofol. Similar to LY333531 and PPS, down‐regulation of both PKCβ2 and PKCθ by each shRNA decreased the F‐actin/G‐actin ratio induced by noradrenaline (Figure 7E, F).

Figure 7.

F‐actin reorganization and the effect of noradrenaline (NE; 1 μM), propofol (prop; 30 μM) and inhibition of PKCβ2 and PKCθ in WKY and SHR VSMCs. (A) F‐actin/G‐actin ratio in the presence of propofol or LY333531 (LY; 0.1 μM) in WKY and SHR VSMCs. ^* P < 0.05, WKY significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); ^& P < 0.05, significantly different from control (con); ^$ P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to WKY in lipid + DMSO group. (B) F‐actin/G‐actin ratio in the presence of propofol and PPS (10 μM) in WKY and SHR VSMCs. ^* P < 0.05, significantly different from corresponding control (con); ^# P < 0.05, significantly different from WKY+noradrenaline (NE); ^& P < 0.05, significantly different from corresponding control (con); ^$ P < 0.05, significantly different from SHR+noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to WKY in lipid group. (C) F‐actin staining in the presence of propofol, LY and PPS in WKY VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to con group. (D) F‐actin staining in the presence of propofol, LY and PPS in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Scale bar: 50 μm. Relative densities are normalized to con group. (E) F‐actin/G‐actin ratio after PKCβ2 silencing in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc comparisons. Relative densities are normalized to vector group. (F) F‐actin/G‐actin ratio after PKCθ silencing in SHR VSMCs. ^* P < 0.05, significantly different from control (con); ^# P < 0.05, significantly different from noradrenaline (NE); one‐way ANOVA followed by Bonferroni post hoc test. Relative densities are normalized to vector group. Data are presented as mean ± SD. n = 5. W, WKY; S, SHR.

PKCβ2 or PKCθ inhibition did not affect the activity of MLC kinase (MLCK)

We additionally measured MLCK activity using an elisa kit. The results indicated that noradrenaline significantly increased MLCK activity in both SHR (Supporting Information Figure S10B) and WKY (Supporting Information Figure S10A) VSMCs. Pretreatment with propofol significantly decreased MLCK activity in both VSMCs. However, neither LY333531 nor PPS inhibited MLCK activity triggered by noradrenaline in SHR and WKY VSMCs. This result excluded the possibility that the effects of LY333531 and PPS were due to inhibition of MLCK.

Discussion

The results from clinical studies and hypertensive animal models clearly demonstrate that the use of propofol for anaesthesia induction is often accompanied with profound hypotension and haemodynamic instability, and its safety in patients with hypertension has been debated for a long time. Although in blood vessel models, propofol has been demonstrated to influence cellular processes including calcium signalling (Lawton et al., 2012; Han et al., 2016), sympathetic neurotransmission (Ebert, 2005; Han et al., 2016) and the function of endothelium (Wang et al., 2010; Gragasin et al., 2013), propofol can also inhibit PKCs in VSMCs (Tanabe et al., 1998; Yu et al., 2006; Kuriyama et al., 2012). Although PKCs have emerged as new targets for treating genetic hypertension in various hypertensive models, the principal finding of the present study is that the enhanced expression and activation of two PKC isoforms, PKCβ2 and PKCθ, in SHR aortic smooth muscle are at least partly involved in the profound relaxation induced by propofol.

We used immunoblotting to demonstrate the increased expression and phosphorylation of PKCβ2 and PKCθ in VSM from SHR aortas. Such an enhanced expression and phosphorylation coincided with activation of both kinases, which is also revealed by increased activity of PKCβ2 and PKCθ and increased phosphorylation of MARCKS, a direct target of PKCs. The activation of PKCβ2 and PKCθ is involved in propofol‐induced more profound relaxation in SHR aortic smooth muscle, because inhibition of either PKCβ2 or PKCθ can also decrease the magnitude of propofol vasodilation (Figure 2C, E). Previous studies by Yu et al. (2006) and our group (Wang et al., 2015) have shown that propofol inhibits vasoconstriction in part by suppressing PKC activation in normal rats. The present data indicate that basal increased PKCβ2 and PKCθ phosphorylation and activity in SHR aortic VSM was further enhanced by noradrenaline and subsequently inhibited by propofol rapidly. This swift inhibitory effect is consistent with our finding that propofol induced a greater inhibition of noradrenaline‐induced contraction in SHR aortas.

Propofol‐induced vascular relaxation has been found to be mediated by both endothelium‐dependent and ‐independent mechanisms (Kassam et al., 2011; Gragasin et al., 2013; Wang et al., 2015). The present results indicate that the endothelium was not engaged in the excessive relaxation by propofol, shown in our model. The experimental evidence for this is provided by the similar magnitude of propofol‐induced relaxation in aortic rings with or without endothelium from SHR (Supporting Information Figure S11A). The association between endothelium dysfunction and impaired NO availability is clear, because l‐NAME could not inhibit propofol‐induced vasodilation in the endothelium‐intact aorta from SHR (Supporting Information Figure S11B). This result agrees with that by Roque et al. showing that ACh‐induced relaxation in mesenteric artery rings from SHR, but not WKY, was significantly decreased (Roque et al., 2013). Although endothelium‐derived relaxing factors include NO, endothelium‐dependent hyperpolarization and prostacyclin, NO plays a dominant role in conduit arteries (Leung and Vanhoutte, 2017).

PKCs are involved in VSM contraction by vascular agonists, by increasing the phosphorylation of Ca²⁺‐sensitization pathways such as CPI‐17–MYPT1–MLC (Woodsome et al., 2001; Moreno‐Dominguez et al., 2013). Consistent with these data, inour model, noradrenaline treatment elicited the phosphorylation of CPI‐17–MYPT1–MLC in aortic smooth muscle from both SHR and WKY, which was inhibited by the addition of propofol. The phosphorylation of CPI‐17–MYPT1–MLC in SHR aorta was higher than that in WKY,under basal conditions. which may partly account for the discrepant vascular response to compounds in SHR aortas and in WKY aortas. The different contributions of individual PKC subtypes to the regulation of myogenic contraction are not well defined. Gu and Bishop (1994) reported that the increased PKC activity, induced by pressure‐overload, was mostly due to increases in the amount ofPKCs β1‐, β2‐ and ε in the surface membrane and nuclear–cytoskeletal fractions in cardiac muscle in rats. Other studies showed that PKCα was activated and localized at the surface membrane in VSMCs of hypertensive rats (Khalil et al., 1992; Liou and Morgan, 1994). Although the present results showed that PKCβ2 plays a leading role in aortic smooth muscle cells of SHR, the dominant role of PKCβ2 could be due to its action in both contractile Ca²⁺‐sensitization pathways as well as actin polymerization in myogenic contraction. The experimental evidence supporting such an argument are as follows: (i) similar to the effect of propofol, the inhibition of PKCβ2 but not PKCθ abolished noradrenaline‐induced phosphorylation of CPI‐17–MYPT1–MLC in VSMCs of SHR; (ii) the inhibition of PKCβ2 suppressed noradrenaline‐induced contraction to a greater extent than PKCθ inhibition; and (iii)calphostin C, a selective cPKC/nPKC inhibitor, and Gö6976, a selective cPKC inhibitor, had the same inhibitory effect on contraction to noradrenaline, as well as the subsequent propofol‐induced relaxation as LY333531, suggesting that the three compounds suppress the response to noradrenaline as well as propofol, all by specifically inhibiting PKCβ2 downstream pathways in SHR aorta smooth muscle.

The inhibitory effect of propofol on noradrenaline‐, angiotensin‐ or endothelin‐1‐induced contraction in arterial smooth muscle from normotensive and hypertensive rats is dependent on vasoactive agent‐induced [Ca²⁺]_i release and Ca²⁺ influx (Tanabe et al., 1998; Samain et al., 2000; Samain et al., 2004; Han et al., 2016). The present results indicate that Ca²⁺ mobilization was not involved in the inhibitory actions of LY333531 and PPS on noradrenaline‐induced contraction as well as propofol‐induced relaxation in aortas from SHR. Removal of intracellular or extracellular Ca²⁺ was unable to abolish the inhibitory effects of both LY333531 and PPS on noradrenaline contraction and propofol relaxation (Figure 3 and Supporting Information Figure S5), although this removal could inhibit the change in magnitude of aorta tension induced by noradrenaline or propofol. This result agrees with those by Snow et al. (2011) and Dimopoulos et al. (2007) showing that PKC inhibitors had no effect on Ca²⁺ mobilization in VSMCs.

Increased expression and activity of PKCs has been shown to enhance VSM contraction and may be involved in hypertension (Sasajima et al., 1997; Novokhatska et al., 2013; Liu et al., 2015; Zhao et al., 2015). Data of the present study indicate that the increased expression of PKCβ2 in SHR VSM is involved in the exaggerated haemodynamic response to the vascular action of propofol. This result agrees with that by Jackson et al. showing that PKCβ inhibition significantly decreased the contraction caused by high glucose in human internal mammary arteries (Jackson et al., 2016). Additionally, targeting the increased expression of PKCβ in diabetic patients can improve insulin‐mediated eNOS activation in vascular tissues. In fact, PKCβ has emerged as a promising target for anti‐diabetes therapy partly through endothelial protection. Given that endothelial dysfunction is a common characteristic of various vascular diseases, PKCβ inhibition is also likely to be beneficial for patients with hypertension. Although there is still little evidence supporting the role of up‐regulation of PKCθ in human hypertension as well as in hypertensive animal models, our results indicate that the increased expression of PKCθ is also involved in SHR VSM relaxation in response to propofol. PKCθ has been demonstrated to be involved in TNF‐α‐induced inflammatory reaction in VSMCs (Phalitakul et al., 2011) and inflammation plays a very important role in early stage of hypertension development (Virdis et al., 2014). Thus, during hypertensive vascular disease, targeting PKCθ may be beneficial not only through mitigating hypertension but also through an anti‐inflammatory effect.

There are some limitations in the present study. The first is the use of aorta instead of resistance arteries, as the latter are directly involved in BP regulation, whereas the former is not. However, the unwanted hypotensive effect of propofol in patients with hypertension presents a severe haemodynamic instability, which is closely related to conduit vessels. In addition, it is already known that PKC inhibitors markedly (80–90%) inhibited agonist‐induced contraction in small mesenteric arteries (Kitazawa and Kitazawa, 2012). Thus, we used aorta to ensure that the magnitude of contraction induced by noradrenaline in the presence of PKC inhibitors was sufficient to evaluate the role of PKCs in mediating propofol‐induced relaxation. Second, in the propofol‐induced vasodilation assay, the use of pharmacological inhibitors of PKCs may carry risk of non‐specific effects. Although either LY333531 or PPS is used as the specific inhibitor of PKCβ and PKCθ (Gayen et al., 2009; Nagareddy et al., 2009), they still lack the specificity of gene knockdown. Despite these limitations, the findings in the present study could provide useful insights into the mechanisms underlying the propofol‐induced, severe hypotension in patients with hypertension. Such insights would be be helpful for the development of new therapeutic agents controlling the haemodynamic instability of postoperative hypertension.

On the basis of these findings, we have proposed a possible mechanism for the excessive vasorelaxation of SHR aorta, induced by propofol (Figure 8). In summary, we used vessel functional measures and cell‐based assays to investigate the role of increased expression and activity of PKCβ2 and PKCθ in propofol‐induced vasorelaxation in SHR aortas. Inhibition of either PKCβ2 or PKCθ reduced the magnitude of propofol‐induced relaxation in SHR aortic rings but not in those from WKY and the suppression was more profound with a PKCβ inhibitor. In addition, inhibition of PKCβ2, but not PKCθ, decreased the noradrenaline‐induced phosphorylation of CPI‐17, MLC and MYPT1 in VSMCs from SHR aortas. Inhibition of either PKCβ2 or PKCθ has similar inhibitory effects on noradrenaline‐elicited cytoskeleton reorganization in SHR VSMCs. Further studies with arteries of various sizes are required to confirm the role of PKCβ2 and PKCθ in propofol‐induced vasodilation in hypertension models.

Figure 8.

The proposed mechanism of propofol‐induced vasodilation in SHR aorta. Dual regulatory pathways by which propofol induces vasodilation by inhibiting the increased PKCβ2 activity in SHR VSM via targeting its downstream calcium‐sensitivity signalling and suppressing the cytoskeleton actin polymerization. (Right) Regulatory model by which propofol induces vasodilation by inhibiting the increased PKCθ activity in SHR VSM by targeting the cytoskeleton actin polymerization.

Author contributions

L.W, A.W. and W.J. contributed to the conception of the project; Y.W. and L.W. designed the study; Y.W., Q.Z. and B.W. performed the experiments; Y.W., Q.Z., H.Z. and X.Z. initially analysed the data and drafted the Methods section; Y.W., Q.Z. and L.W. made the final analysis of data and wrote the manuscript. All authors reviewed this manuscript.

Conflict of interest

The authors declare no conflicts of interest.

Declaration of transparency and scientific rigour

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research recommended by funding agencies, publishers and other organisations engaged with supporting research.

Supporting information

Table S1 Physiological parameters of WKY and SHR. Blood pressure in SHR was significantly higher than WKY (*P < 0.05 vs. WKY, t‐test). Data are presented as mean ± SD. WKY: Wistar‐Kyoto rat; SHR: spontaneously hypertensive rat; BW: body weight; SBP: systolic blood pressure; DBP: diastolic blood pressure; MAP: mean arterial pressure; LDL: low‐density lipoprotein; HDL: high‐density lipoprotein; TG: triglyceride; TC: total cholesterol. Results are expressed as mean ± SD.

Figure S1 Comparison of expression and activity of PKCβ2 and PKCθ in WKY and SHR aortic tissue. (A) Western blotting showing the protein levels of PKCα, PKCβ2, PKCδ, PKCθ, PKCε, PKCζ in WKY and SHR aortic tissue. (B) Kinase activity of PKCβ2 and in WKY and SHR aortic tissue. (C) Kinase activity of PKCθ in WKY and SHR aortic tissue. *P < 0.05 vs. WKY, t‐test. Relative values are normalized to WKY group. Data are presented as mean ± SD. n = 5. PKC, Protein kinase C; WKY, Wistar‐Kyoto rat; SHR, spontaneously hypertensive rats.

Figure S2 Effects of propofol on PDBu‐induced contraction. (A) PDBu (10‐9 ‐ 5 × 10‐6 M)‐induced dose‐dependent contraction in WKY and SHR aorta. (B) Propofol (1‐1000 μM) dilated the aortic rings precontracted by PDBu (0.05 and 0.01 μM for WKY and SHR, respectively) in WKY and SHR in a dose‐dependent manner. *P < 0.05 vs. WKY, two‐way ANOVA followed by Bonferroni post hoc comparisons. Contractile response of each segment is expressed as a percentage of the contraction to KCl (60 mM). The relaxant response to propofol was calculated as the percent change in the tension induced by PDBu. Data are presented as mean ± SD. n = 9 per group. PDBu, Phorbol 12,13‐dibutyrate; prop, propofol.

Figure S3 Effects of LY (0.1, 0.2 and 0.5 μM) and PPS (1, 3 and 10 μM) on NE (0.05 and 0.01 μM for WKY and SHR, respectively)‐induced contraction. Effects of LY (A) and PPS (B) on NE‐induced contraction in WKY aorta. (C) Effects of LY on NE‐induced contraction in SHR aorta. (*P < 0.05; #P < 0.05, in the absence vs. presence of LY at 0.2 or 0.5 μM, respectively. Two‐way ANOVA followed by Bonferroni post hoc comparisons). (D) Effects of PPS on NE‐induced contraction in SHR aorta. (*P < 0.05; #P < 0.05, in the absence vs. presence of PPS at 3 or 10 μM, respectively. Two‐way ANOVA followed by Bonferroni post hoc comparisons). Contractile response of each segment is expressed as a percentage of contraction by KCl (60 mM). Data are presented as mean ± SD. n = 9 per group. LY, LY333531; PPS, PKCθ pseudo‐substrate inhibitor; NE, noradrenaline.

Figure S4 Effects of propofol (30 μM) on PKCβ2 and PKCθ activity in WKY and SHR VSMCs. Effects of propofol on PKCβ2 in WKY (A) and SHR (B) VSMCs (*P < 0.05, NE vs. con; #P < 0.05, prop + NE vs. NE, one‐way ANOVA followed by Bonferroni post hoc comparisons). Effects of propofol on PKCθ in WKY (C) and SHR (D) VSMCs (*P < 0.05, NE vs. con; #P < 0.05, prop + NE vs. NE, one‐way ANOVA followed by Bonferroni post hoc comparisons). Kinase activity values are normalized to control group. Data are presented as mean ± SD. n = 5. con, control; prop, propofol.

Figure S5 Effects of LY (0.1, 0.2 and 0.5 μM) and PPS (1, 3 and 10 μM) on NE (0.05 and 0.01 μM for WKY and SHR, respectively)‐induced contraction after depletion of SR calcium or in calcium free external solution. NE‐induced contraction in the presence or absence of LY (A) or PPS (B) in WKY aorta after depletion of SR calcium. (C) NE‐induced contraction in the presence or absence of LY in SHR aorta after depletion of SR calcium (*P < 0.05; #P < 0.05, in the absence vs. presence of LY at 0.2 or 0.5 μM, respectively. Two‐way ANOVA followed by Bonferroni post hoc comparisons). (D) NE‐induced contraction in the presence or absence of PPS in SHR aorta after depletion of SR calcium (*P < 0.05; #P < 0.05, in the absence vs. presence of PPS at 3 or 10 μM, respectively. Two‐way ANOVA followed by Bonferroni post hoc comparisons). NE‐induced contraction in the presence or absence of LY (E) or PPS (F) in WKY aorta in calcium free external solution. (G) NE‐induced contraction in the presence or absence of LY in SHR aorta in calcium free external solution (*P < 0.05; #P < 0.05, in the absence vs. presence of LY at 0.2 or 0.5 μM), respectively. Two‐way ANOVA followed by Bonferroni post hoc comparisons). (H) NE‐induced contraction in the presence or absence of PPS in SHR aorta in calcium free external solution (*P < 0.05; #P < 0.05, PPS at 3 or 10 μM), respectively. Two‐way ANOVA followed by Bonferroni post hoc comparisons). Contractile response of each segment is expressed as a percentage of contraction by KCl (60 mM). Data are presented as mean ± SD. n = 9 per group.

Figure S6 Effects of propofol on PKCβ2 and PKCθ phosphorylation in SHR VSMCs. *P < 0.05, NE vs. con; #P < 0.05, prop + NE vs. NE, one‐way ANOVA followed by Bonferroni post hoc comparisons. Relative densities are normalized to control group. Data are presented as mean ± SD. n = 5.

Figure S7 Representative Western blots of PKCβ2 and PKCθ knockdown by shRNAs in SHR VSMCs. *P < 0.05 vs. con, one‐way ANOVA followed by Bonferroni post hoc comparisons. Relative densities are normalized to control group. Data are presented as mean ± SD. n = 5.

Figure S8 Effects of propofol (10 μM) on PKCβ2 and PKCθ activation in WKY aorta. (A) Effects of propofol on PKCβ2 and PKCθ phosphorylation level in WKY aorta. (B) Effects of propofol on PKCβ2 and PKCθ translocation in WKY aorta. (C) Effects of NE (1 μM), propofol and LY (0.01 μM) on MARCKS phosphorylation level in WKY VSMCs. (D) Effects of NE (1 μM), propofol and PPS (10 μM) on MARCKS phosphorylation level in WKY VSMCs. *P < 0.05, NE vs. con; #P < 0.05, prop + NE vs. NE, one‐way ANOVA followed by Bonferroni post hoc comparisons. Relative densities are normalized to control group. Data are presented as mean ± SD. n = 5. MARCKS, myristoylated alanine‐rich C kinase substrate.

Figure S9 Effect of NE (1 μM), propofol (30 μM), LY (0.1 μM) and PPS (10 μM) on phosphorylation of CPI‐17, MYPT1 and MLC in WKY VSMCs. (A) Phosphorylation level of CPI‐17, MYPT1 and MLC in the presence of NE, propofol and LY in WKY VSMCs. (B) Phosphorylation level of CPI‐17, MYPT1 and MLC in the presence of NE, propofol and PPS in WKY VSMCs. *P < 0.05, NE vs. con; #P < 0.05, prop + NE vs. NE, one‐way ANOVA followed by Bonferroni post hoc comparisons. Relative densities are normalized to control group. Data are presented as mean ± SD. n = 5. CPI‐17, protein kinase C‐potentiated protein phosphatase inhibitor protein 17 kDa; MYPT1, myosin light chain phosphatase; MLC, myosin light chain.

Figure S10 Effects of LY (0.1 μM) and PPS (10 μM) on MLCK kinase activity in WKY and SHR VSMCs. Effects of LY and PPS on MLCK kinase activity in WKY (A) and SHR (B) VSMCs. *P < 0.05 vs. con, one‐way ANOVA followed by Bonferroni post hoc comparisons. Kinase activity values are normalized to control group. Data are presented as mean ± SD. n = 5. MLCK, myosin light chain kinase.

Figure S11 Role of endothelium in propofol‐induced vasodilation of SHR and WKY aorta. (A) Comparison of the dose‐dependent relaxation of propofol in endothelium‐denuded and endothelium‐intact aortic rings in SHR and WKY rats. (*P < 0.05, WKY EC(+) vs. WKY EC(‐), two‐way ANOVA followed by Bonferroni post hoc comparisons). (B) Comparison of the dose‐dependent relaxation of propofol in endothelium‐intact aortic rings in the presence or absence of L‐NAME (100 μM) in SHR and WKY rats. (*P < 0.05, WKY EC(+) vs. WKY EC(+) + L‐NAME, two‐way ANOVA followed by Bonferroni post hoc comparisons). The relaxant response to propofol was calculated as the percent change in the tension induced by NE. Data are presented as mean ± SD. n = 9. EC(+): endothelium‐intact; EC(‐): endothelium‐denuded; L‐NAME: NG‐nitro‐L‐arginine methyl ester.

Wang, Y. , Zhou, Q. , Wu, B. , Zhou, H. , Zhang, X. , Jiang, W. , Wang, L. , and Wang, A. (2017) Propofol induces excessive vasodilation of aortic rings by inhibiting protein kinase Cβ2 and θ in spontaneously hypertensive rats. British Journal of Pharmacology, 174: 1984–2000. doi: 10.1111/bph.13797.