The Central Role of Protein Kinase C Epsilon in Cyanide Cardiotoxicity and Its Treatment

Abstract

In adult mouse myocytes, brief exposure to sodium cyanide (CN) in the presence of glucose does not decrease ATP levels, yet produces profound reduction in contractility, intracellular Ca²⁺ concentration ([Ca²⁺]_i) transient and L-type Ca²⁺ current (I_Ca) amplitudes. We analyzed proteomes from myocytes exposed to CN, focusing on ionic currents associated with excitation-contraction coupling. CN induced phosphorylation of α1_c subunit of L-type Ca²⁺ channel and α₂ subunit of Na⁺-K⁺-ATPase. Methylene blue (MB), a CN antidote that we previously reported to ameliorate CN-induced reduction in contraction, [Ca²⁺]_i transient and I_Ca amplitudes, was able to reverse this phosphorylation. CN decreased Na⁺-K⁺-ATPase current contributed by α₂ but not α₁ subunit, an effect that was also counteracted by MB. Peptide consensus sequences suggested CN-induced phosphorylation was mediated by protein kinase C epsilon (PKCε). Indeed, CN stimulated PKC kinase activity and induced PKCε membrane translocation, effects that were prevented by MB. Pretreatment with myristoylated PKCε translocation activator or inhibitor peptides mimicked and inhibited the effects of CN on I_Ca and myocyte contraction, respectively. We conclude that CN activates PKCε, which phosphorylates L-type Ca²⁺ channel and Na⁺-K⁺-ATPase, resulting in depressed cardiac contractility. We hypothesize that this inhibition of ion fluxes represents a novel mechanism by which the cardiomyocyte reduces its ATP demand (decreased ion fluxes and contractility), diminishes ATP turnover and preserves cell viability. However, this cellular protective effect translates into life-threatening cardiogenic shock in vivo, thereby creating a profound disconnect between survival mechanisms at the cardiomyocyte level from those at the level of the whole organism.

Cyanide (CN) prevents the cyclic reoxidation of the metallo-enzymes of mitochondrial complexes (Cooper and Brown, 2008), thereby mimicking the effects of acute reduction in O₂ supply. Toxic CN exposure may occur in employees involved in the manufacture of plastic or foam (Vogel et al., 1981) and in victims exposed to smoke inhalation during a fire in buildings (Baud et al., 1991; Fortin et al., 2006). Clinically, acute CN toxicity is lethal due to rapid inhibition of brainstem respiratory neuron activity (Solomon, 2000), along with direct reduction in cardiac contractility (Fortin et al., 2006; et al., 2010; ) and induction of ventricular arrhythmias (Bebarta et al., 2012; Haouzi et al., 2017). CN-induced cardiotoxicity is classically ascribed to the mismatch between the high adenosine triphosphate (ATP) demand by contracting cardiac myocytes and reduced ATP supply due to cessation of oxidative phosphorylation. More recent in vivo (Elliott et al., 1989) and in vitro (Cheung et al., 2018) studies indicate that reduction in myocardial and myocyte contractility occurs soon after CN exposure when ATP levels are unchanged. Specifically, we have found that within minutes of CN exposure, L-type Ca²⁺ current (I_Ca) and [Ca²⁺]_i transient amplitudes are severely reduced, thereby adversely affecting excitation-contraction (EC) coupling with resultant diminished myocyte contraction amplitudes (Cheung et al., 2018).

To explore the mechanisms by which CN alters contractility and [Ca²⁺]_i homeostasis, we analyzed proteomes from adult mouse left ventricular myocytes exposed to CN, both in the absence and presence of methylene blue (MB). Methylene blue is a Food and Drug Administration (FDA) approved drug for the treatment of methemoglobinemia. Methylene blue is also an effective antidote for CN poisoning (Brooks, 1933; Eddy, 1931; Geiger, 1932, 1933; Hanzlik, 1933) by restoring Krebs cycle activity and mitochondrial membrane potential through its potent redox properties. In an attempt to simulate a clinically relevant scenario, we specified three essential experimental conditions that are met in vivo. First, myocytes were exposed to NaCN at 100 µM for 10 min, concentrations compatible with in vivo intoxication that could be reversed by antidotes. We have demonstrated in a previous study that at this CN concentration, myocyte contractility was severely depressed but could be rescued by specific antidotes (Cheung et al., 2018). Second, previous studies on CN toxicity were complicated by concomitant inhibition of glycolysis to achieve total metabolic blockade (Borchert et al., 2013; Ju and Allen, 2005; Rodrigo and Standen, 2005; Ruiz-Meana et al., 2009; Taniguchi et al., 1983), thereby making it difficult to extrapolate the experimental findings to in vivo models and to human intoxication. Therefore, all our experiments were performed in the presence of physiological concentrations of glucose. Third, to simulate clinical utility in the field as a CN antidote, MB was administered after NaCN exposure.

MATERIALS AND METHODS

Isolation of adult murine cardiac myocytes

Cardiac myocytes were isolated from the left ventricular (LV) free wall and septum from mice according to the protocol of Zhou et al. (2000) and modified by us (Song et al., 2008, 2012; Tucker et al., 2006; Wang et al., 2009, 2010, 2011). In all experiments, myocytes were used within 2–6 h of isolation. All protocols applied to the mice in this study were approved and supervised by the Institutional Animal Care and Use Committee at Temple University.

In vitro study protocol

For proteomics analysis, protein kinase C epsilon (PKCε) immunofluorescence imaging, PKC kinase assay, membrane fractionation followed by western blotting, I_Ca and Na⁺-K⁺-ATPase current (I_pump) measurements, either phosphate-buffered saline (PBS), MB (20 µg/ml) or NaCN (100 µM) was added at time zero. In a separate group of myocytes, MB (20 µg/ml) or PBS was added 3 min after NaCN (100 µM). Media were changed at 10 min before measurements were performed. In experiments involving PKCε activator (N-myristoylated-HDAPIGYD) or inhibitor (N-myristoylated-EAVSLKPT) peptide (Genemed Synthesis Inc., San Antonio, TX), myocytes were pretreated with the myristoylated peptide (10 µM in 0.02% DMSO final concentration) or DMSO (0.02%, vol/vol) for 30 min, or with phorbol 12-myristate 13-acetate (PMA, 1 µM in 0.02% DMSO, 10 min), washed and bathed in appropriate extracellular recording solutions (±100 µM NaCN) for 10 min before measurements of I_Ca and myocyte contraction.

Mass spectrometry analysis

The label-free proteomics analysis using modified in-stage tip method was performed using the nanoelectrospray ionization tandem MS with a LTQ Orbitrap Elite mass spectrometer (Thermo Scientific, Rockford, IL) as previously described (Zhang, 2018). Specifically, proteins were digested with trypsin according to our published protocols (Barrero et al., 2013; Boden et al., 2008; et al., 2014, 2015; Shanmughapriya et al., 2018; Tomar et al., 2016; Zhang, 2018). The peptides were acidified and loaded onto an activated in-house-made cation stage tip. The peptides were purified and eluted into six fractions using elution buffers as previously described (Zhang, 2018). The purified peptides were analyzed by mass spectrometer with complete system controlled by Xcalibur software (Version 3.0.63). Mass spectra processing was carried out using Mascot Distiller and Sequest HD with Proteome Discoverer 2.0. The generated deisotoped peak list were submitted to an in-house Mascot server 2.2.07 for searching against the Swiss-Prot database (TaxID = 9606, released 2017-05-10. 42, 153 sequences). Mascot search parameters were set as follows: species, mus musculus; enzyme, trypsin with maximal 2 missed cleavage; fixed modification, cysteine carboxymethylation; variable modification, phosphorylation; 20 ppm mass tolerance for precursor peptide ions; 0.2 Da tolerance for MS/MS fragment ions. All peptide matches were filtered using an ion score cutoff of 20. Quantified proteins were selected and clustered by biological functions.

Electrophysiological measurements

Ca²⁺ current (Feldman et al., 2016; Judenherc-Haouzi et al., 2016; Zhang et al., 2015) and I_pump (Song et al., 2012; Wang et al., 2010, 2014) were measured in isolated LV myocytes (30°C) with whole cell patch-clamp. Fire-polished pipettes (tip diameter 4–6 µm) with resistances of 0.8–1.4 MΩ when filled with pipette solutions were used. Compositions of solutions and voltage protocols are given in figure legends.

PKC kinase activity assay

LV myocytes were incubated in recording solution (RS consisting of HEPES-buffered (20 mM, pH 7.4) medium 199 containing 5.5 mM glucose and 1.8 mM [Ca²⁺]_o) and exposed to NaCN (100 µM) or NaCN followed by MB (20 µg/ml added at 3 min) for 10 min, after which myocytes were suspended in lysis buffer (in mM: 50 Tris [pH 8.0], 150 NaCl, 1 Na⁺ orthovanadate, 1 PMSF, 100 NaF, 1 EDTA and 1 EGTA with 0.5% NP-40, 10 µg/ml leupeptin, and 10 µg/ml aprotinin), sonicated and centrifuged at 10 000 rpm and 4°C for 10 min. The supernatant was assayed for PKC kinase activity using the PKC kinase activity assay kit (ab139437; Abcam, Cambridge, MA).

PKCε translocation by cell fractionation measurements

LV myocytes were incubated in RS and exposed to NaCN (100 µM) or NaCN followed by MB (20 µg/ml added at 3 min) for 10 min, after which myocytes were fractionated into cytosol and crude membranes using a two-step centrifugation protocol described previously (Song et al., 2005). Briefly, myocytes were washed 2 times with PBS and suspended into 400 μl of ice-cold Buffer I containing (in mM): 10 Tris (pH 7.5), 1 Na⁺ vanadate, 1 PMSF, 100 NaF, 1 EGTA, and a Complete Protease Inhibitor Cocktail Tablet (Cat^#1697498, Boehringer Mannheim, Indianapolis, IN). After sonication (3 × 15 s), 400 μl of ice-cold Buffer II containing (in mM): 10 Tris (pH 7.5), 300 KCl, 1 Na⁺ vanadate, 1 PMSF, 100 NaF, 1 EGTA, 20% sucrose, and a Complete Protease Inhibitor Cocktail Tablet were added. Myocyte sonicates were centrifuged (10 000 g) at 4°C for 10 min. About 100 µl of the supernatant was stored as “total lysate” at –20°C. The remaining supernatant was centrifuged (100 000 g) at 4°C for 1 h. The resultant supernatant was collected as the cytosolic fraction. After washing the pellet with ice-cold PBS, the crude membrane fraction was suspended in 80 µl of Buffer I and Buffer II (40 µl each) containing 1% Triton X-100 for 10 min at 4°C. Proteins in cytosol and crude membranes were fractionated (10% SDS-PAGE, β-mercaptoethanol), transferred, probed with PKCε antibody (1:1500; Cell Signaling Technology Inc., Boston, MA) and signals were detected by enhanced chemiluminescence (Thermo Scientific). PKCε signal intensities were normalized against the membrane marker calsequestrin (CLSQ).

PKCε translocation by immunofluorescence

Freshly isolated mouse LV myocytes were plated on laminin-coated coverslips, allowed to adhere for 3 h, and treated with RS, NaCN (100 µM), NaCN followed by MB (20 µg/ml added at 3 min), or PMA (1 µM) for 10 min. Myocytes were washed 3 times with PBS containing 2 mM EGTA (PBS-EGTA), followed by fixation with 4% paraformaldehyde in PBS-EGTA for 30 min. After 2 rinses with PBS-EGTA, myocytes were permeabilized for 5 min with 0.05% Triton X-100. Myocytes were rinsed 2 times with PBS-EGTA and once with Blotto (5% nonfat dry milk, 0.1 M NaCl, and 50 mM Tris–HCl; pH 7.4). PKCε antibody (1:100; Research & Diagnostic Antibodies, Las Vegas, NV) diluted in Blotto were added to the cells, incubated at 4°C in the dark overnight and rinsed 3 times with Blotto. Secondary antibodies (Alexafluor 488-labeled goat antirabbit IgG, Invitrogen, Eugene, OR; 1:250) diluted in Blotto were added to the cells, incubated at room temperature in the dark for 60 min, and followed by three PBS-EGTA rinses. Coverslips were mounted to slides with Prolong gold antifade mounting solution (Invitrogen). Confocal images (×63 oil objective, 510 Meta, Carl Zeiss) were acquired at 488-nm excitation and 510-nm emission. The percentage of myocytes showing a predominantly cross-striated fluorescence (translocated PKCε) or predominantly cytosolic/perinuclear fluorescence (nontranslocated PKCε) was determined by blind counting of myocytes on 3 different slides in each treatment group (Miyamae et al., 1998).

Myocyte shortening measurements

Myocytes adherent to coverslips were bathed in 0.7 ml of air- and temperature-equilibrated (37°C) RS. Myocytes were field stimulated to contract (2 Hz) between platinum wire electrodes spaced 2 mm apart. Images of myocytes viewed through an Olympus DApoUV ×40/1.30 numerical aperture oil objective situated in a Zeiss IM35 inverted microscope were captured by a charge-coupled device video camera (Myocam; Ionoptix, Milton, MA). Edge detection algorithm was used to measure myocyte motion, and data were analyzed off-line by Ionoptix software as previously described (Song et al., 2008, 2012; Tucker et al., 2006; Wang et al., 2009, 2010, 2011).

Statistics

All results are expressed as means ± SE. For analysis of I_Ca, I_pump, PKCε/CLSQ, PKC kinase activity, % striated myocytes on immunofluorescence, and the effects of PMA, PKCε activator or inhibitor peptides on I_Ca and contraction amplitudes, one-way ANOVA was used. A commercially available software package (JMP version 14 Pro, SAS Institute, Cary, NC) was used. In all analyses, p ≤ .05 was taken to be statistically significant.

RESULTS

NaCN Acutely Phosphorylates α1_c Subunit of L-type Ca²⁺ Channel and α₂ Subunit of Na⁺-K⁺-ATPase: Reversal by MB

We have previously reported that CN rapidly decreased I_Ca and that administration of MB 3 min after CN restored I_Ca to control levels (Cheung et al., 2018). Because the time course of changes in I_Ca was rapid, we focused on translational modifications on the channel revealed by proteomics. After 10 min of exposure, CN caused phosphorylation of L-type Ca²⁺ channel subunit α1_c at T96 (Figure 1; peptide consensus sequence QGGTtATRPPR with t96 italicized). Addition of MB 3 min after CN prevented phosphorylation of α1_c at T96. It is important to recall that CN has no effect on Na⁺/Ca²⁺ exchange current (I_NaCa) (Cheung et al., 2018). In agreement, our proteomic analysis did not detect any CN-induced phosphorylation changes on the Na⁺/Ca²⁺ exchanger. Another novel finding is that the α₂ but not α₁ subunit of Na⁺-K⁺-ATPase was phosphorylated at S461 and S464 by CN (Figure 2; peptide consensus sequence CIELsCGsVR with the serines italicized) and phosphorylation was prevented by MB.

Figure 1.

Identification of voltage-dependent L-type Ca²⁺ channel α1_c subunit peptide phosphorylation sites in NaCN-treated cardiac myocytes that were dephosphorylated by MB treatment. Freshly isolated mouse myocytes were exposed to PBS (CTL), NaCN (100 µM), MB (20 µg/ml), or NaCN followed by MB (3 min after) for 10 min before subjected to mass spectrometry. Comparisons were made between CTL and CN, CN and CN + MB, and CTL and MB groups. The extent of phosphorylation was determined by quantitating the precursor abundance. The precursor abundance of the phosphorylated peptide in CN treated cells was 1.84e⁶. This peptide was not detectable with MB treatment. MS/MS spectrum of the doubly charged precursor ion at m/z 1221.56624 (fragmentation pattern shown) indicates the phosphorylated site at T96 (italicized). Note the positive charge (R) at +3 suggests phosphorylation is mediated by PKCα, ε, γ, or η (Rust and Thompson, 2011).

Figure 2.

Identification of Na⁺-K⁺-ATPase α₂ subunit peptide phosphorylation sites in NaCN-treated cardiomyocytes that were dephosphorylated by MB treatment. Proteomic analysis was performed in the 4 groups of myocytes as described in Figure 1. The extent of phosphorylation was determined by quantitating the precursor abundance. The precursor abundance of the phosphorylated peptide in CN treated cells was 3.87e⁴ as compared with 1.27e⁴ with MB treatment. MS/MS spectrum of the doubly charged precursor ion at m/z 613.72175 (fragmentation pattern shown) demonstrated the phosphorylated sites at S461 and S464 (italicized). Note GsVR (S464) indicates phosphorylation is mediated by PKCε, whereas EXs (for S461) suggests phosphorylation is mediated by β-adrenergic receptor kinase (Rust and Thompson, 2011).

NaCN Acutely Depresses I_α2: Reversal by MB

To functionally confirm our proteomics results, we treated myocytes with CN (100 µM) for 10 min and observed ∼36.4% decrease (p = .0074) in I_α2 but not I_α1 (p = .1135) when compared with control (Figure 3). Addition of MB 3 min post-CN restored I_α2 to levels indistinguishable from control values (p = .1495) (Figure 3). Methylene blue treatment alone for 10 min had no effects on either I_α1 (p = .8752) or I_α2 (p = .8135) when compared with control (Figure 3).

Figure 3.

NaCN depresses Na⁺-K⁺-ATPase current contributed by α₂ subunit in adult myocytes: reversal by MB. Whole cell patch-clamp recordings were performed at 30°C (Song et al., 2008; Wang et al., 2010). Pipette solutions contained (in mM): 70 Na⁺-aspartate, 20 K⁺-aspartate, 8 CsOH, 7 MgSO₄, 11 EGTA, 10 TEA-Cl, 1 CaCl₂, 5 HEPES, 5 Na₂ATP, and 0.2 GTP (pH 7.2). External solution contained (in mM): 137.7 NaCl, 18 KCl, 2.3 NaOH, 1 MgCl₂, 2 BaCl₂, 1 CdCl₂, 5 HEPES, and 10 glucose, pH 7.4. Holding potential was 0 mV. Ionic current due to Na⁺-K⁺-ATPase activity (I_pump) was separated into two components by taking advantage of the differential sensitivity of α₁ and α₂ subunits to digitalis glycosides. A, After baseline I_pump was recorded from a myocyte previously incubated in PBS (CTL) for 10 min, dihydroouabain (DHO, 5 µM) was added (first arrow), and the difference current is taken to be I_α2. DHO (1 mM) was then added (second arrow), and the additional decrease in current is taken to be I_α1. B, I_α1 and I_α2 measured in a myocyte previously incubated in CN (100 µM) for 10 min. C, I_α1 and I_α2 measured in a myocyte previously exposed to CN (100 µM), followed by addition of MB (20 µg/ml) at 3 min for a total of 10 min before I_pump was measured. D and E, Means ± SE of I_α1 and I_α2 are shown. There are 5 CTL (3 mice), 3 MB (2 mice), 8 CN (5 mice), and 10 CN + MB (5 mice) myocytes. *p = .0074, CTL versus CN. There are no differences (p = .1495) in I_α2 between CTL and CN + MB myocytes. There are no differences (p = .5087) in I_α1 among the 4 groups of myocytes.

NaCN Translocates PKCε From Cytosol to Sarcolemma and Stimulates PKC Kinase Activity: Reversal by MB

CN exposure resulted in phosphorylation of T96 of α1_c subunit of L-type Ca²⁺ channel (Figure 1). The positive charge (R) at +3 (peptide consensus sequence QGGTtATRPPR) suggests T96 phosphorylation was mediated by PKCα, ε, γ or η (Rust and Thompson, 2011). In addition, phosphorylation at S464 of α₂ subunit of Na⁺-K⁺-ATPase by CN (Figure 2) was mediated by PKCε (peptide consensus sequence GsVR indicates PKCε) (Rust and Thompson, 2011). PKCα, ε, γ and η are expressed in rodent cardiac myocytes (Singh, 2017) with PKCε perhaps a major PKC isoform (Bogoyevitch et al., 1993). We therefore evaluated whether CN exposure would activate PKCε in the heart. The canonical model of PKC activation involves translocation from cytosolic compartment to particulate fractions (Singh, 2017; Steinberg, 2012). CN exposure for 10 min resulted in significant translocation of PKCε from cytosol (GAPDH) to crude membrane (CLSQ) fractions (Figure 4). Administration of MB 3 min after CN addition resulted in reduced PKCε translocation to crude membranes (Figure 4). In agreement with previous studies (Bogoyevitch et al., 1993), we could not detect PKCα in the mouse heart (Figure 4).

Figure 4.

NaCN induces translocation of PKCε from cytosol to membrane: reversal by MB. Freshly isolated mouse myocytes were treated with PBS (CTL), NaCN (100 µM), CN followed by MB (20 µg/ml at 3 min) for 10 min, before fractionation into cytosol and crude membranes (Methods). PKCα and PKCε in total, cytosol (GAPDH), and crude membranes (calsequestrin, CLSQ) were detected by Western blotting. A, Representative western blot from 1 of 3 separate myocyte isolations. B, Means ± SE of the ratio PKCε/CLSQ (to normalize loading amounts of crude membrane fractions) are shown. *p < .04, CTL versus NaCN, or NaCN versus NaCN + MB, or CTL versus NaCN + MB.

To corroborate PKCε Western blot data, we used immunofluorescence localization to assess the degree of PKCε translocation in myocytes exposed to NaCN. Under resting conditions, PKCε was diffusely localized in the cytosol in myocytes (Figure 5). Stimulation of myocytes with the nonisoform specific PKC activator PMA resulted in translocation of PKCε to the sarcolemma and t-tubules, as indicated by the appearance of cross-striations (Figure 5). Exposure of myocytes to NaCN for 10 min resulted in PKCε localization in the plasma membrane and t-tubules (Figure 5). By contrast, addition of MB 3 min following NaCN exposure resulted in loss of striated pattern in myocytes (Figure 5). Blind scoring revealed that the percentage of myocytes demonstrating a predominantly cross-striated pattern of fluorescence was significantly (p ≤ .0295) greater in myocytes treated with PMA (77.2 ± 17.5%) or NaCN (67.5 ± 3.7%) when compared with untreated myocytes (18.5 ± 2.7%). Myocytes treated with NaCN followed by MB had similar % of cells demonstrating predominantly cross-striated fluorescence pattern (19.5 ± 2.8%) when compared with untreated myocytes (p = .8141).

Figure 5.

PKCε localizes in the sarcolemma and t-tubules in the presence of NaCN: reversal by MB. Confocal images of adult mouse left ventricular myocytes labeled with anti-PKCε antibodies under control conditions (no antibody), after 10 min of incubation in HEPES-buffered (20 mM, pH 7.4) medium 199 containing 5.5 mM glucose and 1.8 mM [Ca²⁺]_o (recording solution, RS), after 10 min of exposure to PMA (1 µM) or NaCN (100 µM) or NaCN + MB (20 µg/ml of MB added 3 min after NaCN). Under control conditions, PKCε is present diffusely in the cytoplasm. Note appearance of cross-striation pattern in myocytes treated with PMA or NaCN, indicating PKCε translocation to the sarcolemma and t-tubules. By contrast, myocytes incubated in recording solution or exposed to NaCN + MB showed diffuse cytosolic fluorescence. Scale bar = 20 µm. Figures in right panel are magnified views of the boxed sections of myocytes in the left panel.

Finally, we evaluated the effects of CN ± MB on PKC (nonisoform specific) kinase activity. Exposure to NaCN for 10 min increased PKC kinase activity, which was reversed by MB added 3 min after CN (Figure 6). Thus using 3 fundamentally different methods (membrane fractionation, immunofluorescence and biochemical kinase assay), NaCN activated PKCε and the activation was reversed by MB.

Figure 6.

NaCN enhances PKC kinase activity: reversal by MB. Freshly isolated mouse myocytes were treated with PBS (CTL), NaCN (100 µM), CN followed by MB (20 µg/ml at 3 min) for 10 min, followed by assay for PKC kinase activity using the Abcam ab139437 kit according to the manufacturer’s instructions. PKC kinase activity is normalized to CTL activity for each experiment (n = 4). *p = .05, CTL versus CN.

PKCε Activation Mimics the Effects of NaCN on I_Ca

Activation of PKC (nonisoform specific) is known to reduce myocyte contractility (Tucker et al., 2006) and inhibit I_Ca (McHugh et al., 2000; Satoh, 1992). Pretreatment of myocytes with PMA (1 µM, 10 min) resulted in ∼28% decrease in maximal I_Ca amplitude (measured at 0 mV) compared with myocytes pretreated with the vehicle DMSO, thereby mimicking the effects of CN on I_Ca (Figure 7). Exposure to the vehicle DMSO (0.02%) for 10–30 min had no effect on I_Ca or inhibition of I_Ca by CN (Figure 7).

Figure 7.

Inhibition of L-type Ca²⁺ current (I_Ca) in adult myocytes exposed to NaCN: role of PKCε. Ca²⁺ current was measured in isolated adult cardiac myocytes under patch-clamp (holding potential –90 mV; 30°C) (Cheung et al., 2018; Judenherc-Haouzi et al., 2016; Zhang et al., 2015). Pipette solution contained (in mM): 110 CsCl, 20 TEA-Cl, 10 HEPES, 5 MgATP, and 10 EGTA, pH 7.2. Extracellular bathing solution contained (in mM): 137 N-methyl-d-glucamine, 5.4 CsCl, 2 CaCl₂, 1.3 MgSO₄, 20 HEPES, 4 4-aminopyridine, and 15 glucose, pH 7.4. Our solutions were designed to be Na⁺- and K⁺-free. To ensure steady-state SR Ca²⁺ load, 6 conditioning pulses (from –70 to 0 mV, 100 ms, 2 Hz) were delivered before the arrival of each test pulse (from –90 to 0 mV, 60 ms). PBS (CTL), NaCN (100 µM), DMSO (0.02%, vol/vol), or PMA (1 µM) was added at time zero, and I_Ca was measured at 10 min. Another group of myocytes were preincubated with DMSO (0.02%, vol/vol), or the myristoylated PKCε activator (ACT) or inhibitor (INH) peptide (10 µM each) for 30 min, followed by exposure to PBS (for ACT) or NaCN (100 µM; for DMSO and INH) for 10 min before I_Ca measurements. Upper panel: representative I_Ca tracings from a CTL, NaCN, DMSO, DMSO/CN, PMA, ACT and INH/CN myocyte. Lower panel: Means ± SE of peak I_Ca amplitudes (at 0 mV) are shown. There are 6 CTL (5 mice), 7 CN (5 mice), 4 DMSO (4 mice), 4 DMSO/CN (1 mouse), 4 PMA (4 mice), 5 ACT (4 mice), and 3 INH/CN (2 mice) myocytes. *p < .0001, CTL versus CN or DMSO versus DMSO/CN; ^#p = .0005, DMSO versus PMA or DMSO versus ACT; ^$p < .020, DMSO/CN versus INH/CN. There were no differences in maximal I_Ca amplitudes between CTL and DMSO (p = .8204) or between CN and DMSO/CN (p = .4761), indicating DMSO at 0.02% (vol/vol) had no effects on I_Ca or its inhibition by NaCN.

Activation of PKCε is known to reduce I_Ca in cardiac myocytes (El Khoury et al., 2014; Hu et al., 2000). In addition, our present findings indicate that CN induced phosphorylation of T96 of α1_c subunit of L-type Ca²⁺ channel (Figure 1) and S464 of α₂ subunit of Na⁺-K⁺-ATPase (Figure 2) was likely by PKCε and that CN activated PKCε (Figs. 4–6). These considerations led us to focus our attention on whether activation of PKCε could mimic the deleterious effects of CN. Preincubation with the cell-permeable myristoylated PKCε peptide activator (HDAPIGYD) (Hu et al., 2000) resulted in ∼22.8% reduction in I_Ca amplitude compared with DMSO control, an effect that is indistinguishable from that of PMA (∼28.0%; Figure 7). Conversely, preincubation with myristoylated PKCε peptide inhibitor (EAVSLKPT) (El Khoury et al., 2014; Hu et al., 2000) before CN exposure blunted the decrease in I_Ca amplitude when compared with CN alone (Figure 7).

PKCε Activation Mimics the Effects of NaCN on Contraction

Exposure to either CN (Cheung et al., 2018) or PMA (Tucker et al., 2006) resulted in reduction in myocyte contraction amplitudes: observations which were confirmed in the present study (Figure 8). Exposure to the vehicle DMSO (0.02%) for 10–30 min had no effect on maximal contraction amplitude or its inhibition by CN whereas preincubation with myristoylated PKCε peptide activator significantly inhibited myocyte contraction (Figure 8), thereby mimicking the deleterious effects of CN. By contrast, preincubation with myristoylated PKCε peptide inhibitor partly ameliorated contractile dysfunction in myocytes exposed to CN (Figure 8).

Figure 8.

Inhibition of contraction in adult myocytes exposed to NaCN: role of PKCε. Freshly isolated myocytes from mouse left ventricle and septum were plated on laminin-coated coverslips, bathed in HEPES-buffered (20 mM; pH 7.4) Medium 199 ([Ca²⁺]_o 1.8 mM; 5.5 mM glucose) and paced (2 Hz) to contract (37°C) (Methods). Myocytes were treated with PBS (CTL), NaCN (100 µM), DMSO (0.02%, vol/vol) or PMA (1 µM) for 10 min before contraction was measured. Another group of myocytes were preincubated with DMSO (0.02%, vol/vol), or the myristoylated PKCε activator (ACT) or inhibitor (INH) peptide (10 µM each) for 30 min, followed by exposure to PBS (for ACT) or NaCN (100 µM; for DMSO and INH) for 10 min before contraction measurements. There were no differences in maximal contraction amplitudes in myocytes exposed to DMSO for 10 or 30 min and the data are grouped together. A, Representative contraction tracings from a CTL, NaCN, DMSO, DMSO/CN, PMA, ACT, and INH/CN myocyte. B, Means ± SE of contraction amplitudes are shown. There are 11 CTL (3 mice), 23 CN (4 mice), 12 DMSO (2 mice), 11 DMSO/CN (1 mouse), 7 PMA (2 mice), 7 ACT (2 mice), and 6 INH/CN (2 mice) myocytes. *p < .0001, CTL versus CN or DMSO versus DMSO/CN; ^#p < .008, DMSO versus PMA or DMSO versus ACT; ^$p < .007, DMSO/CN versus INH/CN. There were no differences in maximal contraction amplitudes between CTL and DMSO (p = .9288) or between CN and DMSO/CN (p = .8296), indicating DMSO at 0.02% (vol/vol) had no effects on myocyte contraction amplitudes or its inhibition by NaCN.

DISCUSSION

In isolated perfused ferret hearts, exposure to NaCN in the presence of glucose for 5–10 min reduced developed pressure by 47 ± 2% but ATP levels remained unchanged (Elliott et al., 1989). Similarly, in paced mouse LV myocytes incubated with physiological concentrations of glucose, NaCN rapidly depressed contraction amplitudes by ∼45% without significant effect on cellular ATP levels (Cheung et al., 2018). These observations indicate that decreased myocyte contractility at early stages of CN intoxication is unlikely due to the direct effect of ATP depletion. The CN-induced ∼45% reduction in myocyte contraction amplitude was comparable to the ∼39% reduction in [Ca²⁺]_i transient amplitude and to the ∼48% decrease in peak I_Ca amplitude (Cheung et al., 2018), suggesting decreased contraction amplitude was due to acute alterations in myocyte Ca²⁺ homeostasis by CN.

The determinants of [Ca²⁺]_i transient amplitude are: amount of Ca²⁺ trigger to initiate sarcoplasmic reticulum (SR) Ca²⁺ release, SR Ca²⁺ content, and sensitivity of ryanodine receptors (RyR2) to Ca²⁺ (gain). L-type Ca²⁺ channel provides the Ca²⁺ trigger for SR Ca²⁺ release and contributes to SR Ca²⁺ filling and is therefore a key determinant in [Ca²⁺]_i transient amplitudes. Thus, the first major finding is that CN exposure rapidly induced activation of PKCε which phosphorylated T96 of α1_c subunit of L-type Ca²⁺ channel, resulting in depressed I_Ca which led to reduced [Ca²⁺]_i transient and myocyte contraction amplitudes (Cheung et al., 2018). The relationship between CN exposure and PKCε activation in the heart is supported by the results of experiments utilizing cell-permeable PKCε activator and inhibitor peptides on I_Ca and contraction amplitudes, CN stimulation of PKC kinase activity and CN-induced PKCε translocation to the sarcolemma. Previous studies demonstrated that KCN rapidly induced PKC translocation to particulate fractions and increased PKC enzymatic activities in rat cerebellar and hippocampal slices (Rathinavelu et al., 1994). In addition, PKC activation decreased myocyte contraction (Tucker et al., 2006) and I_Ca amplitudes (McHugh et al., 2000; Satoh, 1992). Specifically, PKCε activation inhibited I_Ca in cardiac myocytes (El Khoury et al., 2014; Hu et al., 2000). Finally, pretreatment with the PKC inhibitor 1-(5-isoquinoline-sulfonyl)-2-methylpiperazine (H-7) before NaCN administration (4.2 mg/kg) blunted the decrease of phosphocreatinine in the brain and improved Yucatan micropig survival from 40 to 80% 60 min post-CN injection (Maduh et al., 1995).

The second major finding is that brief CN exposure resulted in phosphorylation of α₂ but not α₁ subunit of Na⁺-K⁺-ATPase, functionally manifested as a decrease in I_α2 but not I_α1. To our knowledge this is the first report of deleterious effects of CN on Na⁺-K⁺-ATPase activity. Phosphorylation of one of the serines (S464) by CN is mediated by PKCε. This conclusion is supported by the observation that PKCε activation inhibited I_pump in rabbit myocytes (Buhagiar et al., 2001; White et al., 2009). Compared with α₁ subunit, the α₂ subunit of Na⁺-K⁺-ATPase is much more abundant in the transverse (t) tubules (Berry et al., 2007; Despa et al., 2003; Swift et al., 2007) in which the EC coupling machinery (L-type Ca²⁺ channel, RyR2, voltage gated Na⁺ channel and Na⁺/Ca²⁺ exchanger) is concentrated (Bers, 2002; Scriven et al., 2000). Current opinion favors the concept that α₂-containing pumps are principally concerned with regulation of myocyte contractility, whereas α₁-containing pumps are involved with control of bulk intracellular Na⁺ (Fuller et al., 2013; James et al., 1999; Swift et al., 2007). Selective inhibition of α₂ subunit of Na⁺-K⁺-ATPase by CN will increase submembranous Na⁺ concentration in the t-tubules (Despa et al., 2002), thereby reducing the thermodynamic driving force of the Na⁺/Ca²⁺ exchanger to extrude Ca²⁺, resulting in enhanced SR Ca²⁺ load (Nuss and Houser, 1992) and increased myocyte contractility. This conclusion is in agreement with observations in α₂ haplo-insufficent hearts in which contractility is enhanced (James et al., 1999). Viewed in this context, selective inhibition of α₂ subunit of Na⁺-K⁺-ATPase by CN is likely a compensatory mechanism to partially preserve myocyte contractility when I_Ca is severely reduced.

Activation of PKCε has been reported in ischemic heart injury (Chen et al., 2001; Dorn et al., 1999; Inagaki et al., 2006). Both CN intoxication and acute ischemia share the common pathway of altering oxidative phosphorylation. By systematically comparing the coping mechanisms in cells from hypoxia-sensitive and hypoxia-resistant animals subjected to hypoxia, Peter Hochachka proposed some 30 years ago (Hochachka, 1986) that in order to decrease energy demand to survive prolonged periods of hypoxia, cells must decrease cell membrane permeability by regulating ion channel density and activity: the so called “channel arrest” hypothesis (Robertson, 2017). The channel arrest hypothesis was validated in anoxia-tolerant turtle hepatocytes (Buck et al., 1993) and recently revisited (Robertson, 2017). Viewed in the larger context of cellular survival from CN intoxication, PKCε activation in myocytes with the resultant decreases in I_Ca and I_pump (channel arrest) certainly represents a very relevant mechanism of protection, but at the expense of reduced overall cardiac contractility which ultimately is detrimental to survival at the level of the whole animal (Haouzi et al., 2017).

We have recently demonstrated that MB enhanced survival of rats exposed to lethal CN concentrations (Haouzi et al., 2018) and that improved survival was mediated by rapid reversal of CN-induced cardiogenic shock by MB (Haouzi et al., 2017). At the myocyte level, MB restored CN-induced reductions in contraction and [Ca²⁺]_i transient amplitudes, systolic [Ca²⁺]_i, resting membrane potential, action potential amplitude, depolarization-activated K⁺ currents, I_Ca and mitochondrial membrane potential ΔΨ_m, and elevations in superoxide levels towards normal (Cheung et al., 2018). In addition to the previously proposed mechanisms by which MB counteracts CN cardiotoxicity including: (1) restoration by reduced MB (leucomethylene blue; LMB) of the redox environment of cardiac ion channels altered directly or indirectly by CN through reactive oxygen species (ROS) production (Cheung et al., 2018; Zima and Blatter, 2006); (2) a direct effect of MB/LMB on the mitochondrial electron transport chain complexes (Wiklund et al., 2007; Zhang et al., 2006); and (3) an increased capacity of hemoglobin to trap CN (Haouzi et al., 2018); the results of the present study add a fourth mechanism: MB reverses PKCε phosphorylation of membrane ion channels after brief CN exposure.

In summary, exposure to NaCN at a concentration potentially lethal to humans induced rapid phosphorylation of α1_c subunit of L-type Ca²⁺ channel and α₂ subunit of Na⁺-K⁺-ATPase in adult cardiac myocytes, resulting in decreases in ionic currents and adversely affecting EC. Examination of peptide consensus sequence of CN-induced phosphorylation sites, and experimental results with cell-permeable PKCε activation and inactivation peptides indicated CN-induced phosphorylation was mediated by PKCε. NaCN rapidly stimulated PKC kinase activity and induced PKCε translocation to cardiac sarcolemma. Methylene blue administered after NaCN reduced PKC kinase activity, PKCε translocation to membranes and phosphorylation of L-type Ca²⁺ channel and Na⁺-K⁺-ATPase, and restored Ca²⁺ and Na⁺ pump currents to normal. We conclude that MB prevented cardiovascular collapse at early stages of CN intoxication partly by inhibiting PKCε activation, thereby preserving ionic currents intimately associated with EC.

ETHICS STATEMENT

All protocols and procedures applied to the mice in this study were approved by the Institutional Animal Care and Use Committees of Temple University and Pennsylvania State University College of Medicine.

DECLARATION OF CONFLICTING INTERESTS

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

FUNDING

National Heart, Lung, and Blood Institute (RO1-HL123093, RO1-HL137426); National Institute of Neurological Disorders and Stroke (UO1-NS097162, R21-NS098991).

AUTHOR CONTRIBUTION STATEMENT

J.Y.C. and P.H. conceived and designed the research, analyzed experimental data, interpreted experimental results, drafted manuscript, edited and revised and approved manuscript. S.M., J.W., X.Q.Z., J.S., C.M., D.T., H.Y., and A.J.-H. designed and performed experiments, analyzed experimental data, interpreted experimental results, prepared figures, edited and revised, and approved manuscript.