Pyridostigmine improves cardiac function and rhythmicity through RyR2 stabilization and inhibition of STIM1‐mediated calcium entry in heart failure

Abstract Heart failure (HF) is characterized by asymmetrical autonomic balance. Treatments to restore parasympathetic activity in human heart failure trials have shown beneficial effects. However, mechanisms of parasympathetic‐mediated improvement in cardiac function remain unclear. The present study examined the effects and underpinning mechanisms of chronic treatment with the cholinesterase inhibitor, pyridostigmine (PYR), in pressure overload HF induced by transverse aortic constriction (TAC) in mice. TAC mice exhibited characteristic adverse structural (left ventricular hypertrophy) and functional remodelling (reduced ejection fraction, altered myocyte calcium (Ca) handling, increased arrhythmogenesis) with enhanced predisposition to arrhythmogenic aberrant sarcoplasmic reticulum (SR) Ca release, cardiac ryanodine receptor (RyR2) hyper‐phosphorylation and up‐regulated store‐operated Ca entry (SOCE). PYR treatment resulted in improved cardiac contractile performance and rhythmic activity relative to untreated TAC mice. Chronic PYR treatment inhibited altered intracellular Ca handling by alleviating aberrant Ca release and diminishing pathologically enhanced SOCE in TAC myocytes. At the molecular level, these PYR‐induced changes in Ca handling were associated with reductions of pathologically enhanced phosphorylation of RyR2 serine‐2814 and STIM1 expression in HF myocytes. These results suggest that chronic cholinergic augmentation alleviates HF via normalization of both canonical RyR2‐mediated SR Ca release and non‐canonical hypertrophic Ca signaling via STIM1‐dependent SOCE.


| INTRODUC TI ON
Heart failure (HF) is a significant cause of mortality and morbidity in the United States. With the ageing population, HF incidence is expected to increase over time. 1 Autonomic imbalance is a key component of the pathophysiology of HF. 2 Following a decrement in cardiac output, a compensatory increase in sympathetic outflow results in increased norepinephrine release, which acutely improves ventricular contractility and heart rate to maintain cardiac output.
Over time, however, chronic sympathetic stimulation leads to maladaptive cardiac remodelling. Conversely, parasympathetic activity is withdrawn in patients with HF, 3 which results in decreased heart rate variability (HRV) and baroreflex sensitivity (BRS) that are correlated with increased mortality. 4 , 5 Treatments to restore autonomic balance by increasing parasympathetic outflow have shown utility improving HF morbidity and mortality, although the results have not been consistent. [6][7][8][9][10][11][12][13][14][15][16] Indeed, vagal nerve stimulation has been shown to increase survival in post-MI rats and improve autonomic balance in dogs with HF. [6][7][8] In humans, evidence from initial clinical trials [9][10][11] suggested that vagal nerve stimulation (VNS) may be a promising treatment for patients with HF via improvement in ejection fraction and reduced end-diastolic volume.
On the other hand, pharmacological agents that increase acetylcholine levels in the neuro-effector junction would be predicted to increase cholinergic transmission similar to vagal nerve stimulation. Pyridostigmine (PYR), an FDA-approved acetylcholinesterase inhibitor, prevents degradation of acetylcholine (ACh) thus increasing ACh concentration in the synaptic cleft. Lataro et al 12 demonstrated improved cardiac performance associated with increased VEGF production following chronic PYR administration in post-MI rats. Moreover, in human HF, PYR was shown to prevent premature ventricular complexes, improve heart rate recovery following exercise and improve short-term HRV. [13][14][15] Thus, a developing line of evidence implicates PYR as a potentially non-invasive therapeutic option for cardiovascular disease. 12,[15][16][17][18][19][20][21][22][23][24][25][26][27][28][29] However, the mechanisms of PYR treatment and therapeutic efficacy in HF remain to be determined.
Alterations of calcium (Ca) release via cardiac ryanodine receptors (RyR2) are thought to contribute to several key pathologies in HF including hypertrophy, arrhythmogenesis and reduced contractility. 30 Dysfunctional RyR2s in HF have been linked to altered sympathetic regulation associated with β-AR-dependent stimulation of CaMKII with subsequent hyper-phosphorylation of RyR2 at serine 2814. 31,32 More recently, hypertrophy, HF and arrhythmia have been linked to up-regulation of store-operated Ca entry (SOCE). 33,34 SOCE occurs when lowering of luminal Ca prompts the SR protein, STIM1, to actuate Ca entry through plasmalemmal Ca channels. 35 SOCE, considered to be most prevalent in non-excitable cells, has been shown to operate in diseased cardiomyocytes in parallel with RyR2-mediated Ca signaling. 36 However, the relative roles of these Ca signaling mechanisms in cardiac disease and in the beneficial effects of muscarinic stimulation remain to be elucidated.

| Transverse aortic constriction
Approximately 50 age-matched (2 month) male C57BL/6J mice (Wild-Type, WT, Jackson Labs #000664) were anaesthetized with 2% isoflurane and intubated for artificial ventilation at 120-160 breaths per minute, of 0.2-0.35 mL. Heating pads were used to keep body temperature at 37°C throughout the procedure. The transverse aorta was accessed via a left lateral thoracotomy and 6-0 suture used to ligate the aorta overlying a blunted 25-or 27-gauge needle. Mice were recovered on a heating pad and assessed for cardiac dysfunction using echocardiography.

| Osmotic pump implantation
Mice subjected to TAC surgery recovered for 7 days prior to osmotic pump implantation. Experimental groups were defined as CTL (Control, no TAC surgery, no osmotic pump), TAC (TAC surgery, implanted with 0.9% saline osmotic pump) and TAC + PYR (TAC surgery, implanted with 2-10 mg/kg pyridostigmine bromide). ( Figure S1A) Prior to implantation, osmotic pumps (Alzet model 1004) were filled with pyridostigmine bromide (PYR) or sterile 0.9% saline. Pumps were primed by incubation in sterile 0.9% saline at 37°C for 30 minutes prior to insertion. Mice were anaesthetized with 2% isoflurane, and osmotic pumps were implanted subcutaneously above right hind limb. PYR was supplied at range of 2-10 mg/kg/day for 28 days at a volume of 0.11 µL/hr.

| Cardiomyocyte isolation
Intact ventricular myocytes were obtained by enzymatic digestion as previously described. 37 Briefly, mice were anaesthetized with 5% isoflurane in 95% oxygen until a deep plane of anaesthesia was achieved. Hearts were rapidly excised and cannulated through the aorta for perfusion with ice-cold calcium-free Tyrode's solution containing (in mM) 140 NaCl, 5.4 KCl, 0.5 MgCl 2 , 10 HEPES and 5.5 glucose with pH 7.4. Cannulated hearts were then switched to a gravity flow Langendorff apparatus containing calcium-free Tyrode's solution with a temperature of 37˚C. Hearts were perfused for 5 minutes before switching to a perfusion solution containing Liberase TH (0.24 U; Roche) for digestion of connective tissue.
Following enzymatic digestion, hearts were minced and triturated in perfusion solution containing BSA (20 mg/mL).

| Echocardiography
Echocardiographic analysis was performed on mice anaesthetized with isoflurane (1.5% in 1 L/min oxygen). Mice were immobilized on a heated imaging stage during image acquisition. Long-and shortaxis analyses were conducted using the GE LOGIQ E ultrasound machine. Analysis was conducted (M-Mode) following acquisition using at least three non-adjacent contractions. Operators were blinded to experimental group.

| Electrocardiography
Electrocardiography (ECG) recordings were performed before and after epinephrine and caffeine challenge as previously described. 38 Briefly, ECG recordings were obtained from mice anaesthetized with isoflurane (1-1.5%). Subcutaneous electrodes were placed in the left, right upper and right lower limbs for ECG recording (PL3504 PowerLab 4/35, ADInstruments). After a baseline recording (5 minutes), a stress challenge was performed by administering an intraperitoneal injection with epinephrine (Epi, 1.5 mg/kg) and caffeine (Caff, 120 mg/kg). ECG recording continued for 15 minutes after challenge. Analysis was performed using the LabChart 7.3 program (ADInstruments). Ventricular arrhythmias were defined as frequent ectopies, bigeminies and/or ventricular tachycardia (VT).

| Acetylcholinesterase assay
Blood samples (80-150 µL) were collected sublingually from mice in tubes containing 3% heparin. Blood was centrifuged at 8000 RPM for 4-5 minutes at room temperature, and plasma was collected. Acetylcholinesterase activity was detected using Abcam acetylcholinesterase assay kit (Abcam) following manufacturer's instructions. Plasma AChE activity was detected as absorbance change (410 ± 5 nm) using a Fisher Scientific MultiSkan FC spectrophotometer.

| Calcium imaging
Ventricular myocyte cytoplasmic Ca was recorded as described previously. 39 Myocytes were plated on 12 mm coverslips covered by laminin (50 mg/mL). Cells were incubated with a Ca-sensitive dye Fluo-4 AM (9 μM, Thermo Fisher Scientific) in a low Ca  Images were processed with ImageJ software (NIH).

| mRNA expression analysis
RNA was extracted from whole heart tissue using TRIzol

| Store-operated calcium entry (SOCE)
SOCE events were measured following previously published methods. 36 Briefly, cardiac myocytes were loaded with Fluo-4 AM for

| Statistical analysis
Statistical analyses were completed using Origin and/or Microsoft Excel. Unpaired one-tailed Student's t test or 1-way analysis of variance (ANOVA) with post hoc Fisher's test or Tukey HSD was used to test statistical significance between experimental groups. Outlier data points were excluded by using the Grubbs outlier test with significance level of Alpha 0.05.

| RE SULTS
We investigated the effects of chronic treatment with the acetylcholinesterase inhibitor, pyridostigmine bromide (PYR), on in vivo cardiac function and myocyte Ca handling in TAC mice. For chronic PYR treatment, 7 days after TAC surgery mice were implanted F I G U R E 1 Pyridostigmine improves in vivo cardiac function following TAC surgery but does not prevent TAC-induced structural remodelling of the heart. (A) Plot of acetylcholinesterase activity 28 days after osmotic pump implantation (*P < 0.05 vs. CTL, †P < 0.05 vs. TAC), 1-way ANOVA + Tukey HSD. For acetylcholinesterase activity, n = 10 mice per group. (B) Plot of ventricular ejection fraction (EF, expressed as a %) TAC resulted in a significant reduction in EF vs. CTL mice (*P < 0.000001 vs. CTL). However, chronic pyridostigmine treatment significantly increased EF vs. TAC mice ( †P < 0.0001 vs. TAC), 1-way ANOVA + Tukey HSD. For echocardiographic measurements, analysis was conducted following acquisition using at least three non-adjacent contractions. (C) Plot of heart weight (HW) to tibia length (TL) ratio. TAC mice exhibited a significant increase in HW/TL ratio vs. CTL mice (*P < 0.0001 vs. CTL) and chronic pyridostigmine treatment failed to reduce HW/TL following TAC (*P < 0.0001 vs. CTL), 1-way ANOVA + Tukey HSD. (D) Plot of heart weight (HW) to bodyweight (BW) ratio. TAC hearts show a significant increase in HW/BW vs. CTL hearts (*P < 0.0001 vs. CTL), and pyridostigmine failed to reduce HW/BW ratio in response to TAC (*P < 0.0001 vs. CTL), 1-way ANOVA + Tukey HSD. For hypertrophy measurements, n = 7-10 mice per group. For all plots, mean ± SD of data indicated by line and a minimum of three experiments per group with osmotic pumps delivering the drug (at 0.11 µL/hr) for 28 days ( Figure S1A). To determine in vivo inhibition of acetylcholinesterase, plasma was separated from blood following 28 days of osmotic pump treatment. As shown in Figure 1A, chronic PYR treatment resulted in a significant inhibition of plasma acetylcholinesterase activity compared to CTL and TAC samples at 28 days post-implantation.

| PYR improves ventricular function in TAC mice
Echocardiographic analysis was conducted to determine in vivo ventricular function and hypertrophy following TAC surgery ( Figure 1B and Figure S1B). A significant decrease in ventricular function measured by ejection fraction (EF) was observed in untreated TAC mice consistent with pressure overload-induced HF ( Figure 1B).
Interestingly, chronic PYR treatment significantly increased EF compared to TAC mice, although the values of these parameters stayed below CTL levels ( Figure 1B). Consistent with previous studies, 40 EF in TAC mice showed substantial variability, although with no obvious batch-dependent correlation between the TAC and PYR groups ( Figure S1C). Echocardiographic analysis also showed increased interventricular septal thickness at end-diastole and end-systole (IVSd/ IVSs) in TAC hearts indicating cardiac hypertrophy (Table S2). PYR treatment resulted in a significant, albeit incomplete improvement in IVSd in TAC hearts (Table S2). Structural remodelling was further assessed by measuring the heart weight/body weight (HW/BW) and heart weight/tibial length (HW/TL) in CTL, TAC and TAC + PYR mice ( Figure 1C,D). TAC mice exhibited increased HW/TL and HW/ BW relative to the CTL group, but these parameters were not significantly different from the chronic PYR treatment group ( Figure 1C,D).  Figure 2C,D). PYR treatment resulted in a partial normalization of MYH6 mRNA and nearly complete return to CTL values of MYH7, NPPA and NPPB mRNA levels (Figure 2A-D). Collectively, these results suggest PYR treatment improves ventricular function and partially alleviates adverse ventricular remodelling in pressure overload-induced HF.

| PYR reduces arrhythmia susceptibility in TAC mice
Pressure overload HF is associated with increased risk of ventricular arrhythmia. Therefore, we performed ECG measurements to assess the effects of PYR on arrhythmia vulnerability in TAC mice. ECG measurements were performed in anaesthetized CTL, TAC and TAC + PYR mice challenged with epinephrine and caffeine. In congruence with previous reports, untreated TAC mice exhibited enhanced predisposition to arrhythmogenesis indicated by frequent premature ventricular contractions (PVCs) ( Figures 3A,C) Notably, the stress challenge failed to induce PVCs in the TAC + PYR group ( Figure 3B-C). Taken together, these data suggest that PYR treatment confers protection against arrhythmia in pressure overload HF.

| PYR improves myocytes Ca handling in TAC myocytes
Aberrant SR Ca release in the form of spontaneous cytosolic Ca waves is a characteristic feature of HF and an established cause  (Table S1). Chronic PYR treatment had no effect on Ca transient amplitude, while decreasing the decay rate of the Ca transients in TAC myocytes (Table S1). SR Ca content in the three experimental groups was assessed through application of caffeine (10 mM). We found no significant alterations in the SR Ca content in TAC and TAC + PYR myocytes relative to control and TAC, respectively ( Figure S2).  (Figure 4). Notably, chronic treatment with PYR reduced the incidence of arrhythmogenic calcium waves in TAC myocytes. (Figure 4) Overall, our studies indicate PYR reduces aberrant calcium release and confers protection against arrhythmia in pressure overload HF.

| PYR improves dysregulated CAMKII-RyR2 S2814 signaling
Aberrant SR Ca release in HF has been associated with increased phosphorylation of RyR2 at serine 2814 (S2814) by CaMKII in both human and animal models. 31,32 We examined the effects of PYR treatment on RyR2 CaMKII phosphorylation in TAC hearts using immunoblot assays. In TAC hearts, Western blot analyses revealed a significant increase in RyR2 S2814 phosphorylation, consistent with previous reports (Figure 5A,B). Notably, chronic PYR treatment resulted in a significant reduction in RyR2 S2814 phosphorylation compared to untreated TAC hearts ( Figure 5A,B). Additionally, we investigated the upstream signaling effector CAMKII to determine the link between PYR treatment and decreased RyR2 S2814 phosphorylation. As shown in Figure 6A and C, TAC resulted in a significant increase in CaMKII activation level indexed by CAMKII phosphorylation at T287. Notably, PYR treatment blunted the increase in CaMKII-T287 phosphorylation following TAC surgery. Taken together, our results indicate a mechanistic link for PYR protection in response to pressure overload by preventing dysregulation of SR Ca release via reduced CaMKII phosphorylation of RyR2 S2814.

| PYR reduces STIM1-dependent SOCE in HF myocytes
Up-regulated STIM1-dependent Ca signaling has been implicated in pathologic hypertrophy, HF and arrhythmogenesis. 34,42 We investigated the effects of PYR treatment on SOCE and its effectors, STIM1 and Orai1, in TAC myocytes. We performed measurements of SOCE in the form of local Ca entry events (LoCEs, 36 ) in CTL, TAC and TAC + PYR myocytes. As recently reported, LoCEs represent Ca entry via store-operated Ca entry sites predominantly localized at myocyte intercalated discs (IDs). 36 In accordance with our previous report, 36 LoCEs occurred mainly at myocyte ends ( Figure 6A).
LoCEs magnitude and incidence were significantly increased in TAC myocytes compared with CTL cells (Figure 6A-E). Notably, PYR significantly reduced SOCE in TAC myocytes ( Figure 6D-E). To assess whether the observed changes in SOCE with PYR treatment were associated with similar changes with STIM1 and ORAI1 expression, we measured protein levels with Western blot. STIM1 levels were significantly increased in TAC compared to both CTL and TAC + PYR groups ( Figure 7A  levels between the three groups ( Figure S3). Collectively, these results suggest that the beneficial effects of PYR on TAC hearts are associated with down-regulation of STIM1-mediated Ca signalling. Thus, our results regarding the effects of PYR on myocyte Ca handling in the mouse might be also relevant to human. Further studies are needed to define the prevailing mechanisms of the beneficial effects, and the therapeutic efficacy of AChE inhibitors is needed to support their use in patients with cardiac disease.

CO N FLI C T O F I NTE R E S T
The authors declare they have no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Stephen

DATA AVA I L A B I L I T Y S TAT E M E N T
All data are contained within the manuscript.