Cardiac sympathetic nerve transdifferentiation reduces action potential heterogeneity after myocardial infarction

Abstract

Cardiac sympathetic nerves undergo cholinergic transdifferentiation following reperfused myocardial infarction (MI), whereby the sympathetic nerves release both norepinephrine (NE) and acetylcholine (ACh). The functional electrophysiological consequences of post-MI transdifferentiation have never been explored. We performed MI or sham surgery in wild-type (WT) mice and mice in which choline acetyltransferase was deleted from adult noradrenergic neurons [knockout (KO)]. Electrophysiological activity was assessed with optical mapping of action potentials (AP) and intracellular Ca²⁺ transients (CaT) in innervated Langendorff-perfused hearts. KO MI hearts had similar NE content but reduced ACh content compared with WT MI hearts (0.360 ± 0.074 vs. 0.493 ± 0.087 pmol/mg; KO, n = 6; WT, n = 4; P < 0.05). KO MI hearts also had higher basal ex vivo heart rates versus WT MI hearts (328.5 ± 35.3 vs. 247.4 ± 62.4 beats/min; KO, n = 8; WT, n = 6; P < 0.05). AP duration at 80% repolarization was significantly shorter in the remote and border zones of KO MI versus WT MI hearts, whereas AP durations (APDs) were similar in infarct regions. This APD heterogeneity resulted in increased APD dispersion in the KO MI versus WT MI hearts (11.9 ± 2.7 vs. 8.2 ± 2.3 ms; KO, n = 8; WT, n = 6; P < 0.05), which was eliminated with atropine. CaT duration at 80% and CaT alternans magnitude were similar between groups both with and without sympathetic nerve stimulation. These results indicate that cholinergic transdifferentiation following MI prolongs APD in the remote and border zone and reduces APD heterogeneity.

NEW & NOTEWORTHY Cardiac sympathetic neurons undergo cholinergic transdifferentiation following myocardial infarction; however, the electrophysiological effects of corelease of norepinephrine and acetylcholine (ACh) have never been assessed. Using a mouse model in which choline acetyltransferase was deleted from adult noradrenergic neurons and optical mapping of innervated hearts, we found that corelease of ACh reduces dispersion of action potential duration, which may be antiarrhythmic.

INTRODUCTION

The autonomic nervous system is the primary mechanism responsible for the cardiac fight-or-flight response, whereby heart rate and contractility increase rapidly to meet physiological demands. A fundamental feature of autonomic control of the heart is its yin-yang nature: sympathetic nerves release norepinephrine (NE), which acts primarily on β-adrenergic receptors to increase heart rate, contractility, and conduction velocity, whereas parasympathetic nerves release acetylcholine (ACh), which produces opposing effects via muscarinic receptor stimulation. However, a subset of sympathetic neurons that innervate noncardiac targets produce ACh, and multiple studies have shown that sympathetic neurons can undergo phenotypic remodeling in diseased hearts (8, 16, 19), where transdifferentiation of sympathetic neurons results in corelease of NE and ACh.

In a previous study (16), we demonstrated that sympathetic neurons produce ACh together with NE after myocardial infarction (MI) in the mouse heart. Furthermore, selective deletion of choline acetyltransferase (ChAT), the enzyme that synthesizes ACh, in adult sympathetic neurons prevents the MI-induced rise in cardiac ACh. The functional electrophysiological consequences of sympathetic cholinergic transmission in the heart, however, remain unknown. We previously showed in the rabbit heart that coadministration of exogenous ACh and NE blunted NE-induced increases in intracellular Ca²⁺ transients (CaT) and prolonged action potential duration (APD), suggesting diminished contractility and reduced electrical adaptation to fast heart rates (16), which may have detrimental effects during sympathetic activation and the resulting cardiac fight-or-flight response.

In the present study, we sought to investigate the electrophysiological impact of post-MI sympathetic cholinergic transdifferentiation using a mouse model in which ChAT was deleted from adult noradrenergic neurons. Ex vivo, fully innervated Langendorff-perfused hearts (5) from wild-type (WT) and inducible ChAT knockout (KO) mice were simultaneously optically mapped with voltage-sensitive and Ca²⁺-sensitive dyes. Our results indicate that sympathetic cholinergic transmission following MI alters the dispersion of APD with minimal effects on Ca²⁺ handling and contractility.

MATERIALS AND METHODS

Animals.

Wild-type (WT) C57BL/6J mice were obtained from Jackson Laboratories (West Sacramento, CA). Inducible dopamine beta hydroxylase (DBH-CreERT2) driver mice were generated as previously described (21) and were obtained from Dr. Hermann Rohrer (Max Planck Institute, Frankfurt, Germany). Choline acetyltransferase (ChAT) is the enzyme that synthesizes ACh, and ChAT^lox/lox mice were obtained from Dr. Josh Sanes (Harvard University, Cambridge, MA). To generate mice whose adult sympathetic neurons lack ChAT, we crossed DBH-CreERT2 mice with ChAT^lox/lox mice. Offspring that were homozygous for ChAT^lox/lox and expressed at least one copy of DBH-CreERT2 (ChAT^{DBH-CreERT2/lox}) were treated with tamoxifen (2 mg/day ip) for 7 days to stimulate expression of Cre recombinase and generate mice whose noradrenergic neurons lacked ChAT (KO) (16).

All mice were kept on a 12-h:12-h light-dark cycle with ad libitum access to food and water. Male and female mice (12–18 wk old) were used for all experiments. Animals from differing genotypes were age-matched and sex-matched for each experiment. All procedures were approved by Institutional Animal Care and Use Committees at the Oregon Health and Science University or the University of California-Davis and comply with the Guide for the Care and Use of Laboratory Animals, published by National Academies Press (8th edition). The experimental groups used were sham-operated animals and animals that underwent ischemia-reperfusion surgery, with tissue collection and heart mapping for both sham and MI groups occurring 14 days after surgery. A minimum of four animals were assigned to each group for each experiment, and tissue was processed together for each type of analysis. Experimenters were blinded throughout data collection and analysis.

Genotyping.

For DBH-CreERT2 samples, forward (TCAGAGATACCTGGCCTG) and reverse (CTGAAGGGTCTGGTAGGA) primers were used with a 58°C annealing temperature (21). For ChAT^lox/lox samples, a common forward primer (GCCCTGCCAGTCAACTCTA) was used in combination with two different reverse primers. To identify the WT allele, the reverse primer GAAATCCTGACAGATTCCAACA was used, and the product was 525 bp. To identify the mutant allele, the reverse primer TTTCCGCCTCAGGACTCTTC was used and the product was 400 bp. The annealing temperature was set at 60°C for both WT and mutant samples.

Myocardial ischemia-reperfusion.

The experimental protocol for ischemia-reperfusion followed current guidelines (11) and was carried out as previously described (2, 17). Adult mice were placed in an induction chamber and anesthetized with 4% isoflurane. Mice were intubated, mechanically ventilated, and maintained with 1–2% isoflurane mixed with 100% oxygen. Core body temperature was maintained at ~37°C, and a 2-lead ECG was monitored throughout the surgery using a PowerLab data acquisition system (AD Instruments, Colorado Springs, CO). A left lateral thoracotomy was performed in the fourth intercostal space and the pericardium was opened. The left anterior descending coronary artery (LAD) was reversibly ligated with an 8-0 suture for 45 min and then reperfused by release of the ligature. Occlusion was confirmed with ST segment elevation, regional cyanosis, and wall motion abnormalities. Reperfusion was confirmed by the return of color to the myocardium distal to the ligation and disappearance of ST elevation. The suture remained within the wound for the identification of the ligature site, and the chest and skin were closed in layers. After surgery, animals were returned to individual cages and given regular food and water until shipment to the University of California, Davis, or euthanasia and tissue harvest. Buprenorphine (0.1 mg/kg) was administered as needed to ensure the animals were comfortable following surgery. All surgical procedures were performed under aseptic conditions. Sham animals underwent the procedure described above except for the LAD ligation.

HPLC analysis of NE; mass spectrometry analysis of ACh.

NE levels were measured by HPLC with electrochemical detection as described previously (17), and ACh was quantified by mass spectrometry in the Oregon Health & Science University Bioanalytical Core facility as described previously (2, 16). Hearts were excised and cut in 2-mm transverse cross sections. The left ventricle below the site of LAD occlusion was dissected under a microscope to obtain viable peri-infarct tissue, which was frozen and stored at −80°C. Sham tissue underwent similar processing so that any ACh degradation by acetylcholinesterase during the dissection was comparable. Tissue samples from each heart were homogenized and neurotransmitters extracted at room temperature with 300 μL perchloric acid (0.1 M) containing 1.0 μM of the internal standard dihydroxybenzylamine to correct for NE sample recovery. Catecholamines were purified from 100 μL of the supernatant by alumina extraction before analysis by HPLC. Detection limits were ~0.05 pmol with recoveries from the alumina extraction > 60%. ACh was quantified in a second aliquot of 100 μL that was filtered at 4°C before analysis on the mass spectrometer.

Cardiac function assessment.

Mice were anesthetized with 1.5% isoflurane and imaged in the supine position using a Vevo 2100 Imaging System equipped with phased-array transducers with a frequency range of 18 to 55 MHz. Body temperature was maintained between 34 and 38°C, and a two-lead ECG was continuously recorded throughout the procedure. Images were obtained in the parasternal long-axis plane and parasternal short-axis planes at the midpapillary level. Measurements of left ventricle end-diastolic and end-systolic area (short axis) and end-diastolic and end-systolic length (long axis) were used to measure cardiac output as an estimate of left ventricle function.

Innervated whole‐heart preparation and Langendorff perfusion.

Isolated, innervated mouse hearts were prepared using a method described by Paton (18) with modifications (5, 27). Mice were administered an intraperitoneal injection of heparin (100 IU) and euthanized with pentobarbital sodium (150 mg/kg ip) and bisected subdiaphragmatically. The preparation was submerged in ice-cold cardioplegic solution, consisting of (in mmol/L) 110 NaCl, 1.2 CaCl₂, 16 KCl, 16 MgCl₂, and 10 NaHCO₃, and dissected from C1 to T12. The thoracic cavity was opened via incisions on both sides, and the anterior portion of the ribcage was removed. The descending aorta was cannulated with a 22-gage blunt needle. Following cannulation, the pericardium was cut and 5 mL of ice-cold cardioplegic solution was flushed through the preparation to remove blood. The preparation was then positioned supine in a glass-jacketed perfusion chamber.

After transfer of the preparation to the perfusion chamber, the heart and spinal cord were Langendorff perfused via the descending aorta with oxygenated (95% O₂-5% CO₂) modified Tyrode solution consisting of (in mmol/L) 128.2 NaCl, 1.3 CaCl₂, 4.7 KCl, 1.05 MgCl₂, 1.19 NaH₂PO₄, 20 NaHCO₃, and 11.1 glucose (pH 7.4 ± 0.05). For all experiments, pyridostigmine bromide (10 μM; Sigma-Aldrich, St. Louis, MO), a cholinesterase inhibitor, was added to the perfusate to reduce the enzymatic breakdown of ACh. The perfusate was pumped from a reservoir of 0.75 L through an in-line filter and two bubble traps before passing via the cannula to the preparation and then recirculated from the perfusion chamber back to the reservoir and regassed. Flow rate was adjusted at ~10 mL/min to maintain a perfusion pressure of 40–60 mmHg. Major blood vessels from the ribcage were ligated to limit perfusion leak and increase perfusion pressure. Two Ag/AgCl disk electrodes were positioned in the bath to record an ECG analogous to a lead I configuration. A customized bipolar pacing electrode was positioned on the apex of the right ventricular epicardium for cardiac pacing.

Dual optical mapping of transmembrane potential and intracellular Ca²⁺.

Dual optical mapping of transmembrane potential (V_m) and intracellular Ca²⁺ was performed as described previously (5, 27). Upon initiation of perfusion, Blebbistatin (Tocris Bioscience; 20 μM), an excitation-contraction uncoupler, was added to the perfusate to eliminate motion artifact during optical recordings. Hearts were then loaded with the acetoxymethyl ester form of the fluorescent intracellular Ca²⁺ indicator Rhod-2 (Rhod-2 AM; Invitrogen). An amount of 100 μL of 1 mg/mL of Rhod-2 AM in DMSO containing 10% Pluronic F127 was loaded through an in-line injection port over 1 min. Hearts were subsequently stained with the voltage-sensitive dye RH237 (15 μL of 1 mg/mL in DMSO). The anterior epicardial surface was excited using LED light sources centered at 531 nm and bandpass filtered from 511 to 551 nm (LEX-2; SciMedia). The emitted fluorescence was collected through a THT-macroscope (SciMedia) and split with a dichroic mirror at 630 nm. The longer wavelength moiety containing the V_m signal was long-pass filtered at 700 nm, and the shorter wavelength moiety containing the Ca²⁺ signal was bandpass filtered with a 32 nm filter centered at 590 nm. The emitted fluorescence signals were then recorded using 2 CMOS cameras (MiCam Ultima-L; SciMedia) with a sampling rate of 1 kHz and 100 × 100 pixels with a 10 × 10 mm field of view.

Optical mapping and sympathetic nerve stimulation protocol.

Following dye loading, baseline electrophysiological parameters were recorded during normal sinus rhythm as well as right ventricle epicardial pacing at cycle lengths of 150, 100, 90, and 80 ms. Hearts were then subjected to sympathetic nerve stimulation (SNS) in the absence and presence of the muscarinic receptor antagonist atropine (10 μM; Sigma-Aldrich). SNS was performed as previously described (5, 27). Briefly, a 2-Fr octapolar-pacing catheter (0.2-mm electrode, 0.5-mm spacing; CIBer Mouse-EP Catheter; NuMed, Hopkinton, NY) was inserted in the spinal canal at the 12th thoracic vertebra, and the tip was advanced to the level of the second thoracic vertebra to stimulate cardiac sympathetic outflow. The spinal cord was stimulated with a single-channel constant-current square-pulse stimulator (Grass Instruments) at 7.5 Hz and 7.5 V for 15–20 s, and then the stimulus was turned off to allow heart rate to return to baseline between recordings. SNS was also performed with simultaneous right ventricle cardiac pacing to collect rate-matched data.

Optical mapping data analysis.

Optical mapping data analysis was performed using two commercially available analysis programs (BV_Analyze, Brainvision, Tokyo, Japan; Optiq, Cairn, UK). V_m and Ca²⁺ data sets were spatially aligned and processed with a Gaussian spatial filter (radius 3 pixels). For both action potentials (APs) and Ca²⁺ transients (CaT), activation time was determined as the time at 50% of the maximal amplitude. For APs, repolarization time at 80% return to baseline was used to calculate action potential duration at 80% repolarization (APD₈₀). For CaT, repolarization (decay) time at 80% was used to calculate CaT duration at 80%. The spectral method, which has been used clinically for detecting microvolt T-wave alternans (25), was used to detect the presence and magnitude of CaT alternans and APD alternans (13). The spectral method was chosen because of its high sensitivity and relative immunity to noise. This approach allowed us to determine if an area within the mapping field of view was experiencing significant APD or CaT alternans (greater than the background noise levels) as well as the spatial extent of significant alternans. A spectral magnitude of ≥2 was used as the minimum threshold for significant APD or CaT alternans, corresponding to a beat-to-beat change in APD₈₀ ≥ 5 ms or beat-to-beat change in CaT amplitude ≥ 5%, respectively (28). The mapping field of view was subdivided into infarct zones (IZ), border zones (BZ), and remote zones (RZ) as follows: IZ was defined as V_m signals collected within 5 min of dye loading that had a signal-to-noise ratio < 5. BZ was defined as the area within 10 pixels bordering the IZ. RZ was defined as the remaining area and typically included most of the basal right and left ventricles. These selections were cross-referenced with color photographs of the heart to assure agreement between the selected areas and visible infarct tissue.

Statistics.

All values are presented as means ± SD. For most data, a two-way ANOVA followed by Tukey’s multiple comparisons test was used. If the two-way ANOVA revealed a significant main effect (e.g., difference in genotype regardless of treatment or difference in treatment regardless of genotype), the main effect is indicated on the plot or in the legend of the plot. When only two groups were compared (Fig. 1, A, C, and D, and Fig. 4, B and E), a Student’s t-test was used. A probability value of P < 0.05 was considered statistically significant.

Fig. 1.

Effects of choline acetyltransferase deletion from sympathetic neurons on cardiac function and norepinephrine (NE) and ACh content. In vivo heart rate in unoperated wild-type (WT; n = 11) and iChAT knockout (KO; n = 14) mice (A). Cardiac output 14 days after sham (n = 4/group) or myocardial infarction (MI; n = 5–6) surgery (B). Left ventricular ACh (C) and NE (D) content 14 days after MI (n = 4–7). Data are means ± SD; *P < 0.05, ***P < 0.001.

Fig. 4.

Heterogeneity of action potential duration (APD) in myocardial infarction (MI) hearts at baseline and following application of atropine. A: choline acetyltransferase (ChAT) deletion led to shorter APD at 80% repolarization (APD₈₀) in the remote zone (RZ) and border zone (BZ) of knockout (KO) MI hearts compared with the matched location in wild-type (WT) MI hearts. APD₈₀ in the RZ of KO MI hearts was also significantly shorter than the infarct zone (IZ) of KO MI hearts. B: APD dispersion (measured as inner 90th percentile of APD₈₀: IP₉₀) was increased in KO MI vs. WT MI hearts. C: example APD₈₀ maps illustrating increased APD dispersion and shorter RZ and BZ APD with ChAT deletion. D–F: APDs as in A–C following application of atropine (ATR), which eliminated differences between WT and KO groups. APD₈₀ in the RZ of WT MI hearts with ATR is significantly shorter than the IZ, similar to KO hearts. Data are means ± SD; WT, n = 5; KO, n = 8; *P < 0.05, **P < 0.01, ***P < 0.001. All data were collected with right ventricular pacing from apex at a cycle length of 100 ms.

RESULTS

Effects of ChAT KO on cardiac function, NE, and ACh content.

Basal in vivo heart rates before surgery were not different between WT and KO mice (Fig. 1A). Following sham or MI surgery, echocardiography demonstrated impaired cardiac output in the MI mice of both genotypes (Fig. 1B); however, no differences in cardiac function were apparent between genotypes. As expected, WT MI hearts had increased ventricular ACh levels because of sympathetic transdifferentiation, which is prevented in the KO hearts following MI (Fig. 1C). Ventricular NE levels were not different between genotypes of MI hearts (Fig. 1D).

Dual optical mapping of V_m and intracellular Ca²⁺ in ex vivo innervated hearts.

To investigate the spatial and temporal impact of physiological sympathetic nerve stimulation (SNS) on the whole heart, we developed an ex vivo innervated heart model (5) where sympathetic nerves were intact and functional upon stimulation (15). A photograph of the ex vivo innervated Langendorff-perfused mouse heart is shown in Fig. 2A. Simultaneous optical mapping of V_m and intracellular Ca²⁺ was performed using voltage-sensitive and Ca²⁺-sensitive dyes (RH237 and Rhod-2 AM) with example V_m and intracellular Ca²⁺ activation maps and optical AP and CaT traces shown in Fig. 2, B and C. Upon activation of SNS, heart rate and CaT amplitude increased (Fig. 2, C–E; note that CaT amplitude is normalized to baseline conditions as signals are uncalibrated). Sham and MI hearts from both WT and KO groups showed similar increases in heart rate during SNS (Fig. 2, D and E). Baseline ex vivo heart rate was slightly elevated in the KO versus WT MI hearts (Fig. 2E), consistent with lower ACh levels in the KO MI hearts (Fig. 1C).

Fig. 2.

Impact of sympathetic nerve stimulation (SNS) on action potentials (APs), Ca²⁺ transients (CaTs), and heart rate in the isolated innervated heart. Photograph of innervated heart preparation for optical mapping with approximate infarct location indicated by dashed line (A). Example maps of transmembrane potential (V_m) and Ca²⁺ activation (B). Example optical APs and CaTs showing increased frequency and CaT amplitude (when normalized to baseline CaT as signals are uncalibrated) with SNS (C). Heart rate at baseline and with SNS in sham (D) and myocardial infarction (MI) (E) hearts. SNS significantly elevates heart rate in all groups. Data are means ± SD; sham, n = 4/group; MI wild type (WT), n = 5; MI knockout (KO), n = 8; *P < 0.05, **P < 0.01, ***P < 0.001. All data were collected during sinus rhythm. LV, left ventricle; RV, right ventricle.

Effects of physiological SNS on APD in sham and MI hearts.

Shortening of APD upon β-adrenergic receptor activation has been consistently reported in larger mammals (9, 22, 23, 26, 30). In the rodent heart, however, a range of responses has been observed with β-adrenergic agonists, including no change, shortening, or prolongation of APD (3, 7, 10, 29). In our previous studies in the innervated mouse heart (5, 27), we found that APD was significantly prolonged with physiological SNS. Consistent with this, we found that when the heart rate was held constant with ventricular pacing at a cycle length of 100 ms, SNS tended to prolong APD in sham and MI hearts from both genotypes, but this prolongation was not statistically significant (Fig. 3, A and B). Following MI, the KO hearts had significantly shorter APD₈₀ and APD at 50% repolarization compared with WT hearts (Fig. 3, B and D). Although atropine, a muscarinic blocker, did not significantly change APD compared with baseline in either group, it did eliminate genotype APD differences in MI hearts (Fig. 3C), suggesting that APD differences may be due to sympathetic ACh production and release in WT hearts following MI. SNS in the presence of atropine significantly prolonged APD in both genotypes (Fig. 3C).

Fig. 3.

Action potential (AP) duration (APD) in sham and myocardial infarction (MI) hearts at baseline and with sympathetic nerve stimulation (SNS) during pacing at a cycle length of 100 ms. Mean APD at 80% repolarization (APD₈₀) from entire mapping field of view at baseline and with SNS in sham (A), MI (B), and MI hearts treated with atropine (ATR) (C). APD₈₀ was significantly shorter in MI knockout (KO) vs. MI wild-type (WT) hearts (**P < 0.01, main effect WT vs. KO, 2-way ANOVA) (B), and this effect was no longer significant in the presence of ATR (C). SNS significantly increased APD₈₀ in the presence of ATR (*P < 0.05, main effect ATR vs. ATR + SNS, 2-way ANOVA) (C). Mean APD at 50% repolarization from MI hearts at baseline and with SNS (*P < 0.05; **P < 0.01, main effect WT vs. KO, 2-way ANOVA) (D). Example maps of transmembrane potential (V_m) and Ca²⁺ activation (*earliest activation site where placing electrode was placed), APD₈₀, and Ca²⁺ transient duration (CaTD) from sham and MI hearts (E). Example optical action potentials from MI hearts at baseline and with SNS (F). Data are means ± SD; sham, n = 4/group; MI WT, n = 5; MI KO, n = 6–8. All data were collected with right ventricular pacing from apex at a cycle length of 100 ms. LV, left ventricle; RV, right ventricle.

Spatial dispersion of APD in MI hearts.

One of the advantages of cardiac optical mapping is the ability to obtain electrophysiological data with high spatial resolution, which allows for comparison of regional differences. When the mapping field of view was divided into IZ, BZ, and RZ (Fig. 4, C and F), the APD of WT hearts was significantly longer in the RZ and BZ compared with matched regions of KO hearts (Fig. 4, A and C), resulting in significantly increased APD dispersion in KO hearts (Fig. 4B). Atropine eliminated the regional APD differences between WT and KO hearts, resulting in an increase in APD dispersion in WT hearts, similar to KO hearts (Fig. 4, D–F). These data suggest that sympathetic nerve-released ACh in the remote and border zones of WT hearts may act to reduce post-MI APD dispersion.

Effects of SNS on intracellular Ca²⁺ handling in MI hearts.

Under baseline conditions, intracellular Ca²⁺ handling was similar between genotypes, (Fig. 5A), and this is consistent with echocardiographic data, which demonstrated similar contractility in the MI hearts from both genotypes (Fig. 1B). As expected, SNS shortened CaT duration (CaTD) significantly in both genotypes (Fig. 5, A and B). Further examination of regional CaTD showed no differences at baseline (Fig. 5C). Note that there was a trend for shorter CaTD in the RZ from both WT and KO hearts, which became statistically significant with SNS in both genotypes (Fig. 5, C and D).

Fig. 5.

Ca²⁺ transient duration (CaTD) in myocardial infarction (MI) hearts. A: mean CaTD from the entire mapping field of view at baseline and with sympathetic nerve stimulation (SNS) in MI hearts showing no differences between groups, but SNS significantly shortens CaTD at 80% (CaTD₈₀) compared with baseline (*P < 0.01; main effect baseline vs. SNS; 2-way ANOVA). B: example CaTD maps at baseline and with SNS. C: at baseline, there are no regional CaTD₈₀ differences between groups or between regions. D: with SNS, CaTD₈₀ in the remote zone (RZ) is significantly shorter than the infarct zone (IZ) in both groups. Data are means ± SD; wild type (WT), n = 5; knockout (KO), n = 8; **P < 0.01. All data were collected with right ventricular pacing from apex at a cycle length of 100 ms. BZ, border zone.

When hearts were paced at a cycle length of 90 ms (667 beats/min), CaT alternans, defined as beat-to-beat variation in the amplitude of the intracellular CaT, developed in the MI hearts of both genotypes (Fig. 6A). CaT alternans has been shown to underlie clinically observed T-wave alternans, a marker of increased electrical instability and ventricular fibrillation (14). WT and KO hearts showed similar susceptibility to CaT alternans throughout the heart at baseline (Fig. 6A). SNS tended to reduce CaT alternans magnitude in the RZ and BZ of both groups (Fig. 6B), but there were no statistical differences between groups or regions.

Fig. 6.

Ca²⁺ transient (CaT) alternans (Alt) in myocardial infarction (MI) hearts. There are no regional differences in CaT alternans magnitude between wild-type (WT) and knockout (KO) MI hearts at baseline (A) or with sympathetic nerve stimulation (SNS) (B). Example optical maps of alternans magnitude (C) and representative optical CaTs (D). Data are means ± SD; WT, n = 5; KO, n = 7. All data were collected with right ventricular pacing from apex at a cycle length of 90 ms. BZ, border zone; RZ, remote zone; IZ, infarct zone; A.U., arbitrary units.

DISCUSSION

Neuronal control of cardiac function undergoes significant remodeling with cardiovascular disease and consequently contributes to cardiac dysfunction and sudden cardiac death (1, 6, 12, 20, 24). Previously, we found that cardiac noradrenergic neurons undergo cholinergic transdifferentiation following MI, resulting in corelease of ACh and NE (16). Here, we sought to investigate the functional electrophysiological consequences of sympathetic neuronal transdifferentiation. Using an inducible KO mouse model in which ChAT was deleted from adult sympathetic neurons, combined with a novel innervated Langendorff-perfused heart preparation, we found that corelease of ACh may have antiarrhythmic effects by decreasing APD dispersion following MI.

In the normal heart, cardiac sympathetic neurons synthesize and release NE, which plays a key role in positive chronotropy, inotropy, and lusitropy during the fight-or-flight response. These cardiac responses are primarily mediated by NE binding to β1-adrenergic receptors, activation of the stimulatory G protein, G_αs, and subsequent generation of cAMP. Parasympathetic nerves, on the other hand, primarily release ACh, which binds to cardiac M2 muscarinic receptors, and activates the inhibitory G protein, G_αi, to oppose the effects of NE. M2 stimulation also results in an increase in I_KACh via G_βγ, which additionally contributes to slowing of heart rate upon parasympathetic stimulation because of membrane hyperpolarization. We therefore hypothesized that sympathetic corelease of ACh following MI would oppose the electrophysiological effects of NE both at rest (i.e., with basal levels of nerve activity and catechol release) and during elevated sympathetic activity via nerve stimulation. Our results are consistent with this hypothesis, and we found that corelease of ACh may have antiarrhythmic effects in the post-MI heart via reduced APD dispersion.

As expected, deletion of ChAT from sympathetic neurons reduced ventricular ACh levels without impacting ventricular NE content (Fig. 1, C and D). We previously reported that ventricular ACh levels rise transiently (between 7 and 14 days post-MI) in WT MI but not KO MI or sham-operated hearts from either genotype (16). Additionally, we previously showed that ACh levels only rise in the remote and border zones of WT MI hearts because the infarct remains devoid of sympathetic innervation (16). Likewise, cell bodies in the stellates of WT MI hearts stain positive for ChAT, suggesting that ACh is released from sympathetic fibers throughout the noninfarct ventricle as well as from atrial sympathetic neurons. Consistent with this, WT MI hearts had lower ex vivo heart rates at baseline compared with KO hearts (Fig. 2E), suggesting increased basal ACh release in the atria. Although cardiac output was decreased following MI when compared with sham-operated hearts, corelease of ACh in WT hearts did not further impair ventricular function compared with KO MI hearts (Fig. 1B). This might be due to the relatively small amount of ACh in the left ventricle, which was an order of magnitude lower than NE content (Fig. 1). Alternatively, contractile deficits at this post-MI time point (14 days) may be primarily due to structural remodeling rather than Ca²⁺ mishandling. Indeed, intracellular Ca²⁺ handling was not different between genotypes either with or without SNS (Figs. 5 and 6).

When APDs were assessed from the entire mapping field of view (anterior ventricles), there were no significant differences between genotypes for sham hearts, and both groups responded similarly to SNS with slight APD prolongation (however, this was not statistically significant; Fig. 3, A and B). The KO MI hearts, however, had slightly shorter APD₈₀ and APD at 50% repolarization (Fig. 3, B and D), and APD₈₀ was no longer different following atropine application (Fig. 3C), suggesting that release of ACh and subsequent activation of muscarinic receptors may underlie these APD differences. Regional assessment of APD in the MI hearts revealed that APD was significantly shorter in the remote and border zones of KO versus WT hearts, whereas APDs of the infarct zone were similar between genotypes (Fig. 4A). This result is likely a consequence of nonuniform sympathetic nerve distribution following MI. Indeed, we have previously demonstrated that sympathetic nerve density remains low in reperfused infarcts, with normal nerve density in the remote and border zones (7). Thus, sympathetic nerve-released ACh may be most prevalent in the noninfarct regions, where nerve density is greatest. The underlying APD differences between remote and infarct regions of the KO MI hearts led to an increase in APD dispersion (Fig. 4, B–C). Interestingly, atropine rendered APD of the WT hearts more similar to KO hearts by slightly decreasing APD in the remote and border zones, which resulted in significantly different APDs in the RZ versus IZ of WT hearts (similar to KO hearts) and similar APD dispersion between genotypes (Fig. 4, D–F). These intriguing results suggest that corelease of ACh in the remote and border zones following MI may blunt some of the effects of basal nerve-released NE in these areas, leading to more homogeneous APD between infarct and remote regions. In this regard, sympathetic nerve transdifferentiation may be somewhat antiarrhythmic.

Despite regional differences in APD between genotypes, we did not observe regional differences in CaTD either with or without SNS (Fig. 5). It is unclear why nerve-released ACh in the remote and border zones did not impact CaTD in the WT compared with KO hearts. It is possible that the increased ACh in WT hearts following MI has a more significant impact on transmembrane potential versus intracellular Ca²⁺ handling. Likewise, Ca²⁺ alternans magnitude was similar between genotypes both with and without SNS (Fig. 6). SNS tended to reduce Ca²⁺ alternans magnitude most prominently in the RZ of both groups (Fig. 6B), but there were no significant differences between groups or regions.

Limitations.

Here, we utilized a novel ex vivo innervated heart model to study cardiac electrophysiology with intact sympathetic innervation. The ex vivo innervated heart (27), similar to the traditional Langendorff-perfused heart, has substantially lower baseline heart rate (HR) versus in vivo. Similar maximal HRs were achieved in all groups in response to the SNS stimulation parameters used in this study. Varying strengths of SNS may have revealed differential HR responses. Because of the small size of the mouse heart and excellent metabolic state of mechanically uncoupled perfused heart preparations, inducible and spontaneous arrhythmias are rare (4). We did not observe any spontaneous arrhythmias or any sustained arrhythmias in response to pacing or SNS in any hearts in the present study. Therefore, the definitive effect of transdifferentiation on arrhythmogenicity remains an important area for future work. We previously showed that sympathetic transdifferentiation resolves by 21 days post-MI (16); however, we have not examined whether preventing transdifferentiation has a long-term impact on post-MI remodeling or arrhythmia risk. Differences in muscarinic receptor expression or activity were not evaluated, and a full characterization of parasympathetic nerve remodeling and muscarinic responsiveness in the post-MI heart is a critical area for future work. A full range of pacing frequencies was not tested; therefore, the frequency-dependence of Ca²⁺ alternans and alternans thresholds were not examined.

Conclusions.

This study is the first to report on the specific electrophysiological effects of sympathetic cholinergic transdifferentiation following MI. Our data indicate that elevated ACh in WT MI hearts reduced basal heart rate and reduced APD dispersion, the latter of which may be antiarrhythmic by decreasing dispersion of repolarization and preventing unidirectional conduction block following MI. These novel roles for sympathetic nerve-released ACh should be considered when assessing and developing post-MI therapeutic approaches that aim to target nerve activity or transdifferentiation.

GRANTS

This work was supported by the National Institutes of Health Grants HL-093056 (to B. A. Habecker) and HL-111600 (to C. M. Ripplinger).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

L.W., B.A.H., and C.M.R. conceived and designed research; L.W., A.O., S.D.F.S., S.T., M.R.B., W.R.W., B.A.H., and C.M.R. performed experiments; L.W., A.O., S.D.F.S., S.T., M.R.B., W.R.W., B.A.H., and C.M.R. analyzed data; L.W., S.D.F.S., S.T., M.R.B., W.R.W., B.A.H., and C.M.R. interpreted results of experiments; L.W., A.O., M.R.B., W.R.W., B.A.H., and C.M.R. prepared figures; L.W., B.A.H., and C.M.R. drafted manuscript; L.W., A.O., S.D.F.S., S.T., M.R.B., W.R.W., B.A.H., and C.M.R. edited and revised manuscript; L.W., A.O., S.D.F.S., S.T., M.R.B., W.R.W., B.A.H., and C.M.R. approved final version of manuscript.