Acute isoproterenol leads to age-dependent arrhythmogenesis in guinea pigs

Abstract

Sudden cardiac death from ventricular arrhythmias is more common in adult patients with with heart failure compared with pediatric patients with heart failure. We identified age-specific differences in arrhythmogenesis using a guinea pig model of acute β-adrenergic stimulation. Young and adult guinea pigs were exposed to the β-adrenergic agonist isoproterenol (ISO; 0.7 mg/kg) for 30 min in the absence or presence of flecainide (20 mg/kg), an antiarrhythmic that blocks Na⁺ and ryanodine channels. Implanted cardiac monitors (Reveal LINQ, Medtronic) were used to monitor heart rhythm. Alterations in phosphorylation and oxidation of ryanodine receptor 2 (RyR2) were measured in left ventricular tissue. There were age-specific differences in arrhythmogenesis and sudden death associated with acute β-adrenergic stimulation in guinea pigs. Young and adult guinea pigs developed arrhythmias in response to ISO; however, adult animals developed significantly more premature ventricular contractions and experienced higher arrhythmia-related mortality than young guinea pigs treated with ISO. Although there were no significant differences in the phosphorylation of left ventricular RyR2 between young and adult guinea pigs, adult guinea pigs exposed to acute ISO had significantly more oxidation of RyR2. Flecainide treatment significantly improved survival and decreased the number of premature ventricular contractions in young and adult animals in association with lower RyR2 oxidation. Adult guinea pigs had a greater propensity to develop arrhythmias and suffer sudden death than young guinea pigs when acutely exposed to ISO. This was associated with higher oxidation of RyR2. The incidence of sudden death can be rescued with flecainide treatment, which decreases RyR2 oxidation.

NEW & NOTEWORTHY Clinically, adult patients with heart failure are more likely to develop arrhythmias and sudden death than pediatric patients with heart failure. In the present study, older guinea pigs also showed a greater propensity to arrhythmias and sudden death than young guinea pigs when acutely exposed to isoproterenol. Although there are well-described age-related cardiac structural changes that predispose patients to arrhythmogenesis, the present data suggest contributions from dynamic changes in cellular signaling also play an important role in arrhythmogenesis.

INTRODUCTION

Although pediatric heart failure (HF) is less prevalent than adult HF, it is associated with high mortality and high costs of care. The most common cause of noncongenital HF in children is idiopathic dilated cardiomyopathy (IDC) (65). In adults, IDC is one of the most common nonischemic causes of HF (36). Importantly, therapies that improve outcomes in adult patients with IDC fail to provide such benefits in pediatric patients with IDC (54, 55, 65). We have previously shown fundamental differences in the myocellular mechanisms involved in cardiac pathology between children and adults with IDC, which we propose lead to differences in response to therapies (7, 38, 42, 63, 71). Further investigation of these differences will improve understanding of both patient populations and ultimately lead to identification of age-specific targeted treatments that could improve clinical outcomes.

One clinically apparent age difference in patients with HF is the incidence of ventricular arrhythmias and sudden death. For example, primary prevention implantable defibrillators are the standard of care in adults with systolic HF for the prevention of sudden cardiac death, whereas pediatric patients with IDC show a much lower incidence of arrhythmias and sudden death and rarely require implantable defibrillators (13, 49, 53). In addition, adult patients with IDC demonstrate an increased incidence of arrhythmias and sudden death when treated with phosphodiesterase 3 inhibitors, whereas phosphodiesterase 3 inhibitors are well tolerated by pediatric patients with HF, and this therapy is often used as a bridge to transplantation (12, 15, 16, 20, 44, 47, 48). It has been proposed that the adult heart is more prone to arrhythmias because of the development of fibrotic regions, whereas fibrosis in the pediatric failing heart is less prominent (57, 71). We hypothesize that age-dependent differences in adaptations to HF, including alterations in proteins involved in Ca²⁺ handling, contribute to differences in propensities to develop arrhythmias between the two populations.

Posttranslational modifications of Ca²⁺-handling proteins occur in response to changes in circulating catecholamines. Catecholamine excess is present in HF regardless of age, resulting in stimulation of the β-adrenergic receptor (β-AR), leading to phosphorylation of downstream targets through PKA or Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) (18, 35, 41, 52, 62, 74). Furthermore, β-AR stimulation leads to increased production of ROS, which can also lead to posttranslational modifications of Ca²⁺-handling proteins through oxidation (5). In particular, phosphorylation and oxidation of the sarcoplasmic reticulum Ca²⁺ release channel, ryanodine receptor 2 (RyR2), can cause Ca²⁺ leak, leading to delayed afterdepolarizations and Ca²⁺ waves that propagate between cardiomyocytes. These alterations have the potential to trigger aberrant contractions (4, 9, 14, 31, 39, 45, 66, 69) or premature ventricular contractions (PVCs). PVCs that occur during a vulnerable repolarization period in the heart can lead to ventricular tachyarrhythmias, inadequate perfusion, and death (34, 43).

In the present study, we used guinea pigs to investigate age-related differences in catecholamine-induced ventricular arrhythmias. We hypothesized that acute exposure to the β-AR agonist isoproterenol (ISO) would result in arrhythmogenesis and RyR2 changes in an age-dependent fashion. We found that ISO treatment resulted in the occurrence of ventricular arrhythmias and oxidation of RyR2 that was improved with flecainide treatment. Importantly, the arrhythmia burden and oxidation of RyR2 in response to ISO was higher in adult animals compared with young animals.

MATERIALS AND METHODS

Animal Protocols

Young (24 days old) male and female Dunkin-Hartley guinea pigs (average weight: 260 g, n = 71) and adult (over 6 mo of age) male and female Dunkin-Hartley guinea pigs (average weight: 981.2 g, n = 35, Charles River Laboratories, Kingston, NY) were housed at the University of Colorado Anschutz Medical campus. Briefly, animals were anesthetized with isoflurane and received an intraperitoneal injection of saline (0.9% NaCl), flecainide (20 mg/kg, Sigma-Aldrich, St. Louis, MO), or quinidine (37.5 mg/kg, Sigma-Aldrich) before incision. Flecainide is a Vaughan Williams class Ic agent with secondary RyR2 inhibitory effects used clinically to treat catecholaminergic polymorphic ventricular tachycardia (CPVT), whereas quinidine is a class Ia agent acting on Na⁺ channels without RyR2 blockade. These inhibitors were used to differentiate the effects of functional RyR2 inhibition from Na⁺ channel effects in the guinea pig model (21, 67, 68). A subcutaneous incision was made between the scapulae, 1 in. behind the head. Blunt dissection through the incision created a subcutaneous pocket on each side of the animal. REVEAL LINQ Implantable Cardiac Monitors (Medtronic, Minneapolis, MN) were placed into the pocket on the animal’s upper left side. Osmotic pumps (2ML1, 2ML2, and 2ML4, Alzet, Cupertino, CA) containing either saline (control) or ISO (16 or 32 mg·kg⁻¹·day⁻¹, Sigma-Aldrich) were implanted on the upper right side as previously described (58, 60). The incisions were closed with staples. Animals were either monitored for a full hour after surgery or euthanized with Fatal Plus [phenobarbital (100 mg/kg), MWI Veterinary Supply, Boise, ID] 30 min after the surgery was completed. For survival analyses, animals who received either 16 or 32 mg·kg⁻¹·day⁻¹ ISO were grouped, as there were no differences in survival or collapse between these two dosages. All other assessments and experiments were conducted on animals that received 32 mg·kg⁻¹·day⁻¹ ISO. Heart rhythm was recorded from 4−10 and 24–30 min after surgery. Chest compressions (estimated 144 compressions/min) were performed on any animal that collapsed until visible evidence of return of spontaneous circulation or for 10 min of total resuscitation time. Rhythm was collected on the LINQ device specifically during this time to capture initiation of the rhythm disturbance in addition to the prespecified recordings outlined above. Hearts were excised immediately after euthanasia and perfused with saline, and the right ventricle and left ventricle were then separated and flash frozen at −80°C.

All animal experiments were approved by University of Colorado Institutional Animal Care and Use Committee and were in accordance with National Institutes of Health guidelines.

Analysis of LINQ Tracings

ECG recordings from REVEAL LINQ Implantable Cardiac Monitors (Medtronic) were obtained 4–10 min and 24–30 min after pump implantations. Recordings were converted to .csv files using a proprietary script created by Medtronic. ECG parameters, including P-R interval, duration of QRS, Q-T interval, interval from the peak of the T wave to the end of the T wave (T_p-e), R-R interval, corrected QT interval (QT_c), and Q-T_apex, were manually calculated using an average of 5–10 beats derived from the .csv files. T_p-e is a measurement of transmural dispersion of repolarization. Further analysis was done to calculate heart rate (HR) based on R-R interval, and the T_p-e-to-QT_c ratio (a marker of arrhythmogenesis) was calculated (19, 29, 50). Numbers of PVCs were counted in tracings recorded 4–10 min and 24–30 min after surgery.

Protein Isolation

RyR2 protein was isolated as previously described with modifications (6). Briefly, 100 mg of frozen left ventricular tissue were homogenized using a bead homogenizer in 20 mM HEPES, 250 mM sucrose, and protease and phosphatase inhibitors. The homogenates were centrifuged at 4°C at 1,500 × g for 15 min, and the supernatant was transferred to ultracentrifuge tubes. The pellet was resuspended in buffer and centrifuged again at 4°C for 15 min at 1,500 × g, and this supernatant was added to the first supernatant. The combined supernatants were centrifuged at 110,000 × g for 60 min at 4°C, and the resulting pellets were resuspended using 20 mM HEPES, 250 mM sucrose buffer with inhibitors, and 2% N-dodecyl-β-d-maltoside (DDM; Sigma-Aldrich). Protein concentration was determined using the DC assay (Bio-Rad, Hercules, CA).

Protein was isolated from 10–20 mg of frozen left ventricular tissue in isoelectric focusing buffer (8 M urea, 2.5 M thiourea, 4% CHAPS, 2 mM EDTA, 1 mM DTT, and 1% TBP) with protease and phosphatase inhibitors and homogenized at 4°C as previously described (60). Protein concentration was determined using the Bradford assay (Bio-Rad).

Western Blot Analysis

Western blots were performed as previously described with the following modifications (61).

RyR2.

Sarcoplasmic reticulum-enriched proteins were isolated as described above and separated using a 4–20% Criterion gel (Bio-Rad) for 4 h at 130 V. Proteins were transferred to 0.45 μM PVDF membranes in transfer buffer (National Diagnostics, Atlanta, GA) + 0.01% SDS at 30 V for 16 h and 80 V for 1 h on ice. Blots were blocked with 5% BSA in Tris-buffered saline plus 0.1% Tween (Bio-Rad). Dephosphorylated Ser²⁰³⁰ (1:1,000, a gift from Dr. Valdivia), Ser²⁸⁰⁸ (1:8,000, Badrilla, Leeds, UK), Ser²⁸¹⁴ (1:3,000, Badrilla), and total RyR2 (1:2,000, a gift from Dr. Marks) were incubated on separate blots at 4°C overnight, and the appropriate secondary antibodies were applied for 1 h at room temperature (goat anti-rabbit, 1:25,000, Jackson Immunology Research Laboratories, West Grove, PA; and sheep anti-mouse, 1:25,000, GE Healthcare Life Sciences, Pittsburg, PA).

Phospholamban.

Proteins isolated by isoelectric focusing buffer were separated using 15% Criterion gels (Bio-Rad) at 130 V for 2 h and transferred onto 0.2 μM PVDF membranes (EMD Millipore, Billerica, MA) at 80 V for 80 min. Blots were blocked with 10% milk in Tris-buffered saline plus 0.1% Tween. Ser¹⁶ (Badrilla), Thr¹⁷ (Badrilla), and total phospholamban (PLN; EMD Millipore) were incubated on separate membranes at 4°C overnight, and the appropriate secondary antibodies were applied for 1 h (goat anti-rabbit, 1:50,000, Jackson Immunology Research Laboratories; and sheep anti-mouse, 1:25,000, GE Healthcare Life Sciences).

Oxyblot.

Carbonyl groups on proteins from the sarcoplasmic reticulum-enriched protein fraction were derivatized to 2,4-dinitrophenylhydrazone (DNP) by treatment with 2,4-dinitrophenylhydrazine as instructed in the OxyBlot Protein Oxidation Detection kit (EMD Millipore). As a negative control, protein samples from the same animal were exposed to a control buffer lacking the enzyme. Proteins were then separated on a 4–20% Criterion gel (Bio-Rad) and treated as described above for the RyR2 Western blots. Blots were incubated with anti-DNP (1:150, EMD Millipore) for 1 h at room temperature followed by anti-rabbit secondary (1:300, EMD Millipore), and membranes were exposed as described above. RyR2 protein was identified based on molecular weight and confirmed by incubation with total RyR2 antibody. Oxidized RyR2 was normalized to total RyR2.

All blots were incubated with ECL reagent (West Pico or West Femto, ThermoFisher Scientific, Pittsburg, PA) and exposed to autoradiography film. Proteins were normalized to GAPDH, and ratios of the phosphorylated protein to total protein were calculated (except for oxyblot samples, total oxidized RyR2 was normalized to total RyR2, Santa Cruz Biotechnology, Dallas, TX) and quantified using ImageJ (version 1.49v).

Data Analysis and Statistics

Statistical analyses were performed using Prism Version 6 (GraphPad Software, San Diego, CA). Consistent with American Physiological Society recommendations on studies that include higher-level mammals and the goal of identifying putative mechanisms that could lead to further studies, statistical significance was set a priori at P < 0.1, and all data are presented as means ± SD (8). Data were tested for normality, and data that were non-normally distributed were log transformed before statistical testing. Mortality data were assessed using the Mantel-Cox log-rank test. Comparison of three or more groups was conducted using one-way ANOVA, and Bonferroni’s post hoc tests were performed. Since transformation of the PVC data was not possible with the zero counts, PVC data were analyzed using a Kruskal-Wallis test with Dunn’s multiple-comparisons test.

RESULTS

Adult Guinea Pigs Are More Susceptible to Sudden Death With acute ISO Exposure Than Young Guinea Pigs

Acute exposure to either concentration of ISO (16 or 32 mg·kg⁻¹·day⁻¹) resulted in guinea pigs collapsing. Overall, 50–60% of guinea pigs (11 of 18 adult guinea pigs and 37 of 71 young guinea pigs) collapsed within 1 h of completion of the surgery (Fig. 1A). Although chest compressions rescued 68% of young animals (25 of 37 guinea pigs), none of the adult animals (0 of 11 guinea pigs) survived (Fig. 1B). As a result, mortality was significantly higher in adult guinea pigs within the first hour of acute ISO exposure when compared with young guinea pigs (Fig. 1C).

Fig. 1.

Age-specific responses to acute isoproterenol (ISO) exposure. A: percentage of guinea pigs that collapsed in the first hour of ISO exposure; 37 of 71 young guinea pigs collapsed (52%) and 11 of 18 adult guinea pigs collapsed (61%). B: percent survival in guinea pigs that collapsed when exposed to ISO. Sixty-eight percent of young guinea pigs (25 of 37 guinea pigs) recovered after collapse, whereas none of the adult guinea pigs (0 of 11 guinea pigs) recovered after collapse. Significantly more adult ISO-treated guinea pigs died after collapse than did young ISO-treated guinea pigs by Mantel-Cox log-rank test (P < 0.0001). C: percent survival in guinea pigs in the first hour of ISO exposure; 59 of 71 young guinea pigs (83%) survived, whereas 7 of 18 adult guinea pigs (39%) survived. Mortality in adult guinea pigs was statistically significant by Mantel-Cox log-rank test (P = 0.04). Young control group: n = 36, young ISO-treated group: n = 71, adult control group: n = 13, and adult ISO-treated group: n = 18.

ISO Treatment Increased PVCs and Repolarization Abnormalities to a Greater Extent in Adult Guinea Pigs

Heart rhythm data from LINQ recorders demonstrated increased PVCs in response to acute exposure to ISO (32 mg·kg⁻¹·day⁻¹). When these PVCs occurred during ventricular repolarization, animals developed ventricular fibrillation and collapsed (Fig. 2A). Importantly, all episodes of collapse captured on LINQ recorders were associated with ventricular tachycardia/ventricular fibrillation, and no ventricular tachycardia/ventricular fibrillation was documented in animals that did not collapse. PVC number was significantly higher in ISO-treated animals compared with control animals regardless of age (Fig. 2B). However, adult guinea pigs treated with ISO developed significantly more PVCs than young guinea pigs treated with ISO. Young animals exposed to ISO had significantly higher HRs than young control animals and adult animals exposed to ISO (Fig. 2C). There were no differences in QRS duration, QT_c, or P-R interval between ISO-treated young and adult animals (data not shown). Adult guinea pigs exposed to ISO had prolonged T_p-e compared with age-matched control guinea pigs (Fig. 2D) and young ISO-treated guinea pigs. The T_p-e-to-QT_c ratio was also significantly higher in adult guinea pigs treated with ISO compared with age-matched control guinea pigs (Fig. 2E) and young ISO-treated guinea pigs. There was no difference in T_p-e or the T_p-e-to-QT_c ratio between control and ISO-treated young animals.

Fig. 2.

Arrhythmogenic changes in guinea pigs upon acute isoproterenol (ISO) exposure. A: representative images of cardiac rhythms from young and adult guinea pigs exposed to ISO demonstrating premature ventricular contractions (PVCs) leading to polymorphic ventricular tachycardia/ventricular fibrillation. These episodes were associated with collapse of the animal and sudden death. B: quantification of PVCs in young and adult guinea pigs for 14 min of recording within 30 min of pump implantation. Both young and adult guinea pigs had significantly more PVCs than age-matched control guinea pigs when exposed to ISO (young: P = 0.002 and adult: P < 0.01). Adult guinea pigs treated with ISO had more PVCs than young guinea pigs treated with ISO (P < 0.01). Statistical analysis of data was by Kruskal-Wallis test with Dunn’s multiple comparisons. Young control group: n = 15, young ISO-treated group: n = 29, adult control group: n = 9, and adult ISO-treated group: n = 7. C: heart rate (HR) averaged over first 10 min after pump implantation. Within the first 10 min of ISO exposure, young guinea pigs had a significantly higher HR (P = 0.03). There was a significant difference between young and adult guinea pig HRs with exposure to ISO (P = 0.003). Statistical analysis of transformed data was done using one-way ANOVA with Bonferroni’s multiple-comparisons analysis. Young control group: n = 6, young ISO-treated group: n = 5, adult control group: n = 5, and adult ISO-treated group: n = 6. D: ISO exposure for 10 min significantly increased the time from the peak to the end of the T wave (T_p-e) in adult guinea pigs (P = 0.03). Adult guinea pigs exposed to ISO had longer duration of T_p-e than young guinea pigs treated with ISO (P = 0.06). Statistical analysis of transformed data was done using one-way ANOVA with Bonferroni’s multiple-comparisons analysis. Young control group: n = 6, young ISO-treated group: n = 5, adult control group: n = 5, and adult ISO-treated group: n = 6. E: time from T_p-e over corrected QT interval (T_p-e/QT_c) increased in adult guinea pigs treated with ISO (P = 0.02). Adult guinea pigs treated with ISO had a significantly higher T_p-e/QT_c than young guinea pigs treated with ISO (P = 0.04). Statistical analysis of transformed data was done using one-way ANOVA with Bonferroni’s multiple-comparisons analysis. Young control group: n = 6, young ISO-treated group: n = 5, adult control group: n = 5, and adult ISO-treated group: n = 6.

Flecainide But Not Quinidine Decreases Sudden Death Due to Acute ISO exposure by Decreasing PVCs

Treatment with flecainide before acute ISO exposure lowered the incidence of arrhythmias and sudden death in both young and adult guinea pigs (Fig. 3). Although 52% of young guinea pigs (37 of 71 guinea pigs) treated with ISO collapsed in the first hour, none of the young guinea pigs (0 of 27 guinea pigs) treated with ISO + flecainide collapsed or died (Fig. 3, A–C). In contrast, although 61% of adult guinea pigs (11 of 18 guinea pigs) collapsed when exposed to ISO, 26% of the adult guinea pigs (5 of 19 guinea pigs) that received ISO + flecainide collapsed (Fig. 3A). Importantly, although none of the adult guinea pigs exposed to ISO survived if they developed ventricular fibrillation, even with chest compressions (0 of 11 guinea pigs), 40% of guinea pigs (2 of 5 guinea pigs) survived when treated with ISO + flecainide and chest compressions (Fig. 3B). Young guinea pigs treated with ISO + flecainide developed significantly fewer PVCs than young guinea pigs treated with ISO (Fig. 3D). Although adult guinea pigs that received ISO + flecainide had fewer PVCs compared with adult guinea pigs treated with ISO, these guinea pigs still experienced significantly more PVCs than young guinea pigs that received ISO + flecainide. Interestingly, young guinea pigs treated with ISO + quinidine still collapsed (2 of 6 guinea pigs), and there was no difference in the number of PVCs compared with young control and ISO-treated guinea pigs (Fig. 3, E and F).

Fig. 3.

Flecainide (Flec) decreases sudden death due to acute isoproterenol (ISO) exposure, but quinidine (Quin) does not. A: percentage of guinea pigs that collapsed in the first hour of ISO exposure with and without Flec; 52% of young guinea pigs (37 of 71 guinea pigs) collapsed with ISO exposure, whereas none of the young guinea pigs (0 of 27 guinea pigs) treated with ISO + Flec collapsed, and 61% of adult guinea pigs (11 of 18 guinea pigs) exposed to ISO collapsed, whereas 26% of adult GPs (5 of 19 guinea pigs) treated with ISO + Flec collapsed. The percentage of adult GPs that collapsed was statistically significant by a Mantel-Cox log-rank test (P = 0.03). Young control group: n = 8, young ISO-treated group: n = 29, young ISO + Flec-treated group n = 27, adult control group: n = 9, adult ISO-treated group: n = 7, and adult ISO + Flec-treated group: n = 19. B: percent survival of guinea pigs that collapsed when treated with ISO or ISO + Flec. When exposed to ISO, 68% of young guinea pigs (25 of 37 guinea pigs) survived after collapse, whereas 100% of young guinea pigs treated with Flec survived. When treated with ISO, none of the adult guinea pigs (0 of 11 guinea pigs) that collapsed survived, but when treated with Flec at the time of pump implantation, 40% of the guinea pigs (2 of 5 guinea pigs) that collapsed survived (Mantel-Cox log-rank test, P < 0.0001). Young control group: n = 8, young ISO-treated group: n = 29, young ISO + Flec-treated group: n = 27, adult control group: n = 9, adult ISO-treated group: n = 7, and adult ISO + Flec-treated group: n = 19. C: percent survival in guinea pigs in the first hour of ISO exposure; 83% of young guinea pigs (59 of 71 guinea pigs) survived acute ISO exposure, 100% of young guinea pigs (27 of 27 guinea pigs) treated with Flec survived ISO exposure, 39% of adult guinea pigs (7 of 18 guinea pigs) survived acute ISO exposure, and 84% of adult guinea pigs (16 of 19 guinea pigs) treated with Flec survived ISO exposure. Mortality in adult guinea pigs was statistically significant by Mantel-Cox log-rank test (P = 0.013). Young control group: n = 8, young ISO-treated group: n = 29, young ISO + Flec-treated group: n = 27, adult control group: n = 9, adult ISO-treated group: n = 7, and adult ISO + Flec-treated group: n = 19. D: quantification of premature ventricular contractions (PVCs) from 14 min of recordings within 30 min of pump implantation. Flec significantly decreased the number of PVCs in young animals compared with young animals treated with ISO alone (P = 0.001). Adult animals treated with ISO + Flec had significantly more PVCs than young animals treated with ISO + Flec (P = 0.02). Statistical analysis of data was performed by Kruskal-Wallis test with Dunn’s multiple comparisons. Young control group: n = 15, young ISO-treated group: n = 29, young ISO + Flec-treated group: n = 16, adult control group: n = 9, adult ISO-treated group: n = 7, and adult ISO + Flec-treated group: n = 9. E: percentage of young guinea pigs that collapsed in 30 min of ISO exposure with and without Quin; 52% of young guinea pigs (37 of 71 guinea pigs) collapsed with ISO exposure, whereas 33% of young guinea pigs (2 of 6 guinea pigs) collapsed when treated with ISO + Quin (P = 0.07). Young control group: n = 8, young ISO-treated group: n = 29, and young ISO + Quin-treated group: n = 5. F: young guinea pigs treated with ISO + Quin had significantly more PVCs than young guinea pigs treated with saline (P = 0.003). There were no differences in the number of PVCs in young animals treated with ISO + Quin compared with young animals treated with ISO. Statistical analysis of data was performed by Kruskal-Wallis test with Dunn’s multiple comparisons. Young control group: n = 8, young ISO-treated group: n = 29, and young ISO + Quin-treated group: n = 5.

There Are No Significant Differences Between Age or Treatment Groups in Phosphorylation of Important Ca²⁺-Handling Proteins

There were no differences in total RyR2 protein between young and adult guinea pigs under any treatment condition (Fig. 4, A and B). There are two sites on RyR2 phosphorylated by PKA: Ser²⁸⁰⁸ and Ser²⁰³⁰ (4), and no differences in phosphorylation at either site in young or adult GPs across treatment groups were observed (Fig. 4, C–F). In addition, RyR2 is phosphorylated at Ser²⁸¹⁴ by CaMKII. Phosphorylation at this site has also been shown to increase Ca²⁺ leak (4, 70); however, acute exposure to ISO did not alter phosphorylation of RyR2 at Ser²⁸¹⁴, and there were no differences in phosphorylation at Ser²⁸¹⁴ between young and adult guinea pigs (Fig. 4, G and H). To determine phosphorylation of other PKA targets, we investigated phosphorylation of PLN at Ser¹⁶. There were no differences in total PLN between age or treatment groups (Fig. 5A). Ser¹⁶ phosphorylation is significantly lower in adult guinea pigs than in young guinea pigs (control young guinea pigss: 1.0 ± 0.33 and control adult guinea pigs: 0.44 ± 0.06, P < 0.04), but there were no differences in phosphorylation with ISO or ISO + flecainide compared with age-matched control guinea pigs. There were also no differences between age or treatment groups in the phosphorylation of PLN at Thr¹⁷ (CaMKII site; Fig. 5, E and F).

Fig. 4.

Isoproterenol (ISO) and flecainide (Flec) do not significantly change phosphorylation of ryanodine receptor 2 (RyR2). A: representative Western blot of total RyR2 in guinea pig left ventricles (LVs). B: quantification of total RyR2/GAPDH normalized to control groups. There were no significant differences in transformed data of total RyR2 expression in young or adult hearts by one-way ANOVA with Bonferroni’s multiple comparisons. Young control group: n = 6, young ISO-treated group: n = 8, young ISO + Flec-treated group: n = 6, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. C: representative Western blot of phosphorylation of RyR2 at Ser²⁸⁰⁸ in guinea pig LVs. D: quantification of phosphorylation of Ser²⁸⁰⁸/total RyR2 normalized to control groups. There were no significant differences in transformed data by one-way ANOVA with Bonferroni’s multiple comparisons. Young control group: n = 5, young ISO-treated group: n = 8, young ISO + Flec-treated group: n = 8, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. E: representative Western blot of phosphorylation of Ser²⁰³⁰ RyR2. F: quantification of RyR2 phosphorylation at Ser²⁰³⁰/total RyR2 normalized to control groups. There were no significant differences in transformed data by one-way ANOVA with Bonferroni’s multiple comparisons. Young control group: n = 4, young ISO-treated group: n = 8, young ISO + Flec-treated group: n = 9, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. G: representative Western blot of RyR2 phosphorylation at Ser²⁸¹⁴ in guinea pig LVs. H: quantification of RyR2 phosphorylation at Ser²⁸¹⁴/total RyR2 normalized to control groups. There were no significant differences in transformed data by one-way ANOVA with Bonferroni’s multiple comparisons. Young control group: n = 3, young ISO-treated group: n = 7, young ISO + Flec-treated group: n = 6, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. The blots for RYR2 and GAPDH are the appropriate and same controls for the adult samples in C and E.

Fig. 5.

Isoproterenol (ISO) and flecainide (Flec) do not significantly change phosphorylation of phospholamban (PLN). A: representative Western blot of total PLN in guinea pig left ventricles (LVs). B: quantification of total PLN/GAPDH normalized to control groups. There were no significant differences in transformed data by one-way ANOVA with Bonferroni’s multiple comparisons. Young control group: n = 7, young ISO-treated group: n = 7, young ISO + Flec-treated group: n = 10, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. C: representative Western blot of phosphorylation of PLN at Ser¹⁶ in GP LVs. D: quantification of phosphorylation of Ser¹⁶/total PLN normalized to control groups. There were no significant differences in phosphorylation of Ser¹⁶ between treatment groups in adult and young guinea pigs. Young control group: n = 7, young ISO-treated group: n = 10, young ISO + Flec-treated group: n = 9, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. E: representative Western blot of phosphorylation of PLN at Thr¹⁷. F: quantification of phosphorylation of PLN at Thr¹⁷/total PLN normalized to control groups. There were no significant differences in transformed data by one-way ANOVA with Bonferroni’s multiple comparisons. Young control group: n = 7, young ISO-treatet group: n = 10, young ISO + Flec-treated group: n = 9, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. Blots for GAPDH and total PLN are the appropriate and same controls for samples presented in A and C (young) and A, C, and E (adult).

Acute Exposure to ISO in Adult Guinea Pigs Increases Oxidation of RyR2

ISO treatment can result in increased protein oxidation (5, 40, 76, 77). To determine if oxidation increases in the left ventricle of guinea pigs in response to 30-min exposure to ISO, expression of carbonyl groups in proteins was assessed. When carbonyl groups were derivatized to DNP by 2,4-dinitrophenylhydrazine (a marker used to indicate oxidative state) (10), there was a trend toward higher overall expression of carbonyl groups in proteins from adult guinea pigs treated with ISO (Fig. 6A). Importantly, RyR2 had significantly more oxidative modifications in adult guinea pigs treated with ISO than young guinea pigs treated with ISO (Fig. 6C). Both young and adult guinea pigs treated with ISO had significantly more oxidation in RyR2 than control guinea pigs. Furthermore, RyR2 oxidation was not different between young animals treated with ISO + flecainide and young control animals, whereas in adult animals treated with ISO + flecainide, there was a significant decrease in the oxidative modifications of RyR2 (Fig. 6C).

Fig. 6.

Oxidation of ryanodine receptor 2 (RyR2) is higher in adult guinea pigs treated with isoproterenol (ISO). A: quantification of total 2,4-dinitrophenylhydrazone (DNP) in protein lysates from guinea pig left ventricles (LVs). There was a trend toward increased DNP in adult ISO-treated animals compared with young ISO-treated animals. Young control group: n = 6, young ISO-treated group: n = 8, young ISO + Flec-treated group: n = 5, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. B: representative image of DNP at RyR2 in guinea pig LVs with and without 2,4-dinitrophenylhydrazine (DNPH) treatment. C: quantification of DNP at RyR2/total RyR2. There was significantly more oxidation of RyR2 in young guinea pigs treated with ISO compared with control guinea pigs (P = 0.1). RyR2 was more oxidized in adult guinea pigs treated with ISO than control or ISO + Flec-treated guinea pigs (P = 0.04 and P = 0.004). RyR2 was significantly more oxidized in adult guinea pigs treated with ISO compared with young guinea pigs treated with ISO (P = 0.01). Statistical analysis of transformed data was by one-way ANOVA with Bonferroni’s multiple comparisons. Young control group: n = 6, young ISO-treated group: n = 8, young ISO + Flec-treated group: n = 5, adult control group: n = 3, adult ISO-treated group: n = 3, and adult ISO + Flec-treated group: n = 3. Flec, flecainide.

DISCUSSION

The present study is the first, to our knowledge, to examine age differences in arrhythmogenic sudden cardiac death in guinea pigs. Implantable loop recorders were used to document changes in the underlying rhythm, electrocardiographic intervals, and timing of arrhythmic events in real time. Similar to what is observed clinically where adult patients with HF are more likely to develop arrhythmias and sudden death than pediatric patients with HF, older guinea pigs also showed a greater propensity to arrhythmias and sudden death than young guinea pigs when acutely exposed to ISO. After acute treatment with ISO, older guinea pigs had significant changes in ECG parameters and more PVCs than young guinea pigs, which were associated with an increased incidence of lethal ventricular arrhythmias. Older guinea pigs also had higher oxidation of RyR2 than did young guinea pigs with ISO-mediated β-AR stimulation. Treatment with the anti-arrhythmic RyR2 channel blocker flecainide blunted arrhythmogenic events and resulted in less RyR2 oxidation in older guinea pigs. Although there are well-described age-related cardiac structural changes, in particular increasing wall thickness (59) and fibrotic changes (23), that predispose patients to arrhythmogenesis (11, 37), the present data suggest contributions from dynamic changes in cellular signaling.

The initial mortality in guinea pigs in response to acute exposure to 32 mg/kg ISO initiated the present study. We chose the dose based on dosing studies in mice and guinea pigs that produced pathological hypertrophy under chronic stimulation (58, 60). Several groups have previously demonstrated early mortality in guinea pigs after ISO treatment at varying doses [Liu et al. (33): 1 mg·kg⁻¹·day⁻¹, Overholser et al. (46): 9.6 µg·kg⁻¹·day⁻¹ and 0.312, 0.48, and 4.8 mg·kg⁻¹·day⁻¹, and Soltysinska et al. (58): 19.2 mg·kg⁻¹·day⁻¹]. In line with the occurrence of sudden death associated with a wide range of ISO doses, we found that high (32 mg·kg⁻¹·day⁻¹) and low (16 mg·kg⁻¹·day⁻¹) doses of ISO led to similar mortality in both age groups. Indeed, Soltysinska et al. (58) found that 20% of guinea pigs that received chronic ISO died suddenly, and the authors attributed this death to arrhythmias but the acute mechanism was not described. The present study extends these findings by not only demonstrating an age difference in sudden death but also documenting the underlying electrocardiographic mechanism as well as a putative molecular mechanism through changes in RyR2 oxidation.

We hypothesized that these age-specific differences in sudden death may be due to differences in rhythm characteristics between young and adult guinea pigs. Greer-Short et al. (17) reported increased ectopic activity in Langendorff-perfused adult guinea pig hearts stimulated acutely with ISO when compared with young guinea pig hearts. In our in vivo model, we found that ISO increased the number of PVCs (ectopic activity) in both adult and young animals (Fig. 2B) but that older animals had a greater PVC burden than the young.

Soltysinska et al. (58) also demonstrated that chronic ISO administration (daily injections with increasing dosage over 3 mo) lengthened ventricular repolarization in guinea pigs. In the present study, although there were no age differences at baseline, ISO induced an age-dependent change in two repolarization parameters associated with increased arrhythmogenesis and sudden death. T_p-e (a marker for transmural dispersion of repolarization) and the T_p-e-to-QT_c ratio (an index of arrhythmogenesis) (19, 29, 50) were significantly higher in adult guinea pigs treated with ISO compared with age-matched control guinea pigs (Fig. 2, D and E) and compared with young guinea pigs treated with ISO. The acute nature of the ISO treatment in the present study suggests that there are significant age-related differences in cellular signaling in response to β-AR agonism that contribute to rhythm disturbances.

CPVT is a condition in humans in which acute adrenergic stimulation results in episodes of ventricular tachycardia (25, 27). The most commonly identified mutations in patients with CPVT occur in RyR2 channels and are hypothesized to lead to increased Ca²⁺ leak (25, 27). CPVT is often treated with flecainide, a class I antiarrhythmic that blocks Na⁺ and RyR2 channels (21, 67, 68). In the present study, administration of flecainide lowered the number of PVCs induced by exposure to ISO in both adult and young guinea pigs (Fig. 3) (21, 67, 73). In contrast, quinidine (a class I antiarrhythmic that blocks Na⁺ channels but not RyR2 channels) did not lower the number of PVCs in ISO-treated guinea pigs. Since there were fewer PVCs in the animals treated with flecainide, we hypothesize that the RyR2 channel is central to the development of PVCs with acute β-AR stimulation.

β-AR stimulation increases phosphorylation of the RyR2 channel, but there is a lack of consensus about how phosphorylation at three well-documented sites (Ser²⁰³⁰, Ser²⁸⁰⁸, and Ser²⁸¹⁴) affects the function of the RyR2 channel (22, 23, 43, 53–63). It has been suggested that phosphorylation of RyR2 at Ser²⁸⁰⁸ and Ser²⁸¹⁴ occurs during HF, and this phosphorylation increases Ca²⁺ leak from the sarcoplasmic reticulum (4, 24, 26, 42a, 51, 56, 69). This increased Ca²⁺ can cause delayed afterdepolarizations and aberrant contractions as well as depletion of sarcoplasmic reticulum Ca²⁺ stores, thereby leading to decreased contractility (3, 25, 26). In guinea pigs acutely exposed to ISO, there were no significant differences in the phosphorylation of the three known sites on RyR2, which suggests that phosphorylation changes in RyR2 channels are not responsible for the PVCs in guinea pigs acutely exposed to β-AR stimulation (Fig. 4).

It is possible that changes in sarcoplasmic Ca²⁺ loading could contribute to changes in cytoplasmic Ca²⁺ and lead to arrhythmogenesis. PLN is phosphorylated at Ser¹⁶ by PKA, resulting in increased sarco(endo)plasmic reticulum Ca²⁺-ATPase 2 activity and increased Ca²⁺ reuptake into the sarcoplasmic reticulum (3). Changes in PLN structure or phosphorylation can alter this Ca²⁺ handling and could potentially increase arrhythmogenic propensity (32). However, there were no significant differences in PLN phosphorylation in adult or young guinea pigs exposed to ISO.

β-AR stimulation also adversely affects cell function through increasing production of ROS. In fact, acute β-AR stimulation has been shown to increase ROS within 15 min of stimulation (28). Increased ROS can lead to increased oxidation of cellular proteins (5, 40, 76, 77). Although several studies have demonstrated that phosphorylation of RyR2 at Ser²⁸⁰⁸ is not sufficient to cause Ca²⁺ leak (5, 75), it has been suggested that oxidation of the RyR2 channel increases Ca²⁺ leak (22, 64, 72). Furthermore, ROS have been shown to oxidize RyR2, altering the activity of the channel (5, 30, 42a, 64, 77). Becerra et al. (2) demonstrated that blockade of oxidative modifications of RyR2 in an ischemia-reperfusion model leads to decreased arrhythmias. Since, in the guinea pigs model, every episode of ventricular arrhythmia was initiated by a PVC, we hypothesized that differences in ROS production between adult and young guinea pigs mediated oxidation of RyR2, which, in turn, influenced PVC generation. Importantly, RyR2 was significantly more oxidized in adult animals treated with ISO than in young animals treated with ISO. Additionally, treatment with flecainide prevented RyR2 oxidation in adult and young animals, and this was associated with fewer PVCs and improved survival. Therefore, we believe acute ISO exposure in guinea pigs leads to ROS production, which oxidizes RyR2 channels, leading to PVC production. A higher PVC burden in combination with the repolarization abnormalities could lead to an increased chance that a PVC will occur during ventricular repolarization (R-on-T phenomena), which leads to ventricular arrhythmias and death.

Limitations

There are several limitations of this study. First, the amount of ISO eluted from the pumps immediately upon implantation is not known. All pumps were primed for similar durations before implant to mitigate any variability. It is also interesting to note that this is an acute phenomenon, and after the first hour postsurgery, no animals experienced ventricular tachycardia. We originally hypothesized that β-adrenergic signaling would activate CaMKII and PKA pathways; however, phosphorylation at known PKA and CaMKII sites for both RyR2 and PLN did not show significant changes. We believe the timing of this experiment might play a role, and it is possible that earlier or later assessment might demonstrate activation of these pathways. Second, there may be differences in the blockade of Na⁺ channels between quinidine and flecainide that were incompletely characterized. It is also possible that there is an age-dependent difference in the response to quinidine that was not captured in the present experiments. However, the remarkable response to flecainide and the known RyR effects in CPVT along with the changes in RyR oxidation in guinea pigs provide significant support for the critical role of RyR in this model. Although we suspect ROS production is related to PVC generation, this is an associative finding, and these experiments do not conclusively define the cause of the PVCs. Finally, there are many factors that contribute to arrhythmogenesis, and RyR2 oxidation does not explain the differences in repolarization demonstrated with acute ISO exposure in adult guinea pigs. It is clear RyR2 is only one contributor, and the differences between young and adult guinea pigs in response to acute ISO stimulation is multifactorial.

Conclusions

In this study, the underlying rhythm, electrocardiographic intervals, and timing of arrhythmic changes in response to acute β-AR stimulation were characterized in guinea pigs. Importantly, we evaluated the relationship between rhythm disturbances responsible for age-related differences in sudden death in guinea pigs. Flecainide, but not quinidine, lowered the number of PVCs, suggesting that RyR2 channels contribute to rhythm changes in response to acute β-AR stimulation in guinea pigs. Importantly, age-specific differences in RyR2 oxidation resulted from acute ISO treatment, and RyR2 oxidation was lower with the use of flecainide. The increased number of PVCs, repolarization abnormalities, and oxidation of RyR2 in adult guinea pigs in response to ISO led to a higher incidence of fatal ventricular arrhythmias compared with their younger counterparts.

GRANTS

This work was supported by funding from National Heart, Lung, and Blood Institute Grants R01-HL-07715 (to B. L. Stauffer), R01-HL-16928 (to S. D. Miyamoto), T32-HL-007171 (to K. C. Woulfe), and R25-HL-103286 (to S. Nau) and American Heart Association Grant 16POST29970010 (to K. C. Woulfe). Medtronic provided research support to investigate the use of LINQ cardiac monitors in animal research.

DISCLAIMERS

The contents are the authors’ sole responsibility and do not necessarily represent official National Institutes of Health views.

DISCLOSURES

C. C. Sucharov, B. L. Stauffer, and S. D. Miyamoto are founders and scientific advisors for CoramiR Biomedical, LLC.

AUTHOR CONTRIBUTIONS

K.C.W. and B.L.S. conceived and designed research; K.C.W., C.E.W., S.N., S.C., and E.K.P. performed experiments; K.C.W., C.E.W., S.N., S.Z., and C.T. analyzed data; K.C.W., S.Z., C.T., C.C.S., S.D.M., and B.L.S. interpreted results of experiments; K.C.W. prepared figures; K.C.W. drafted manuscript; K.C.W., C.C.S., S.D.M., and B.L.S. edited and revised manuscript; K.C.W., C.T., C.C.S., S.D.M., and B.L.S. approved final version of manuscript.