Myocardial fibrosis is regarded as a pivotal proarrhythmic substrate, but there have been no comprehensive studies showing a correlation between the severity of fibrosis and ventricular tachyarrhythmias (VTAs). Our purpose was to document this relationship in a transgenic (TG) strain of mice with fibrotic cardiomyopathy. TG mice with cardiac overexpression of β2-adrenoceptors (β2-AR mice) and non-TG (NTG) littermates were studied at 4–12 mo of age. VTA was quantified by ECG telemetry. The effect of pharmacological blockade of β2-ARs on VTA was examined. Myocardial collagen content was determined by hydroxyproline assay. NTG and TG mice displayed circadian variation in heart rate, which was higher in TG mice than in NTG mice (P <0.05). Frequent spontaneous ventricular ectopic beats (VEBs) and ventricular tachycardia (VT) were prominent in TG mice but not present in NTG mice. The frequency of VEB and VT episodes in TG mice increased with age (P < 0.01). Ventricular collagen content was greater in TG mice than in NTG mice (P <0.001) and correlated with age (r = 0.71, P < 0.01). The number of VEBs or VT episodes correlated with age (r = 0.83 and r = 0.73) and the content of total or cross-linked collagen (r = 0.62∼0.66, all P <0.01). While having no effect in younger β2-TG mice, β2-AR blockade reduced the frequency of VTA in old β2-TG mice with more severe fibrosis. In conclusion, β2-TG mice exhibit interstitial fibrosis and spontaneous onset of VTA, becoming more severe with aging. The extent of cardiac fibrosis is a major determinant for both the frequency of VTA and proarrhythmic action of β2-AR activation.
- ventricular tachyarrhythmias
- β-adrenergic receptor
NEW & NOTEWORTHY
We demonstrated in a mouse model of cardiomyopathy that the severity of myocardial fibrosis correlated significantly with telemetry-quantified frequency of ventricular tachyarrhythmias. Furthermore, severe fibrosis is the prerequisite for the proarrhythmic effect of β2-adrenergic activation in this model. Our findings improve current understanding on fibrosis as an arrhythmia substrate.
ventricular tachyarrhythmias (VTAs) incorporate a spectrum of abnormal ventricular rhythm, including ventricular ectopic beat (VEB), ventricular tachycardia (VT), and fibrillation. VTAs are commonly seen in patients with structural heart diseases and represent the most common cause for sudden cardiac death (SCD) in these patients (11, 23, 37, 40). VTA also form an independent risk for heart failure (HF) and mortality in HF patients (1, 43), and, conversely, catheter-based ablation of VTA is associated with functional improvement (34). Despite significant progress in our understanding of mechanisms and device- or drug-based treatment, the efficacy of current therapies in controlling arrhythmias and related complications is unsatisfactory (37).
Cardiac MRI-based late gadolinium enhancement (LGE) has been increasingly used to detect fibrotic scars or patch fibrosis. Using this technique, clinical studies have established links between the risk of spontaneous or induced VTA and cardiac histopathologies, including the size of fibrotic scars or patch fibrosis (8, 10, 22, 28, 31). In patients with dilated or hypertrophic cardiomyopathy, LGE-based detection of myocardial fibrosis is associated with increased risk of all-cause mortality and SCD (8, 22, 35). In rats with chronic myocardial infarction, the severity of VTA induced by sympathetic nerve stimulation correlated with the size of scar of the left ventricle (LV) (13). However, there has been no clinical or experimental evidence in vivo to show a quantitative relationship between interstitial fibrosis and VTAs. We previously demonstrated that a strain of transgenic (TG) mice with cardiomyocyte-restricted overexpression of β2-adrenoceptors (β2-TG mice) displayed a phenotype of fibrotic cardiomyopathy and premature death due to HF and non-HF causes (15, 19, 46). The arrhythmic phenotype in this β2-TG model and its relationship to the development of cardiac fibrosis has not previously been studied. Two prior studies (12, 26) have revealed the arrhythmogenic nature of β2-AR stimulation in rabbit or human failing hearts. Hence, our aims in the present study were to 1) characterize the arrhythmia phenotype in the β2-TG model by telemetry ECG with particular emphasis on the relationship between the severity of VTA and cardiac fibrosis and 2) to test the significance of β2-adrenergic activation in arrhythmogenesis in relation to fibrosis.
MATERIALS AND METHODS
Male β2-TG and non-TG (NTG) littermate mice aged 4–12 mo were studied. Mice were initially generated onto a mixed background (C57BL/SJL) (15, 30) and were then backcrossed (>10 times) into the C57BL/6 genetic background. In this β2-TG model, male mice exhibit more severe cardiomyopathy, including fibrosis, than female counterparts (19). We therefore studied only male mice. Mice were housed in standard conditions in a 12:12-h light-dark cycle. At the end of the experiment, mice were killed by an anesthetic overdose [ketamine (200 mg/kg)-xylazine (40 mg/kg)-atropine (2.4 mg/kg), intraperitoneally] and followed by heart removal. The experimental procedures described were approved by the local animal ethics committee and conformed with the Australian Code for the Care and Use of Animals for Scientific Purposes (8th edition).
Under isoflurane anesthesia, Data Science (St. Paul, MN) ambulatory telemeters (TA11ETA-F10 and TA10EA-F20) were implanted in mice according to the manufacturer's instructions and as previously described (36). In brief, an ECG telemeter was implanted into the abdominal cavity with the negative and positive leads subcutaneously positioned at the right foreleg and left side of the chest, respectively, approximating a lead II ECG configuration. After implantation, mice were given a recovery period of 14 days. The ECG was recorded in conscious mice while in their home cage at 1 kHz for 48 h using the Dataquest ART Acquisition System (Data Science).
Effects of receptor antagonists in vivo.
To determine the influences of autonomic nervous activity on the circadian variation of heart rate (HR) and frequency of arrhythmias, β2-TG and NTG mice were treated with selective antagonists to block β2-ARs (ICI-118,551 at 1 mg/kg, Sigma, St. Louis, MO), β1-ARs (atenolol at 2 mg/kg, Sigma), and muscarinic receptors (atropine at 1.2 mg/kg, Pfizer). All drugs were given subcutaneously as a single bolus. Doses of drugs were selected based on our previous studies (14, 16) showing effective blockade of the respective receptors.
Recording and analysis of body surface ECG.
Frequent onset of VEB and VT in β2-TG mice by telemetry monitoring implies the possibility that a brief period of ECG recording is able to provide quantitative estimation of the severity of arrhythmias. To test this possibility, we studied separate batches of NTG and β2
Animals were maintained under light anesthesia by inhalation of 1.8% isoflurane. LV size and function were determined by echocardiography using an iE33 ultrasound machine and a 15-MHz linear transducer (Philips, San Jose, CA). Short-axis two-dimensional image-guided M-mode images crossing the LV at the level of papillary muscles were acquired. LV dimensions at diastole and systole were measured. Fractional shortening was calculated as previously described (19).
At the end of the study, hearts were harvested, immersed in saline at 4°C, dissected into separate parts, and weighed. The LV was frozen in liquid nitrogen and stored in −80°C until use. As previously described (15, 19), total collagen content of the apical portion of the LV (∼20 mg) was determined using a hydroxyproline assay. The concentration of soluble and insoluble (cross-linked) collagen was determined by digesting tissue in pepsin with acetic acid for 24 h. This was followed by centrifugation to separate the supernatant (pepsin-soluble collagen) and pellet (insoluble collagen) fractions as previously described (17). Concentrations of soluble and insoluble fractions were then determined by the hydroxyproline assay. Data are expressed as collagen concentration per dry weight.
Gene expression analysis.
RNA was extracted from LV tissues using TRIzol reagent (Sigma) according to the manufacturer's instructions. DNase-treated RNA (2 μg) was reverse transcribed into cDNA using a Moloney Murine Leukemia Virus Reverse Transcriptase Kit (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. Using SYBR green reactions (Roche, Mannheim, Germany) and target primers with the ABI 7500 Fast RT-PCR system, we determined the expression of the following genes related to fibrosis, inflammation, oxidative stress, or electrophysiology: α-smooth muscle actin (α-SMA), transforming growth factor (TGF)-β1, connective tissue growth factor, procollagen types I and III, fibronectin, matrix metalloproteinase-2, IL-6, monocyte chemoattractant protein 1, isoforms of NADPH oxidase (NOX2 and NOX4), connexin (Cx)43 and Cx45, hyperpolarization-activated cyclic nucleotide-gated K+ channel (HCN)2 and HCN4, Na+/Ca2+ exchanger (NCX)1, and sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA)2a. PCR primers specific to the respective genes were designed using Primer3, maintained by the National Center for Biotechnology Information (Bethesda, MD), and ordered from Sigma. Once received, all primers were verified against their reference sequence for their specificity before use. Gene expression was normalized by the expression level of 18S or GAPDH, and the fold change was calculated according to the 2−ΔΔCT method (where CT is threshold cycle). Results are presented relative to that of the NTG group.
Analysis of telemetry ECG and detection of arrhythmias.
9). A programmed detection method was developed to quantify VEBs and episodes of VT from all 48 h of ECG recordings. This method took advantage of increased amplitudes of R and/or S waves of VEBs by setting positive and negative voltage thresholds beyond the normal sinus rhythm amplitudes. In doing so, VEBs that exhibit morphologically greater positive or negative amplitudes were detected. Episodes of VT were detected using integrals of cumulative slopes derived from consecutive VEBs. To validate the telemetry results obtained by programmed detection, 1-h ECG recordings were chosen for manual screening, and the results were compared with those obtained by the programmed detection. Reproducibility of telemetry data was also examined by comparing two consecutive 24-h period of recording (i.e., days 1 and 2).
Analyses of data were performed using either Prism 6 (GraphPad Software, La Jolla, CA) or SPSS Statistics 19 (IBM, Armonk, NY). Both the absolute number of VEB and VT episodes failed in passing the Shapiro-Wilk's test for normal distribution (P <0.05) and the Levene's test for equal variance (P <0.05). Thus, a log(y + 1) transformation was applied, and the transformed data passed both normality (P > 0.05) and equal variance (P > 0.05) tests (9). Transformed data were then compared using one- or two-way ANOVA. The relationship between the severity of VTA and aging or extent of cardiac fibrosis was assessed using Pearson correlation coefficient analysis. Incidence data were compared using a χ2-test or Fisher's exact test. Two-way ANOVA was used to compare arrhythmia results or scores between groups. A Bonferroni post hoc test was performed if ANOVA indicated a statistically significant difference. As there were no age-related differences in NTG mice in all parameters studied, NTG mice were combined to form a single group for clarity, unless otherwise indicated. Parametric results are expressed as means ± SE. P values of <0.05 were considered statistically significant.
Circadian changes in HR and physical activity in conscious mice.
Male NTG and β2-TG mice of all age groups demonstrated circadian differences in HR expressed either as hourly averages over a 48-h period or averaged level of the light versus dark phase (Fig. 1A). Compared with NTG mice, β2-TG mice at all ages had higher HR levels across 48 h (P < 0.05; Fig. 1A). HR levels of β2-TG mice during the light and dark phase were similar between different age groups. HR was also found to be reproducible between the two 24-h periods (data not shown). NTG and β2-TG mice exhibited similar circadian patterns in physical activity across 48 h, with the value significantly greater during the dark phase in all groups studied (P < 0.01; Fig. 1B).
Arrhythmic phenotype in β2-TG mice.
Typical ECG configurations of NTG and β2-TG mice are shown in Fig. 2, A and B. NTG mice rarely displayed VEBs and never showed VT episodes. Frequent onset of VEB and VT was noticed in β2-TG mice by ECG telemetry (Fig. 2, C–E). VEBs were presented as polymorphic singles or bigeminy (Fig. 2, C and D). Episodes of VT were typically brief in duration (Fig. 2E). In addition to VEBs and VTs as major types of arrhythmias, manual validation of telemetry ECG revealed the presence of other types of arrhythmias in some β2-TG mice, including atrial-ventricular (AV) dissociation with well-maintained ventricular rhythm and atrial flutter (data not shown). We validated the accuracy of programmed detection of VEB and VT by manually screening 1-h ECG traces and conducted paired statistical analysis with automatically detected results of the same period. There was no significant difference between the two sets of independently derived arrhythmia data (data not shown). Our data also suggested similarity in frequencies of VEB and VT on day 1 versus day 2 (data not shown).
Circadian differences were evident in the frequency of VEBs and VT episodes across 48 h in β2-TG mice (Fig. 3, A and B). In β2-TG mice, age was associated with increased frequency and severity of spontaneous VEB and VT (both P < 0.05 by two-way repeated-measures ANOVA; Fig. 3, A and B). For instance, the number of VEBs was ∼400 VEBs/h in 4- to 6-mo-old β2-TG (TG4–6) mice but was 10 times greater in 9- to 11-mo-old β2-TG (TG9–11) mice. The number of VT episodes increased from 3 VEBs/h in TG4–6 mice to 54 VEBS/h in TG9–11 mice.
Under isoflurane anesthesia, 10-min ECG recording also detected a high frequency of VTAs in β2-TG mice. The incidence of spontaneous onset of VEBs was detected in >95% of β2-TG mice (P < 0.05 vs. 6% in NTG mice; Fig. 4A). VT was only observed in β2-TG mice, with its incidence increasing with age (Fig. 4B). An age-dependent increase in the severity of arrhythmias was indicted by the absolute number of VEBs or arrhythmia score (Fig. 4, C and D).
Effects of β1-AR, β2-AR, and muscarinic antagonists on HR and arrhythmias.
Next, we studied the effects of selective antagonists with continuous ECG telemetry recording. The timing of drug administration was determined by pilot experiments to ascertain the duration of the drug action and by taking into consideration the HR circadian pattern. Under conscious conditions, the duration of the drug effect on HR was found to be ∼7 h for ICI-118,551 and ∼2 h for both atenolol and atropine. Assuming that the sharp HR increment around commencement of the dark phase was indicative of sympathetic activation, atenolol was administered 5 min before the dark phase, whereas ICI-118,551 was given 2 h before the commencement of the dark phase. Blockade of β1-AR by atenolol significantly reduced HR in NTG mice (P < 0.001; Fig. 5A). Conversely, the selective β2-AR antagonist ICI-118,551 had no effect on HR in NTG mice but profoundly lowered HR in β2-TG mice (P < 0.001; Fig. 6A). Atropine was administered during the middle of the light phase, when HR was at the lowest level. Blockade of muscarinic receptors significantly increased HR, an effect that was similar in NTG and β2-TG mice and lasted for 2 h (P < 0.01 by repeated-measures ANOVA; Fig. 5B).
Changes in HR confirmed the expected receptor-blocking actions of the drugs. We then examined changes in VTA after drug intervention by comparing the frequency of VEB and VT in the same animals recorded at exactly the same duration of the preceding day. The frequency of VEB and VT was unaffected by atenolol or atropine (Fig. 5C, D). The frequency of VEB in TG4–6 and 7- to 8-mo-old β2-TG (TG7–8) groups was unaffected by treatment with ICI-118,551 but was reduced by ∼40% in the TG9–11 group (P < 0.05; Fig. 6B).
Relationship of cardiac fibrosis, age, and severity of VTA.
An age-dependent increase in the severity of interstitial fibrosis in β2-TG mice was clearly indicated by LV sections with collagen stained by picrosirus red (Fig. 7A). To illustrate the relationship between the severity of VTA, age, and cardiac fibrosis, hearts were harvested from all NTG and β2-TG mice at the end of the telemetry experiment for the hydroxyproline assay. We determined both total collagen content and the percentage of insoluble collagen in the LV. Collagen content was similar in NTG mice aged from 4 to 12 mo (Fig. 7B). Compared with NTG mice, both total collagen content and insoluble collagen were significantly higher in β2-TG mice and correlated significantly with age (Fig. 7, B–D).
The total number of VEB and VT episodes by 24-h telemetry were positively correlated with age in β2-TG mice (r = 0.83 and r = 0.73, respectively; Fig. 8, A and B). Notably, total collagen content or the percentage of insoluble collagen of the LV was positively correlated with the 24-h counts of VEB (r = 0.63 and r = 0.66, respectively, P < 0.01; Fig. 8, C and D) and VT episodes (r = 0.62 and r = 0.66, respectively, P < 0.01; Fig. 8, E and F).
Age-dependent development of cardiomyopathy.
Echocardiography was performed, and organ weights were measured in some of NTG and β2-TG mice studied by ECG telemetry. Echocardiography at four different age groups revealed that the LV dimension at diastole increased in TG9–11 mice, whereas fractional shortening was significantly lower in TG9–11 mice relative to age-matched NTG mice (Fig. 9A). Weights of the LV, right ventricle, atria, and lungs were significantly increased in TG9–11 mice (P <0.05; Fig. 9B). These findings indicate the presence of HF in TG9–11 mice. TG9–11 mice also showed a significant increase in the QRS duration, indicating slow intraventricular conduction (Table 1).
Altered expression patterns of genes.
Expression levels of selected genes were determined using LVs of NTG and β2-TG mice at 4–6 and 9–11 mo of age, respectively. The profile of fibrotic genes in the LV of β2-TG mice provided further support for the fibrotic cardiomyopathy in this model. There was a severalfold increase in α-SMA as a measure of myofibroblasts (Fig. 10), profibrotic factors such as connective tissue growth factor, and extracellular matrix proteins, including procollagen type I, procollagen type III, and fibronectin (Table 2). Meanwhile, the enhanced oxidative stress and inflammation in β2-TG hearts were also indicated by elevated expression of NOX4, NOX2, IL-6, and monocyte chemoattractant protein-1 (Table 2). There was a threefold increase in the gene expression of Cx43 in β2-TG hearts (P < 0.05) together with a trend for increased Cx45 expression (P = 0.09; Fig. 10). Expression of NCX1 was increased by 70–90% (Fig. 10). Downregulation of HCN2 and SERCA2a together with upregulation of HCN4 were observed in the LVs of TG9–11 mice (Fig. 10).
We made three key findings in the present study: first, we demonstrated a previously unexplored phenotype in the β2-TG strain of mice, i.e., spontaneous and frequent onset of VTA that becomes increasingly severe with aging. Second, the frequency of VEB and VT quantified by telemetry correlated positively with both age and the severity of myocardial fibrosis. Finally, pharmacological blockade of β2-ARs lowered the frequency of VTA in old β2-TG mice with severe fibrosis.
Although the genetic targeting is restricted to cardiomyocytes, one of the early signs of cardiomyopathy in this β2-TG model is interstitial fibrosis, which becomes evident as early as 3–4 mo of age based on our previous study (3) and the present study. This implies that cardiomyocyte β2-ARs act in a paracrine fashion leading to extracellular matrix remodeling with excessive collagen deposition. We previously showed that the molecular mechanism involves enhanced oxidative stress due to the activation of NOX with the subsequent activation of inflammatory, fibrogenic, and apoptotic signaling (46). In β2-TG mice, a comparable functionality of transgene-driven β2-ARs is indicated by the similar increment of HR across all age groups. We previously observed in β2-TG mice an age-dependent progression of cardiomyopathy with premature death occurring from 8 mo of age (15, 19, 46). A similar cardiomyopathy phenotype was observed in the present study on β2-TG mice (C57BL/6 background) that exhibited LV dilatation and a decline in fractional shortening from 9 mo of age. This is in keeping with the 44% downregulation of SERCA2a in the hearts of β2-TG mice at 9–11 mo of age.
The β2-TG model is characterized by spontaneous and frequent onset of VTA. VEBs were observed in all β2-TG mice with polymorphic ECG configurations, suggesting multiple ectopic origins. ECG telemetry also revealed frequent episodes of VT, in keeping with operation of the reentry mechanism in the setting of interstitial fibrosis (32). The frequency of VTA is so high that even a brief ECG recording in mice under isoflurane anesthesia is able to detect VTA in a quantitative manner. ECG telemetry did not capture lethal VTA, such as sustained VT or ventricular fibrillation. However, in keeping with previous studies (15, 19), in the present study, four β2-TG mice died, most likely due to lethal arrhythmias or worsening HF due to frequent VTA. In addition to VEB and VT as the major arrhythmia forms, we also observed in β2-TG mice proximal atrial flutter and complete AV dissociation. For the latter, the rates of ventricular rhythm and sinus rhythm were very similar, likely attributable to an enhanced automaticity of the AV node driven by an enhanced β2-adrenergic signaling.
Interstitial fibrosis is generally regarded as a pivotal arrhythmic substrate (32). Thus, the severity of fibrosis would be expected to correlate with the frequency of VTA. However, few clinical or in vivo experimental studies exist that demonstrate such a quantitative relationship. We have provided, for the first time, experimental evidence for a significant and positive correlation between the severity of interstitial fibrosis and frequency of spontaneous VTA. Our finding strongly supports recent clinical reports showing that the presence of fibrotic tissues, measured by LGE, predicts the risk of SCD in patients with cardiomyopathy of diverse etiologies (8, 27, 31, 35). In addition to the severity of interstitial fibrosis, we observed an age-dependent development of LV dilatation and hypertrophy in β2-TG mice at 9–11 mo. We previously observed in the β2-TG model histopathology, such as cardiomyocyte apoptosis and hypertrophy (15, 19, 46), but the interstitial fibrosis, estimated by the expression of fibrotic genes, chemical assays, and quantitative histology, is an early sign of cardiomyopathy (3, 46). In the β2-TG model, both the severity of VTA and interstitial fibrosis are clearly age dependent. Our data imply that the correlation between age and the frequency of VTA is largely attributable to an age-dependent worsening of cardiac fibrosis. Considering the adverse impact by fibrosis on cardiac remodeling and dysfunction and the pivotal role of myofibroblasts in inflammatory, hypertrophic, and apoptotic signaling (18, 44, 45), it is likely that fibrosis is a key player in the age-dependent progression of LV remodeling and dysfunction. The significance of global as well as histological factors contributing to VTA in this model warrant further investigation. It is also critical to address the question of whether the severity of arrhythmias can be ameliorated by antifibrotic interventions.
The presence of patchy or interstitial fibrosis interferes with electrical coupling of adjacent cardiomyocytes, leading to slow conduction, formation of reentry, and afterdepolarization (32, 37, 45). A recent in vitro study (38) has also revealed direct electrical coupling via the formation of gap junctions between myofibroblasts and cardiomyocytes as an additional proarrhythmic mechanism. In this context, it is important to note the two- to threefold upregulation in the β2-TG heart of both the myofibroblast marker α-SMA and Cx43 as a pivotal gap junction protein (2, 33, 38). Altered intracellular Ca2+ handling, due either to attenuated uptake via SERCA2a or elevated extrusion by NCX1, is known to be arrhythmogenic (21). Graf et al. (20) showed in 3- to 5-mo-old β2-TG mice a significant increase in the amplitude of the HCN channel (i.e., If channel) together with upregulated expression of HCN4 but not HCN2 (20). Here, we observed HCN4 upregulation but HCN2 downregulation only in TG9–11 mice, when they exhibited signs of cardiac decompensation. HCN4 is known as the main molecule for the HCN channel, and numerous studies have documented the association of HCN4 upregulation and arrhythmias in heart diseases (39). Interestingly, upregulated expression of NCX1 and fibrotic and inflammatory genes was similar across the age groups studied, whereas changes in the expression of SERCA2a, HCN2, and HCN4 were evident only in TG9–11 mice. This implies that HF-associated remodeling of ion channels may contribute additionally to the onset of VTA.
Numerous clinical and experimental studies have demonstrated the role of the autonomic nervous system in the regulation of cardiac electrophysiology and risk of VTA (6, 41). VTAs can be induced by stimulation of sympathetic nerves or administration of β-agonists (12, 13, 26, 41). Conversely, activation of parasympathetic tone generally improves electrophysiological stability (41). After the demonstration of circadian variation of spontaneous VTA, likely driven by circadian changes in autonomic nervous activity in β2-TG mice, we tested if blockade of cardiac receptors of autonomic neurotransmitters, particularly β2-ARs, might reduce the severity of VTA. HR was similarly increased in NTG and β2-TG mice in response to atropine administered during the light phase, suggesting similar parasympathetic tone. The lack of influence by atropine on VTA indicated limited impact of the parasympathetic nervous activity in the β2-TG model. We found that pharmacological blockade of β1-ARs or muscarinic receptors failed to alter the frequency of VTA in the β2-TG model. Our previous study (5) showed that ICI-118,551 was effective in blocking β2-ARs in this model. In the present study on conscious mice, β2-TG mice showed a clear surge of HR immediately before the commencement of the dark cycle. Unlike NTG mice in which the β1-AR blocker atenolol induced bradycardia, ICI-118,551 induced a profound HR reduction in β2-TG mice, suggesting that the transgene-directed β2-AR becomes predominant over the β1-AR in the heart and that they are able to be activated by endogenous ligands. Here, we have shown that treatment with ICI-118,551 significantly reduced the frequency of VTA in conscious TG9–11 mice, whereas no such effect was observed in TG4–6 and TG7–8 mice. TG9–11 mice exhibited more severe cardiomyopathy, as measured by fibrosis and ventricular remodeling, compared with younger β2-TG animals. Thus, while supporting the findings from rabbit or human failing hearts (12, 26), our observation from the β2-TG model with the use of ICI-118,551 further indicates that the proarrhythmic nature of β2-AR activation is dependent on the severity of myocardial histopathology, particularly interstitial fibrosis.
There have been very limited models that exhibit spontaneous VTA, myocardial fibrosis, and ventricular remodeling that better simulate clinical entities. Animal models of nonischemic VTA are commonly induced by either chemicals or electrical stimulation in vitro and in vivo (29). These models usually lack cardiac histopathologies that constitute arrhythmic substrates. VTA or SCD have been reported in some strains of mice with mutations of proteins that regulate sympathetic neurotransmission (25), ionic channels (channelopathies), and intracellular Ca2+ transient or electromechanical coupling (4, 42). Again, very few models exhibit significant cardiac pathology. Our study in the β2-TG mouse model documented a positive correlation between the severity of interstitial fibrosis and VTA. As far as we are aware, there have been no similar clinical or experimental reports on such a correlation, albeit the correlation exists between the risk of arrhythmias or SCD and presence of myocardial fibrotic scars due to myocardial infarction (7, 13) or cardiomyopathy (8, 22, 28, 31, 35). Further clinical studies are warranted to quantitatively examine such association between interstitial fibrosis by cardiac MRI (24) and the severity of VTA in patients with defined heart diseases.
In conclusion, our study demonstrates that β2-TG mice develop spontaneous and frequent onset of VTA that are correlated significantly with the extent of interstitial fibrosis, implying a potential mechanistic link. Activation of β2-adrenergic signaling is arrhythmogenic in the setting of severe ventricular remodeling and fibrosis. Future research is warranted to illustrate factors that influence the severity of VTA and to perform therapeutic testing, particularly antifibrotic interventions, of potential antiarrhythmic efficacy.
This work was supported by the National Health and Medical Research Council (NHMRC) of Australia Grants ID1081710, ID1032687, and ID1008682 and Paceline and was supported in part by the Victorian Government's Operational Infrastructure Support Program. M.-N. Nguyen is a recipient of an Australian Postgraduate Award. J. R. McMullen and X. J. Du are NHMRC senior research fellows (NHMRC Grants ID1078985 and ID1043026).
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: M.-N.N., H.K., and X.-J.D. conception and design of research; M.-N.N., H.K., X.-M.G., Y.S., A.J., and L.-P.H. performed experiments; M.-N.N., H.K., D.R., Y.S., A.J., L.-P.H., and X.-J.D. analyzed data; M.-N.N. and X.-J.D. interpreted results of experiments; M.-N.N. and X.-J.D. prepared figures; M.-N.N. and X.-J.D. drafted manuscript; M.-N.N., H.K., J.R.M., and X.-J.D. edited and revised manuscript; M.-N.N., H.K., D.R., X.-M.G., Y.S., A.J., L.-P.H., J.R.M., and X.-J.D. approved final version of manuscript.
The authors thank Aya Matsumoto and Geeta Sapra (Baker IDI Heart and Diabetes Institute, Melbourne, Victoria, Australia) for technical support with the ECG telemetry experiments.
- Copyright © 2015 the American Physiological Society