Low-Dose Propranolol Prevents Functional Decline in Catecholamine-Induced Acute Heart Failure in Rats

Severe hyper-catecholaminergic states likely cause heart failure and cardiac fibrosis. While previous studies demonstrated the effects of beta-blockade in experimental models of single-catecholamine excess states, the detailed benefits of beta-blockade in more realistic models of hyper-adrenergic states are less clearly understood. In this study, we examined different therapeutic dosages and the effects of propranolol in rats with hyper-acute catecholamine-induced heart failure, and subsequent cardiopulmonary changes. Rats (n = 41) underwent a 6 h infusion of epinephrine and norepinephrine alone, with additional low-dose (1 mg/kg) or high-dose propranolol (10 mg/kg) at hour 1. Cardiac and pulmonary tissues were examined after 6 h. Catecholamine-only groups had the lowest survival rate. Higher doses of propranolol (15 mg/kg) caused similarly low survival rates and were not further analyzed. All low-dose propranolol rats survived, with a modest survival improvement in the high-dose propranolol groups. Left ventricular (LV) systolic pressure and LV end-diastolic pressure improved maximally with low-dose propranolol. Cardiac immunohistochemistry revealed an LV upregulation of FGF-23 in the catecholamine groups, and this improved in low-dose propranolol groups. These results suggest catecholamine-induced heart failure initiates early pre-fibrotic pathways through FGF-23 upregulation. Low-dose propranolol exerted cardio-preventative effects through FGF-23 downregulation and hemodynamic-parameter improvement in our model of hyper-acute catecholamine-induced heart failure.


Introduction
Catecholamines are an important mediator during physiological stress. Several critical conditions can cause the elevation of serum catecholamines, from endogenous secretions in the setting of septic shock [1] to exogenous administrations during acute resuscitation. Excessive serum catecholamines may likewise induce heart failure [2]. In animal models of hyperadrenergic states, most models were constructed through single catecholamine use, with a focus upon long-term effects. Effects of chronic exposure to epinephrine (E) include biventricular heart failure and ventricular remodeling [3], while rats infused with continuous norepinephrine (NE) developed left ventricular hypertrophy [4]. Fibroblast growth factor 23 (FGF-23), a novel inducer of cardiac hypertrophy and fibrosis through Toxics 2022, 10, 238 2 of 14 pro-fibrotic gene transcription, is speculated to participate in cardiac remodeling with possible reversibility [5,6].
Although the individual effects of E and NE have been studied in various animals, in vivo models investigating the combined effects of E and NE co-administration upon cardiopulmonary physiology are relatively scarce, even though they are more representative of clinical hyperadrenergic states, including sepsis, chronic heart failure, or the iatrogenic supratherapeutic administration of catecholamines [1,3,7]. On the other hand, prospective analyses of patients in shock provide insight into the real-world effects of combined catecholamines. In cohorts with single E use, higher mortality rates with complications including arrhythmia, lactate acidosis, and cardiac stress were observed than in those with additional NE [8]. A recent meta-analysis examined the clinical use of beta-blockers in patients with sepsis and septic shock, with results revealing an improvement in mortality in 4 out of 6 studies [9]. However, detailed mechanisms of these pathways are still lacking, with few real-world applications of beta-blockade in these scenarios. With more information and mechanistic explanations from the bench-side, clinical utilization may become more common.
We previously demonstrated, in our preceding work, that rat models injected with E and NE displayed significant cardiopulmonary impairment and biventricular dysfunction compared with catecholamine monotherapy, in accordance with previous studies [10]. Therefore, we sought to further investigate the consequences and methods of mitigating hyperadrenergic states through an experimental model of greater clinical relevance. Although beta-blockade is commonly known to be beneficial in experimental models of single catecholaminergic states, the mechanisms and benefits of beta-blockade in realistic hyperacute adrenergic states through combined excess E and NE administration are less explored. We hereby propose a catecholamine-induced acute heart failure model in rats through combined catecholamines, with the aim of alleviating hemodynamic dysregulation through propranolol. We hypothesize that non-selective beta-blockade, through the downregulation of beta-adrenergic receptors in catecholamine excess, improves hemodynamic parameters and pre-fibrotic markers such as FGF-23 [11].

Hemodynamic Data Acquisition
Commercial pressure catheters SPR-513 and SPR-407 were connected to PCU-2000 control units (Millar Inc., Houston, TX, USA) and the PowerLab 35 Series data-acquisition system with LabChart Pro and analyzed with a blood pressure analysis program (ADInstrument Inc., Colorado Springs, CO, USA). The SPR-513 catheter was inserted through the right jugular vein into the right ventricle (RV) and confirmed by the typical RV pressure curve. The SPR-407 catheter was placed in the right carotid artery and advanced into the left ventricle (LV). Biventricular systolic and end-diastolic pressures, heart rates, and other hemodynamic parameters were recorded. The contractility index was calculated according to the formula provided by the PowerLab 35 acquisition system. It is calculated as follows: dP/dt max divided by the pressure (p) at the time of max dP/dt max , where max dP/dt is defined as the steepest slope during the downstroke of the pressure curve.

Examination of Cardiac Congestion and Lung Edema
After euthanizing the rats 6 h post-catecholamine infusion, hearts and lungs were removed and weighed. Heart-to-body weight ratio (%) was calculated by dividing heart weight by the body weight of the rats. The cranial, middle, and caudal lobe of the right lungs were weighed. Lung-to-body weight ratio (%) was calculated for lung edema index. Increased cardiac and lung-to-body weight ratio were used as surrogates of accumulated fluids and indicative of acute heart failure.

Statistical Analysis
The data are expressed as mean ± SD, mean or median. All results were calculated using a nonparametric Kruskal-Wallis test, followed by a Mann-Whitney U-test. A p value of <0.05 was considered statistically significant. Statistical analyses were performed using IBM SPSS Statistics Version 20 software (IBM Corp., Armonk, NY, USA, 2011) and GraphPad Prism version 6.01 for Windows, (GraphPad Software. Inc., San Diego, CA, USA, www. graphpad.com).

Overall Survival, Organ-to-Body Weight Ratios and Serum Cardiac Biomarkers
Hourly survival rates during catecholamine infusion were recorded for all subsets ( Figure 1A). Low-dose propranolol (1 mg/kg) groups had the highest survival rates (100%), equal to the sham group. Groups with higher propranolol doses (15 mg/kg) had decreased survival rates, akin to E and NE only groups. This is consistent with working groups studying propranolol overdose in rat models [14]. Trials of greater propranolol doses (20, 25 mg/kg) were administered, however due to 100% mortality, only low-dose and highdose propranolol were selected for therapeutic investigation. Lung-to-body weight ratios increased significantly in combined catecholamine and high-dose propranolol subsets, with low-dose propranolol groups having similar ratios to the sham groups ( Figure 1B). Heart-to-body weight ratios were not attenuated with propranolol, regardless of dosage ( Figure 1C). Heart rates of low-dose propranolol groups were similar to the sham groups at hour 2, 3, and 6. Only the high-dose propranolol subset exhibited decreased heart rates at hour 2 and 3 ( Figure 1D). Serum NT-proBNP was significantly increased in all three experimental groups, compared to sham groups ( Figure 1E). No significant change in serum troponin T was noted among all groups ( Figure 1E).
The administration of high-dose propranolol caused a marked prolongation of both systolic and diastolic durations, observed at hour 2, 3 and 6 (p < 0.05, compared with E and NE, Figure 3C,D). The contractility index changes of the RV were similar to that of
The administration of high-dose propranolol caused a marked prolongation of both systolic and diastolic durations, observed at hour 2, 3 and 6 (p < 0.05, compared with E and NE, Figure 3C,D). The contractility index changes of the RV were similar to that of LV, with maximal contractility indexes observed at 1 h post-catecholamine administration. A subsequent proportional decrease in the contractility index with low-dose and high-dose propranolol was observed ( Figure 3E).

Acute Cardiac Injury in Histopathology
Cardiac injury was investigated with the immunohistochemistry of connexin 43 (Cx43) for lateralization in cardiomyocytes [15]. Lateralization of Cx43 in the LV lateral wall, ventricular septum, and RV lateral wall was observed after catecholamine infusion, without significant changes in the high-dose and low-dose propranolol treatments (see Figure 4). FGF-23 upregulation was observed after continuous catecholamine infusion. The attenuation of FGF-23 in the LV wall was observed after propranolol administration. Maximal FGF-23 downregulation to near-normalization was observed in the low-dose propranolol groups ( Figure 5). An examination of the klotho protein and reactive oxygen species immunohistochemistry denied significant differences among all groups (see Supplementary Figures S1 and S2, respectively). LV, with maximal contractility indexes observed at 1 h post-catecholamine administration. A subsequent proportional decrease in the contractility index with low-dose and high-dose propranolol was observed ( Figure 3E).

Acute Cardiac Injury in Histopathology
Cardiac injury was investigated with the immunohistochemistry of connexin 43 (Cx43) for lateralization in cardiomyocytes [15]. Lateralization of Cx43 in the LV lateral wall, ventricular septum, and RV lateral wall was observed after catecholamine infusion, Figure 3. Trend graphs of the hemodynamic changes of the right ventricle in rat models with excessive catecholamine infusion alone or with additional propranolol treatment. (A) Right ventricular systolic pressure, and (B) right ventricular end-diastolic pressure, were recorded in rat groups of sham, n = 5; E + NE, n = 6; E + NE (pro1), n = 7; E + NE (pro10), n = 6. (C) Systolic durations were measured in groups of sham, n = 5; E + NE, n = 6; E + NE (pro1), n = 7; E + NE (pro10), n = 5. (D) Diastolic durations were measured in groups of sham, n = 5; E + NE, n = 6; E + NE (pro1), n = 7; E + NE (pro10), n = 6. (E) Contractility indexes were measured in groups of n = 5 for all conditions, except for E + NE (pro10), n = 6. Data were presented as median. Mann-Whitney U test was used for statistical analysis of all panels. a p < 0.05 vs. sham group; b p < 0.05 vs. E + NE group; c p < 0.05 vs. E + NE (pro1) group. E + NE, epinephrine and norepinephrine; pro1, low-dose propranolol 1 mg/kg; pro10, high-dose propranolol 10 mg/kg. without significant changes in the high-dose and low-dose propranolol treatments (see Figure 4). FGF-23 upregulation was observed after continuous catecholamine infusion. The attenuation of FGF-23 in the LV wall was observed after propranolol administration. Maximal FGF-23 downregulation to near-normalization was observed in the low-dose propranolol groups ( Figure 5). An examination of the klotho protein and reactive oxygen species immunohistochemistry denied significant differences among all groups (see Supplementary Figures S1 and S2, respectively).

Lung Injury Mediated through Apoptosis
In pulmonary tissue, a significantly downregulated RAGE and an upregulation of pro-surfactant protein C, Cx43, and HMGB-1 expression were observed 6 h post catecholamine infusion. Low-dose and high-dose propranolol subsets failed to modify the expression of pro-surfactant protein C, Cx43, and HMGB-1. However, RAGE was upregulated after high-dose propranolol treatment ( Figure 6).

Lung Injury Mediated through Apoptosis
In pulmonary tissue, a significantly downregulated RAGE and an upregulation of prosurfactant protein C, Cx43, and HMGB-1 expression were observed 6 h post catecholamine infusion. Low-dose and high-dose propranolol subsets failed to modify the expression of pro-surfactant protein C, Cx43, and HMGB-1. However, RAGE was upregulated after high-dose propranolol treatment ( Figure 6).  E + NE (pro1), and E + NE (pro10) groups. n = 4 in all groups except for the sham group in quantitative analysis of Cx43, n = 3. Data are expressed as mean± SD. Mann-Whitney U test was used for statistical analysis of all panels. a p < 0.05 vs. sham group; b p < 0.05 vs. E + NE group; c p < 0.05 vs. E + NE (pro1) group. Cx43, connexin 43; E + NE, epinephrine and norepinephrine; HMGB1, High mobility group box 1; pro1, low-dose propranolol 1 mg/kg; pro10, high-dose propranolol 10 mg/kg; pro-SpC, pro-surfactant protein C; RAGE, receptor for advanced glycation end products. The scale bars represent 50 µm.

Discussion
Our rat models of combined catecholamine-induced heart failure, observed over 6 h, demonstrated the prevention of cardiac dysfunction with a low-dose of bolus propranolol. To the best of our knowledge, no previous animal studies have attempted to realistically simulate hyperadrenergic states through combined catecholamines. Furthermore, a lowdose propranolol bolus effectively mitigated acute heart failure and prolonged survival. Finally, pathological changes of early cardiac dysfunction may be mediated through FGF-23-dependent mechanisms, with maximal prevention after the administration of lowdose propranolol. Our experiment demonstrates that the deleterious effects of adrenergic stimulation are best prevented through low-dose propranolol (1 mg/kg). Therefore, we propose a catecholamine-induced heart failure model, with the prevention of cardiac dysfunction mediated through low-doses of beta-blockade.
Few studies have reported animal models of combined catecholamine-induced cardiomyopathy. Previous animal models were mostly chronically exposed to either E or NE, leading to ventricular dysfunction [16][17][18]. Chronic high-dose E infusion (7.5 mg/kg/min) in rats induced biventricular ischemia and fibrosis, while continuous NE (0.1 mg/kg/h) infusion induced LV hypertrophy and fibrosis [19]. Yet, despite the plethora of animal models utilizing individual catecholamines for heart failure models, a combination of E and NE has not been broadly adopted for acute heart failure models. However, we believe this catecholamine-induced heart failure model is also compatible with clinical scenarios of hyper-catecholaminergic states, including sepsis [1], chronic heart failure [20], and catecholamine administration during resuscitation [21].
In our preceding study, we investigated the physiological changes of excessive catecholamine infusions in rat models with parameters such as echocardiography and ventricular pressure. Infusions of high-dose E (4.5 µg/kg/min) and NE (6.8 µg/kg/min) over 6 h in rats were observed to cause a thickened interventricular septum and increased left ventricular mass upon echocardiography, reflecting cardiomyocyte hypertrophy. A significantly reduced stroke volume, cardiac output, systolic and diastolic function were all found in combined E and NE infusion groups as well, signifying cardiac dysfunction [10].
Of the hemodynamic parameters measured in our current study, the catecholamineonly subsets reflected similar declining LVSP and LVEDP changes as in our previous study groups, reflecting adverse cardiac hemodynamic changes through initially high cardiac output states with subsequent cardiac dysfunction. After low-dose propranolol treatment, LVSP, LVEDP and the contractility indexes improved significantly, preventing cardiac dysfunction through cardiac function optimization. This is consistent with propranolol's beta-1 antagonism properties, maintaining cardiac output and decreasing the myocardial oxygen demand through decreased heart rate, longer diastolic duration and higher end-diastolic volume. These results provide mechanistic explanations for recent studies investigating the benefits of optimized heart rates and hemodynamics with beta-blockers in hyper-catecholaminergic states [22]. In high-dose propranolol groups, the high mortality rate is likely due to its strong membrane action, potential stabilization properties and over-suppression of the adrenergic system as elevated LVEDP was observed [23]. It is worthwhile to note that at even higher doses of propranolol treatment (15 mg/kg), mortality was higher than the high-dose group and on par with the E-and-NE group. Previous lethal dosing of propranolol in rats was documented at 15 mL/kg, with cause of death due to clinical suppression of the central neurological system through membrane action potential stabilization, and direct cardiac depressant effects such as PR prolongation and AV dissociations [14,24]. However, the rats in previous studies were not pre-treated with combined catecholamines, therefore, rat groups with sequential boluses of combined catecholamines and 15 mg/kg propranolol were included in the initial stage of our study. Further analyses of this group were not performed as their mortality rate was similar to the combined-catecholamine rat group.
The hemodynamic trends of RV were slightly different from those of LV. The catecholamine-alone group had an overall slower right ventricular response than the LV, likely due to ventricular anatomical and physiological differences. The RV has less muscle mass with thinner walls and is therefore acutely sensitive to afterload changes leading to dysfunction [25]. Another reason for different ventricular responses may be the heterogenous distribution of adrenergic receptors between the ventricles; however, previous rat studies suggest that the distribution of beta-adrenoreceptors is not significantly different among ventricles [26].
We further investigated the histopathological changes of propranolol upon the myocardium after catecholamine injury. It was previously observed that with supratherapeutic doses of NE and E, the lateralization of Cx43 through reactive oxidative stress pathways occurs in response to severe stress [10]. In vitro studies of cultured rat cardiomyocytes exposed for 24 h to norepinephrine (1-10,000 nM) also increased their expression of Cx43, while in vivo equivalents showed an association of catecholamine stimulation with enhanced gap junctional intercellular communication [27]. Our results are consistent with other working groups that used isoproterenol, a β1and β2 -adrenoceptor agonist, as a medium to stimulate neonatal cardiomyocytes, with findings of upregulated Cx43 protein in both isoproterenol-only and combined isoproterenol-metoprolol mediums [28]. The lateralization of Cx43 was not improved by either of the propranolol groups within our 6 h observation period, implicating clinical-histopathological incompatibility with early signs of pathological acute cardiac injury despite perceived clinical hemodynamic reversal. To the best of our knowledge, the role of catecholamines and possible treatment effects of beta-blockers upon Cx43 expression has not been well-researched in the literature, and our pathological findings suggest that acute cardiac injury and subsequent pathological compensatory mechanisms are persistent despite propranolol treatment within six hours.
Next, we investigated possible participating molecular pathways in acute hemodynamic changes after adrenergic overstimulation through the staining of FGF-23. Faul et al. demonstrated pathological left ventricular hypertrophy through the direct intraventricular injection of FGF-23 [29], likely through klotho-independent pathways [30]. While there are known associations of elevated FGF-23 and cardiovascular morbidities in chronic kidney diseases in both clinical and experimental studies, less is understood in an acute, hyperadrenergic setting [31,32]. Previous investigations associated FGF-23 with causal myocardial fibrosis during acute cardiac injury [33]. In our study, FGF-23 was significantly upregulated in the LV, and though upregulated in the RV, failed to reach significance. This suggests an early initiation of the pro-fibrotic pathway in bilateral ventricles through FGF-23 upregulation in hyper-catecholaminergic states. It is also worthy of note that an increased expression of FGF-23 in the RV was not previously observed in intramyocardial injections of FGF-23 [29].
Low-dose propranolol was the only treatment group to effectively downregulate FGF-23 in both ventricles, likely due to optimal cardiac stabilization as evidenced by hemodynamic parameters, with high-dose propranolol only effective in the RV. FGF-23 reversal is thus corroborative of the compatibility between ventricular function and histological changes, further evident of the pathological prevention of adverse cardiac dysfunction. FGF-23 upregulation was only significant in the LV lateral wall of the highdose propranolol group, likely secondary to the overt adrenergic suppression of the LV, leading to volume overload with acute decompensated heart failure.
Our present catecholamine-induced heart failure model, while similar to various hyper-adrenergic states, may not be strictly compatible with specific clinical scenarios due to their variability. Another potential limitation is the lack of pressure-volume loop analyses, meaning that detailed measurements such as ejection fraction were not collected. Furthermore, immunohistochemistry could be refined with the identification of beta-selective receptors, and downhill kinases could be included for an investigation of different pathways. In addition, we were unable to exclude the role of inflammation in catecholaminergic acute lung injury due to a lack of relevant markers, and the experiment time of 6 h may not have been long enough for the development of lung apoptosis. Finally, we were unable to determine the precedence of cardiac and pulmonary injury as the accruement of murine tissues was only possible at the end of the 6 h experiment, instead of detailed time points at hour 1 through 6. While cardiac function improved with propranolol administration, the negative findings in histopathological markers including HMGB-1, Cx-43 and prosurfactant apoprotein-C may be limited by the short 6 h experiment period; therefore, the compatibility of clinical and histopathological changes is still unclear.

Conclusions
In conclusion, this experiment demonstrated that a single low-dose propranolol of 1 mg/kg was successful in improving survival and hemodynamics in a combined catecholamine-induced acute heart failure model. The therapeutic effects of propranolol were not dose-dependent, as high-dose and higher-dose propranolol both caused lower survival rates. Myocardial FGF-23 expression, a pro-fibrotic marker, was also downregulated with propranolol treatment, implicating the role of non-selective beta blockers in the prevention of cardiac dysfunction. Low-dose propranolol may therefore be useful as a cardioprotective treatment option in the hyper-acute setting of excess catecholaminergic states with acute heart failure.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/toxics10050238/s1, Figure S1: Klotho expression in the heart tissue 6 h post continuous combined catecholamine infusion alone, or with propranolol treatment, Figure S2: Level of reactive oxygen species in heart tissues 6 h post excess catecholamine infusion alone, or with propranolol treatment.