Activation of the sympathetic nervous system is a hallmark of heart failure (HF) and is positively correlated with disease progression. Catecholaminergic (C1) neurons located in the rostral ventrolateral medulla (RVLM) are known to modulate sympathetic outflow and are hyperactivated in volume overload HF. However, there is no conclusive evidence showing a contribution of RVLM-C1 neurons to the development of cardiac dysfunction in the setting of HF. Therefore, the aim of this study was to determine the role of RVLM-C1 neurons in cardiac autonomic control and deterioration of cardiac function in HF rats. A surgical arteriovenous shunt was created in adult male Sprague-Dawley rats to induce HF. RVLM-C1 neurons were selectively ablated using cell-specific immunotoxin (dopamine-β hydroxylase saporin [DβH-SAP]) and measures of cardiac autonomic tone, function, and arrhythmia incidence were evaluated. Cardiac autonomic imbalance, arrhythmogenesis and cardiac dysfunction were present in HF rats and improved after DβH-SAP toxin treatment. Most importantly, the progressive decline in fractional shortening observed in HF rats was reduced by DβH-SAP toxin. Our results unveil a pivotal role played by RVLM-C1 neurons in cardiac autonomic imbalance, arrhythmogenesis and cardiac dysfunction in volume overload-induced HF.
Heart failure (HF) is highly prevalent in the elderly, affecting approximately 20% of the population over 75 years of age [1,2]. This disease is characterized by impairment of autonomic control [3–7] and progressive decline in cardiac function [8,9]. Early in the development of HF, sympathoexcitation acts as an adaptive response to improve general cardiovascular function  but ultimately becomes maladaptive and contributes to disease progression . Importantly, development of cardiac dysfunction occurs in tandem with sympathetic hyperactivation, and both are inversely correlated with survival in HF [5,11]. Interestingly, it has been shown that pre-sympathetic neurons located in the rostral ventrolateral medulla (RVLM-C1), one of the most important autonomic control regions in the brain , become hyperactivated in HF independent of its etiology [5,13,14]. Despite this evidence, no comprehensive studies have addressed the contribution of RVLM-C1 neurons to cardiac dysautonomia, arrhythmogenesis and/or deterioration of cardiac function in HF.
RVLM-C1 neurons are a subpopulation of catecholaminergic pre-sympathetic neurons that reside in the ventrolateral aspect of the brainstem which play an important role in regulating cardiac autonomic function [12,15–17]. Some evidence suggests that RVLM-C1 neurons contribute to development of cardiac dysfunction in HF [5,13]. Indeed, we recently reported that RVLM-C1 neurons are chronically activated in volume overload HF rats ; however, evidence demonstrating a contribution of RVLM-C1 to HF pathophysiology is still lacking. Accordingly, we hypothesize that RVLM-C1 neurons act as a key nodal point in the brainstem mediating increases in sympathetic drive in HF, contributing to cardiac autonomic imbalance, arrhythmogenesis, and cardiac dysfunction. To address this hypothesis, we selectively ablated RVLM-C1 neurons in volume overload HF rats and assessed cardiac autonomic control, arrhythmia incidence, and cardiac function.
Adult male Sprague-Dawley rats (n=45), weighing initially 250 g, were used in the present study. All experiments were performed 8-weeks following induction of HF (Supplementary Figure S1). In accordance with the American Physiological Society and the National Institutes of Health Guide for the care and use of laboratory animals and the Guía para el Cuidado y Uso de los Animales de Laboratorio from CONICYT, all animals were kept at controlled room temperature under a 12-h light/dark cycle with ad libitum access to food and water. All experimental protocols were approved by the Ethics Committee for Animal Experiments of the Pontificia Universidad Católica de Chile. At the end of the appropriate experiments, all animals were humanely euthanized via anesthetic overdose (sodium pentobarbital 100 mg/kg i.p.).
Rat model of volume overload HF
HF was induced with the surgical creation of an arteriovenous anastomosis (volume overload HF) as described previously [5–7,18]. Briefly, under anaesthesia (Isoflurane: 5% for induction; 1.5% for maintenance balanced with O2), the inferior vena cava and abdominal aorta were exposed by opening the abdominal cavity by a midline incision. Both vessels were clamped caudal to the renal artery and to the aortic bifurcation. An 18-gauge needle was pushed through the aorta until it perforated the adjacent cava vein. Immediately afterward, a drop of hystoacril glue was used to seal the puncture point on the aorta. The A-V was confirmed by visualization of bright red arterial blood entering the vena cava through the anastomosis. The peritoneal cavity was closed with absorbable suture (Novosyn 4/0, Braun) and skin was closed with absorbable suture (Novosyn 3/0, Bbraun) and metallic clips (Kent, U.S.A.). Post-operative management consisted of administration of 5 mg enrofloxacin (s.c.), 1 mg ketoprofen (s.c.), 5 ml saline solution (i.p.), and 2% lidocaine hydrochloride jelly (topical). Sham-operated rats underwent the same anesthesia and surgical procedures without implantation of a shunt.
Four weeks after HF induction surgery, cardiac function was evaluated using transthoracic echocardiography under light isoflurane anesthesia (5% for induction; 1.5% for maintenance balanced with O2). The criteria for HF were an increase in EDV and stroke volume (SV) (≥1.5-fold) relative to Sham without changes in ejection fraction (EF). Transthoracic M-mode echocardiography was recorded for quantification of cardiac dimensions at the level of the mid-papillary muscle with the parasternal short-axis view (SonoaceR3 imaging system, Samsung, U.S.A.). Left ventricle end-systolic diameter (LVESD) and left ventricle end-diastolic diameter (LVEDD) were averaged from three consecutive cardiac cycles according to the guidelines of the American Society of Echocardiography . The left ventricular end-systolic volume (LVESV) and left ventricular end-diastolic volume (LVEDV) were calculated using the Teicholz method ([LVESV = 7*LVESD3/(2.4 + LVESD] and [LVEDV = 7*LVEDD3/(2.4 + LVEDD], respectively) [6,19]. The EF and fractional shortening (FS) were calculated from LV diameters [6,19].
Ablation of RVLM-C1 neurons
At 4 weeks post-HF or Sham surgery, rats underwent a second surgery for selective destruction of catecholaminergic neurons in the RVLM region. The rats were anesthetized with a mixture of ketamine (100 mg/kg of body weight i.p; Ketamil, Troy laboratories, Australia) and xylazine (10 mg/kg of body weight i.p.; Centrovet Ltda, Chile), and then fixed to a stereotaxic frame (model 900; David Kopf Instruments, U.S.A.). Body temperature was maintained at 35°C with a servo-controlled heating pad. Bilateral injections of anti-dopamine-β-hydroxylase conjugated with saporin toxin (DβH-SAP; Advanced Targeting Systems, U.S.A.) were delivered to the RVLM to destroy C1 neurons [15,20]. The DβH-SAP immunotoxin (7.5 ng/100 nl of sterile saline solution) was injected bilaterally into the RVLM (12.36 mm caudal to bregma, 2.3 mm lateral to the midline, and 8.5 mm below the dura matter) , using a Hamilton syringe (500 nl) connected to an injection needle (32-gauge point style 3). The dose of DβH-SAP toxin used for the present study was similar to that used in previous studies . Control rats (vehicle) were injected with sterile saline solution (0.9% of NaCl/100 nl). Post-surgical care included application of enrofloxacin (5% s.c.) and injection of ketoprophen (1% s.c.) for inflammation and pain relief, respectively.
Telemetry implant for blood pressure and HR measurements
Arterial blood pressure (BP) was measured using radio-telemetry (Data Science International, U.S.A.). Six weeks post-HF or Sham surgery, rats were anesthetized (2% isofluorane in O2) and a skin incision was made to expose the femoral artery. The tip of a pressure transducer was guided into the femoral artery and the HD-S10 digital transmitter was placed subcutaneously. After surgery, the rats received a subcutaneous injection of ketoprofen (1%) and enrofloxacin (1%). After 10 days of recovery BP recordings were initiated. BP was recorded for 2 h/day (sampling rate of 1000 Hz) between 10 a.m. and 12 p.m. Heart rate (HR) was derived from the dP/dt signal obtained from the BP recordings [5,6]. Systolic BP (SBP), diastolic BP (DBP), pulse pressure (PP), and mean arterial BP (MABP) were calculated from the arterial pressure signal.
Baroreflex control of the HR
Spontaneous baroreflex sensitivity (BRS) was calculated using the sequence method (Hemolab Software Suite 20.0 for 64 bits, http://www.haraldstauss.com/HemoLab/HemoLab.php) as previously described [4,7,13–15,18–20,22–24]. Using the SBP and calculated HR from the BP signal, we searched for sequences where the SBP decreased and HR increased (BRS down sequences) and where the SBP increased and the HR decreased (BRS up sequences). In addition, we determined the total gain, which is calculated as the average between BRS down and up sequences. The data analyzed correspond to 10 min segments taken at random from 2 h of continuous recordings. Only sequences with a correlation coefficient r > 0.80 were used for analysis [5,6].
Cardiac autonomic control
Cardiac sympatho-vagal balance was evaluated by quantifying the HR responses to injections of propranolol (1 mg/kg i.p.) and atropine (1 mg/kg i.p.), respectively [5,7]. Changes in HR in response to propranolol were used as an indicator of sympathetic tone and HR responses to atropine were used as an indicator of parasympathetic tone. Change in HR was expressed relative to baseline values (ΔHR).
Tachograms were constructed from the dP/dt of the arterial pressure signal over 1 h of recording. Irregular heartbeats were visually inspected, and arrhythmic episodes were defined as premature or delayed beats with changes greater than 2.5 standard deviations from the mean beat-to-beat interval duration as previously described [5,6,13,22,24]. Movement artifacts were visually inspected and eliminated from HR time series. All events meeting the stated criteria were identified and recorded to determine an index of events per hour (events/hour) [5,6,13,22,24].
Cardiac function assessment through invasive hemodynamic measurements
Following physiological experiments performed in conscious unrestrained animals, invasive measurements of cardiac function were made under anaesthesia. Briefly, rats were anesthetized with a mixture of urethane and α-chloralose (800 and 40 mg/kg, respectively) and intubated with a 16-gauge cannula. The abdominal cavity was opened first, to visualize the arteriovenous anastomosis (volume overload), and then to perform venous occlusion for the measurement of cardiac function. A conductance catheter transducer (SPR-869, Millar Instruments, U.S.A.) was introduced into the left ventricle (LV) through the right carotid artery as previously described [5,6]. Pressure–volume (PV) loops were recorded (Powerlab, ADInstruments) after 30 min of signal stabilization. Ten to fifteen consecutive stable PV loops were used to calculate the following parameters: left ventricular end-systolic pressure (LVESP), left ventricular end-diastolic pressure (LVEDP), SV; cardiac output, stroke work (SW), arterial elastance (Ea), and EF. Load-dependent parameters of cardiac function were determined by means of the calculation of dV/dtmax, dV/dtmin, dP/dtmax, and dP/dtmin. Load-independent parameters of cardiac function (end-systolic pressure–volume relationship (ESPVR) and end-diastolic pressure–volume relationship (EDPVR) were obtained from transient occlusions of the inferior vena cava with adjustment of data to linear and exponential function, respectively [5,6]. Measurements were calibrated by injecting a hypertonic saline bolus (30% wt/vol. NaCl) to determine conductance, and relative volume units were converted to absolute units using the cuvette calibration method [5,6]. Data were analyzed using the PV loop module of the LabChart7 Pro v7.2 software (ADInstruments).
After completing all physiological experiments, rats were deeply anesthetized with urethane (40 mg/kg i.v.) and perfused through the ascending aorta with 300 ml of saline solution (NaCl 0.9%) followed by 4% phosphate-buffered (0.1 M; pH 7.4) paraformaldehyde (PFA) (MERK, Germany). The brain was removed and stored in the perfusion fixative for 3 h at room temperature and later maintained in 30% sucrose for cryopreservation. All histochemical procedures were done using free-floating sections according to previously described protocols [25,26]. Tyrosine hydroxylase (TH) and phox2b were detected with mouse and rabbit antibodies, respectively (TH: 1:2000 dilution; Chemicon, Temecula, CA, U.S.A.; phox2b: 1:800 dilution; gift from J.-F. Brunet, Ecole NormaleSuperieure, Paris, France). This primary antibody was detected by incubation with appropriate secondary antibody tagged with fluorescent reporters to reveal TH (goat anti-mouse Alexa 488; Invitrogen, Carlsbad, CA, U.S.A.) or phox2b (donkey anti-rabbit Alexa 594; Jackson, U.S.A.).
Rats were perfused with 0.9% saline solution and then fixed with 4% PFA in 0.2 M phosphate buffer (Merck, Germany) for 15 min, followed by incubation with 4% PFA at 4°C for 1 week. Later, heart tissues were dehydrated and embedded in paraffin (Paraplast, Thermo). To quantify cardiac fibrosis, 5 µm thickness cross-sections of the heart were stained with Picrosirius red stain (Diapath, Italy). Photomicrographs of 400× magnification were captured using a CCD camera coupled to an Olympus CX31 microscope (Olympus Corp, U.S.A.). Five images from each animal were analyzed using color deconvolution plugins for ImageJ 1.51n software (NIH, Betheseda, MD, U.S.A.). Percent fibrosis was expressed as the ratio of fibrotic area versus total LV area [25,26].
Cell counting, imaging, and data analysis
A multifunction microscope, Zeiss Axioimager A1 microscope (Zeiss, Muenchen, Germany), was used to image the sections and perform subsequent analysis. Immunofluorescence was examined under epifluorescence illumination. The locations of TH (TH-ir) in the ventrolateral medulla were plotted in sections from 11.12 to 14.96 mm caudal to bregma (17 sections/animal), with the RVLM located between 11.60 and 12.80 mm caudal to bregma (6 sections/animal). The locations of TH in the A5 and A6 region (ventrolateral and dorsolateral pons, respectively) were plotted in sections from 9.00 to 10.68 mm caudal to bregma (8 sections/animal). The profile counts of the animals that received bilateral microinjections of anti-DβH-SAP reflected the sum of both sides of the brainstem and were compared with the control rats. Digital color photomicrographs were acquired using a Zeiss AxiocamHRc camera. Image J (version 1.41; National Institutes of Health, Bethesda, MD) was used for cell counting, and Canvas software (ACD Systems, Victoria, Canada, v. 9.0) was used for line drawings. The neuroanatomical nomenclature employed during experimentation and in this manuscript was defined by Paxinos and Watson .
After euthanizing the rats (n=4 per groups), hearts were removed and quickly frozen on dry ice and stored at −80°C. LV tissue samples were lysed in RIPA buffer containing 1% protease inhibitor cocktail (Sigma–Aldrich, U.S.A.) plus 1% phosphatases inhibitors (Phosphostop® Sigma, U.S.A.). Protein concentration was determined using a BCA protein assay kit (Thermo Scientific, U.S.A.). 50 μg of protein were separated by SDS/PAGE under reducing conditions on gels of 8 and 15% acrylamide-bisacrylamide gels to identify MMP-2 and TIMP-2, respectively. Then, the proteins were transferred to a PVDF membrane (Millipore), using the wet trans-blot system (Biorad, U.S.A.). The membranes were incubated in blocking solution, 5% non-fat milk diluted in Tris-saline buffer (TBS: 20 Mm Tris base, pH 7.2, 0.3 M NaCl), for 1 h at room temperature in constant agitation. Subsequently, the membranes were incubated with primary antibodies, MMP-2 (NB200-193, Novus) and TIMP-2 (NB-100-92000, Novus) at a dilution of 1:1000 in blocking solution at 4°C overnight. Afterward, the membranes were washed with TBS with 0.1% tween-20 and followed by incubation with secondary antibody conjugated with HRP anti-rabbit (074-1506, KPL) at a dilution of 1:2000 in blocking solution plus 0.1% tween-20 at room temperature for 1 h. Later, the membranes were incubated with restore western blot stripping buffer (21059, Thermofisher) according to manufacturer instructions. Then, the membranes were incubated in blocking solution (5% non-fat milk in Tris-saline buffer) for 1 h at room temperature at constant agitation. Subsequently, the membranes were incubated at 4°C overnight with loading control antibodies diluted in the same blocking solution (mouse anti-HSP-90, 1: 3000, TA150046-HSP90AB1, Origene; mouse anti-GADPH, 1:2000, sc-32233, SantaCruz, for 8 and 15% gel, respectively). Membranes were then washed in cold buffer solution and incubated with HRP conjugated anti-mouse antibody (1:2000, 074-1806, KPL). Proteins were visualized using the ECL chemiluminescent solution (Pierce, U.S.A.) and exposed to high sensitivity film (BioMax mitral regurgitation [MR] Film of Carestream, Kodak). Finally, the signal intensities were quantified with the software image studio lite version 5.2 (Li-Cor Bioscience, U.S.A.). The MMP-2/TIMP-2 ratio was calculated based on the signal intensities of both proteins.
Data were evaluated using One-way ANOVA parametric test, followed by a Holm–Sidak post hoc analysis. In addition, Two-way ANOVA, followed by Holm–Sidak post hoc analysis was also used according to data structure. The level of significance was defined as P<0.05. Results were presented as mean ± standard error (SEM). All the statistical analysis was performed with GraphPad Prism 7.0 software (U.S.A.).
Ablation of RVLM-C1 neurons restores normal cardiac autonomic control in volume overload HF rats
DβH-SAP treatment induced a 1.9-fold decrease in the number of TH-positive C1 neurons in the RVLM region in both Sham and HF rats (Figure 1, Supplementary Figure S2). Selectivity of the lesion was assessed by comparing TH-positive neuron viability in regions closest to the RVLM. Neurons located in A5 (−9.00–10.44 mm) and A6 (−9.72–10.68 mm) regions, showed no decrease in the number of TH-positive neurons following DβH-SAP injections (Supplementary Figure S3). DβH-SAP toxin treatment did not modify hemodynamic parameters in HF rats (Table 1).
Immuno-based selective ablation of RVLM-C1 neurons
|Sham+Veh (n=8)||Sham+DβH-SAP (n=6)||HF+Veh (n=8)||HF+DβH-SAP (n=8)|
|SBP (mmHg)||117.0 ± 5.2||118.5 ± 1.3||93.7 ± 9.4*+||110.4 ± 2.3|
|DBP (mmHg)||82.8 ± 3.3||83.0 ± 1.9||65.8 ± 6.3*+||76.2 ± 2.3|
|PP (mmHg)||34.3 ± 2.4||35.5 ± 1.6||27.8 ± 3.4*+||34.2 ± 2.6|
|HR (bpm)||295.3 ± 3.5||290.2 ± 11.7||303.1 ± 9.9||286.4 ± 6.4|
|Sham+Veh (n=8)||Sham+DβH-SAP (n=6)||HF+Veh (n=8)||HF+DβH-SAP (n=8)|
|SBP (mmHg)||117.0 ± 5.2||118.5 ± 1.3||93.7 ± 9.4*+||110.4 ± 2.3|
|DBP (mmHg)||82.8 ± 3.3||83.0 ± 1.9||65.8 ± 6.3*+||76.2 ± 2.3|
|PP (mmHg)||34.3 ± 2.4||35.5 ± 1.6||27.8 ± 3.4*+||34.2 ± 2.6|
|HR (bpm)||295.3 ± 3.5||290.2 ± 11.7||303.1 ± 9.9||286.4 ± 6.4|
Values are mean ± S.E.M. One-way ANOVA followed by Holm–Sidak posthoc analysis. *, P<0.01 vs. Sham+Veh; +, P<0.01 vs. Sham+DβH-SAP.
HF+Veh rats had increased cardiac sympathetic drive compared with Sham+Veh rats (ΔHR to propranolol: −96.0 ± 7.8 vs. −26.7 ± 3.9 beats/min, HF+Veh vs. Sham+Veh, respectively, Figure 2A,B). In HF rats, DβH-SAP toxin treatment resulted in a significant (P<0.05) reduction in cardiac sympathetic drive (ΔHR to propranolol: −50.3 ± 7.3 vs. −96.0 ± 7.8 beats/min, HF+DβH-SAP vs. HF+Veh, respectively, Figure 2B). No effect on cardiac sympathetic tone was found in Sham rats treated with DβH-SAP toxin (Figure 2A,B). In addition, HF+Veh rats displayed reduced parasympathetic tone compared with Sham+Veh group (Atropine ΔHR: 42.3 ± 7.0 vs. 80.9 ± 12.6 beats/min, HF+Veh vs. Sham+Veh, respectively, Figure 2A,C). DβH+SAP toxin treatment had no effect on parasympathetic tone in Sham or HF rats (Figure 2C). Additionally, we do not find significant differences in SBP, DBP, PP, and MABP after propranolol and atropine between all experimental groups (Supplementary Table S1).
RVLM-C1 neuron ablation restores cardiac autonomic control in HF rats
Ablation of RVLM-C1 neurons improves baroreflex function in volume overload HF rats
Compared with Sham animals, HF rats had significantly (P<0.05) reduced spontaneous BRS (2.2 ± 0.3 vs. 4.1 ± 0.2 mmHg/ms, HF+Veh vs. Sham+Veh, respectively) (Supplementary Figure S4). After selective ablation of RVLM-C1 neurons BRS was significantly improved in HF rats (4.2 ± 0.7 vs. 2.2 ± 0.4 mmHg/ms, HF+DβH-SAP vs. HF+Veh, respectively, P<0.05) (Supplementary Figure S4). Ablation of RVLM-C1 neurons in Sham rats had no effect on BRS (Supplementary Figure S4). The number of total BRS sequences were not statistically different between all experimental conditions (Supplementary Table S2).
Progression of cardiac dysfunction in volume overload HF is partially dependent on RVLM-C1 neurons
DβH-SAP treatment had no effect on LVEDD or LVESD in HF animals (Supplementary Table S3). However, DβH-SAP treatment attenuated declines in FS in HF rats compared with the untreated HF group (Figure 3). Indeed, HF+Veh rats displayed reductions in FS between the 4th and 8th week post-HF induction (59.3 ± 5.1 vs. 45.0 ± 1.4%, HF+Veh 4 weeks vs. HF+Veh 8 weeks, respectively) (Figure 3A,B). DβH-SAP markedly reduced this decline in FS in HF rats (57.2 ± 4.7 vs. 51.0 ± 4.8%, HF+DβH-SAP 4 weeks vs. HF+DβH-SAP 8 weeks, respectively) (Figure 3A,B). No significant changes in left ventricular ejection fraction (LVEF) were observed between the 4th and 8th week post-HF induction and/or following DβH-SAP treatment (Figure 3C).
Ablation of RVLM-C1 neurons attenuates the progression of cardiac dysfunction in HF
Ablation of RVLM-C1 neurons improves cardiac function in volume overload HF rats
Cardiac function was evaluated through intraventricular PV loops. Compared with Sham+Veh treated rats, HF+Veh rats showed significant impairment in both diastolic (measured by the EDPVR: 0.009 ± 0.001 vs. 0.003 ± 0.0008 mmHg/μl, HF+Veh vs. Sham+Veh rats, respectively, Figure 4A,B) and systolic cardiac function (measured by the ESPVR: 0.26 ± 0.03 vs. 0.51 ± 0.07 mmHg/μl, HF+Veh vs. Sham+Veh rats, respectively, Figure 4A,C). DβH-SAP toxin treatment significantly improved cardiac function in HF rats (p<0.05). We observed approximately 3-fold reduction in EDPVR and approximately 2.2-fold increase in ESPVR in HF+DβH-SAP rats compared with HF+Veh rats (Figure 4A–C). No significant changes in any other hemodynamic parameter were observed in HF animals treated with DβH-SAP toxin (Supplementary Table S4). Considering that autonomic control and cardiac function were improved, we performed correlation analysis between EDPVR and ESPVR vs. ΔHR during propranolol test. We did not find a significant correlation; however, there is a tendency to observe greater deterioration of cardiac systolic function with greater autonomic dysregulation, which was abolished by DβH-SAP toxin treatment (Supplementary Figure S5).
RVLM-C1 neuron ablation restores cardiac function in HF rats
Cardiac arrhythmogenesis in volume overload HF is dependent on RVLM-C1 neurons
HF rats displayed a large increase in the number of cardiac arrhythmias compared with Sham rats (71.5 ± 15.6 vs. 12.6 ± 1.9 events/hour, HF+Veh vs. Sham+Veh rats, respectively, Figure 5A,B) which were significantly reduced by DβH-SAP toxin treatment (71.5 ± 15.6 vs. 34.8 ± 8.8 events/hour, HF+Veh vs. HF+DβH-SAP rats, respectively, Figure 5A,B). No significant differences in the number of cardiac arrhythmias were found between Sham+Veh and Sham+DβH-SAP rats (Figure 5).
Reduction of cardiac arrhythmias following selective ablation of RVLM-C1 is independent of cardiac hypertrophy and cardiac remodeling in HF rats
To determine if the improvements in cardiac function and cardiac arrhythmogenesis following DβH-SAP treatment in HF rats were secondary to changes in cardiac hypertrophy and/or cardiac remodeling, we examined LV collagen content and calculated a hypertrophy index. Accordingly, LV collagen content and LV expression of metalloprotease and its inhibitor were quantified (Figure 5C,E). HF+Veh rats displayed overt signs of cardiac hypertrophy and tissue fibrosis (Figure 5C,D) compared to Sham+Veh rats and this was not improved by DβH-SAP toxin treatment (7.8 ± 0.7 vs. 7.7 ± 0.7% of tissue, HF+Veh vs. HF+DβH-SAP rats, respectively, Figure 5E). Also, DβH-SAP did not change the expression ratio of MMP-2/TIMP-2 in HF rats (1.1 ± 0.1 vs. 1.1 ± 0.2 MMP-2/TIMP-2 ratio, HF+Veh vs. HF+DβH-SAP rats, respectively, Supplementary Figure S6). Finally, cardiac hypertrophy index was not reduced by DβH-SAP in HF rats (Supplementary Figure S7).
The major findings of this study are as follows: (i) selective ablation of C1 neurons in rats with volume overload HF resulted in marked reductions in cardiac sympathetic tone (i.e. ΔHR during propranolol test) and BRS improvement; (ii) the progression of cardiac dysfunction (FS, before vs. after DβH-SAP toxin) was reduced by ablation of RVLM-C1 neurons; (iii) cardiac dysfunction observed in HF rats was improved by RVLM-C1 neurons ablation; (iv) the incidence of cardiac arrhythmias was reduced by ablation of RVLM-C1; and (v) improvements in cardiac function and arrhythmogenesis after ablation of RVLM-C1 neurons are independent of cardiac hypertrophy and cardiac remodeling in HF rats. Our findings demonstrate that the progression of volume overload HF and impairments in cardiac autonomic control, arrhythmogenesis (i.e. arrhythmia index), and dysfunction are critically dependent on RVLM-C1 neurons integrity.
Role of C1 neurons in sympathoexcitation in volume overload HF rats
Sympathoexcitation is a pathophysiological hallmark of HF and is strongly associated with disease severity, prognosis, and mortality . We have previously shown that volume overload HF in rats mimics several important aspects of human HF including cardiac sympathoexcitation, reduced BRS, arrhythmogenesis, impairment of cardiac function, cardiac hypertrophy, and cardiac remodeling [5–7]. In the present study, we showed that RVLM-C1 neurons are critical to induce/maintain the heightened cardiac sympathetic drive in volume overload HF rats. While numerous regions in the central nervous system have been implicated in the development of autonomic imbalance (i.e. paraventricular nucleus [PVN], nucleus of the solitary tract [NTS]), our results strongly suggest that RVLM-C1 neurons play a critical role in this process in volume overload HF. Previous evidence suggests that RVLM-C1 neurons might serve as key nodal points of sympathetic integration for regulation of cardiovascular function ; however, the precise contribution of RVLM-C1 neurons to autonomic dysregulation in volume overload HF has not been previously addressed. Our study provides comprehensive evidence of a role for RVLM-C1 neurons in the pathophysiology of volume overload HF. Previous work supports the importance of the NTS-PVN-RVLM pathway in HF progression . The neural circuitry of this axis is such that PVN and NTS neurons are reciprocally ‘wired’, and PVN-NTS neurons send glutamatergic projections to the RVLM. Thus, it is reasonable to posit that inputs from the NTS-PVN in volume overload HF are integrated in the RVLM by C1 neurons and orchestrate cardiac sympathetic outflow.
We have previously shown that HF rats have decreased BRS compared with Sham animals . In the present study, we showed that baroreflex impairment in HF was improved by ablation of RVLM-C1 neurons. Importantly, it has been proposed that decreases in baroreflex gain are strongly associated with increased mortality rate in HF patients . Indeed, patients with HF that displayed <3 mmHg/ms of BRS gain have lower probability of survival compared with the patients that exhibit >3 mmHg/ms of BRS gain . It is worth noting that DβH-SAP toxin treatment improved BRS gain from a high-risk zone (∼2.2 mmHg/ms) to low-risk zone (∼4.2 mmHg/ms) in HF animals. Our results showed that ablation of C1 neurons improve BRS in HF; however, it should be noted that BRS can be modulated by sleep-state [30,31]. Indeed, baroreflex gain is higher in rapid eye movement (REM) compared with the non-REM sleep , and it has been proposed that C1 neurons project to orexinergic neurons contributing to arousal state . The possibility that ablation of C1 neurons in the present study affected BRS via sleep-state/arousal-related mechanisms merits consideration, but requires further investigation.
Influence of C1 neurons on cardiac function in volume overload HF rats
Human HF is characterized by diastolic and systolic dysfunction [33,34], and we observed similar changes in cardiac function in the present study using an animal model of volume overload HF. These findings confirm our previous studies showing that rats with HF induced by volume overload display systolic and diastolic dysfunction [5,6]. Previously, Toledo et al. (2017)  showed that diastolic cardiac dysfunction in volume overload HF was improved after acute β-blocker treatment. In the present study, we demonstrated that selective ablation of RVLM-C1 neurons with DβH-SAP toxin significantly improved diastolic function in HF. Taken together these results strongly suggest that alterations in passive cardiac properties in the setting of volume overload HF may be linked to sympathoexcitation and that cardiac dysfunction is critically dependent on RVLM-C1 neurons.
It is important to note that our experimental HF paradigm is based on a physical increase in cardiac pre-load (similar to MR)  by means of the creation of a surgical anastomosis between two major blood vessels (see Data Supplement for complete description of the methods). Thus, it was not expected that targeting C1 neurons would decrease volume overload or changes in cardiac morphological parameters (i.e. LVEDV, SV, hypertrophy index) following RVLM-C1 ablation in HF rats. Nevertheless, RVLM-C1 neuron ablation significantly delayed the progression of cardiac dysfunction since DβH-SAP toxin treatment reduced the progressive decline in FS observed in HF rats. Whether ablation of RVLM-C1 neurons in volume overload HF rats will improve echocardiography measurements and cardiac hypertrophy index in the long-term remains to be determined.
C1 neurons and cardiac arrhythmogenesis in volume overload HF
One of the many important aspects of sympathetic hyperactivity in HF is its potential contribution to cardiac arrhythmogenesis and related mortality (i.e. sudden cardiac death) [35,36]. In our study, RVLM-C1 neuron ablation resulted in a reduction in arrhythmia incidence in HF animals, an effect we attribute to restoration of normal cardiac autonomic balance, since elimination of RVLM-C1 neurons had no effect on other arrhythmogenic substrates (cardiac collagen content and hypertrophy). Even though normalizing cardiac sympathetic tone did not significantly prevent pathological cardiac remodeling in HF rats, it is worth noting that cardiac function was improved by deletion of RVLM-C1 cells.
In our experimental design, animals received bilateral injections of the anti-DβH-SAP toxin or saline (control) directly into the RVLM-C1 region with the aim of eliminating catecholaminergic neurons. Many studies have investigated the role of the C1 neurons in sympathetic reflexes and cardiovascular regulation using intraspinal injections of the anti-DβH-SAP toxin to retrogradely destroy C1 neurons [39–41]. In many aspects, the results from these studies are quite similar, but the lesions are somewhat different. Schreihofer and Guyenet (2000) , showed that after 3–5 weeks of intraspinal DβH-SAP toxin injection, the number of C1 neurons significantly decreased but that this approach also results in the partial elimination of neurons that resides on C3, A5 and A6 regions. Importantly, our injections directly performed on the rostral aspect of the medulla produced a significant lesion in catecholaminergic neurons in the C1 region, but spared the A5 and A6 catecholaminergic neurons in the brainstem [40,41]. It is well described that injections of anti-DβH-SAP effectively eliminate catecholaminergic C1 neurons without affecting TrpOH-expressing neurons in the midline raphe or astrocytes in the RVLM-C1 region.
Also important, it has been shown that C1 neurons send glutamatergic projections to the locus coeruleus and A5 region. Even though our results showed no reductions in the cell count in the A5 region after DβH-SAP toxin injection, it is plausible that this intervention may result in a partial loss of glutamatergic drive to other regions that contribute to sympathetic regulation. Therefore, it is important to consider that partial reduction of RVLM-C1 neurons may trigger an overall loss of glutamatergic drive to several sympathetic control areas within the brainstem. It is also important to point out that the dose of the toxin used in the present study was the same as the dose used in previous publications showing that the anti-DβH-SAP toxin reduced the number of TH cells without affecting the number of Phox2b- or choline acetyltransferase-expressing neurons [40–43].
It is important to emphasize that our experimental model of HF recapitulates some but not all the pathophysiological characteristics of human HF associated with MR and aortic regurgitation [37,38]. HF induced by volume overload results in a deterioration of passive properties of the heart (EDPVR and LVEDP) with preserved EF, both of which are considered diagnostic criteria for human HF. In addition to the nature of the model itself, the time course of our experiments was limited to 8 weeks post-HF induction. Thus, it is not possible to infer the long-term effects of ablation of RVLM-C1 neurons on cardiac physiology or survival in HF rats from these studies. Future investigations should address this important translational aspect. Finally, we measured baroreflex gain by means of the sequence method, which is only an indirect measure of baroreflex gain. The sequence method is limited to estimations of overall sensitivity rather than the precise contribution of each arm of the baroreflex. To characterize sympathetic/parasympathetic contributions to cardiac baroreflex control further studies are needed.
Summary and conclusion
The role of sympathoexcitation in HF pathophysiology has been extensively studied. However, the precise contribution of pre-sympathetic RVLM-C1 neurons, a key area related to autonomic control in the brain, in the progression of HF had not been previously explored. In the present study, we demonstrate for the first time that cardiac autonomic dysfunction, arrhythmogenesis, and cardiac dysfunction in volume overload HF are all significantly improved by selective ablation of RVLM-C1 neurons. In addition, our data show that ablation of RVLM-C1 neurons delayed declines in FS in HF animals. A comprehensive listing of the physiological effects of ablation of RVLM-C1 neurons in HF rats is provided in the Supplementary Table S5. In conclusion, pre-sympathetic C1 neurons of the RVLM play a pivotal role in volume overload HF pathophysiology. Whether immuno-based targeted ablation of C1 neurons is a feasible treatment for human HF requires future investigation.
The definite contribution of RVLM-C1 neurons to altered autonomic function in HF has not been previously explored.
Ablation of pre-sympathetic neurons from RVLM-C1 is sufficient to normalize cardiac autonomic control, arrhythmogenesis and cardiac function in volume overload HF.
Targeting neural components of autonomic regulation may have salutary effects in HF.
We thank Paulina Arias for her technical assistance and Fidel Flores for his help in managing the animal facility.
D.C.A. and C.T. performed data collection and analysis, performed interpretation of the data and contributed to the preparation of the manuscript. H.S.D., C.L., and A.A.A. performed data collection and analysis, and contributed to the preparation of the manuscript. L.M.O. performed data analysis. A.C.T., T.S.M., H.D.S., N.J.M., and J.A. performed interpretation of the data and contributed to the preparation of the manuscript. R.D.R. contributed to the concept of the project and experimental design. R.D.R. performed interpretation of the data and contributed to the preparation of the manuscript. All data collection was undertaken in the laboratory of R.D.R. All authors approved the final version of the manuscript.
This work was supported by FONDECYT [grant number: 1180172] from the National Fund for Scientific and Technological Development of Chile. This work was also supported by the São Paulo Research Foundation [FAPESP; grants numbers: 2016/23281-3 to ACT and 2015/23376-1 to TSM].
The authors declare that there are no competing interests associated with the manuscript.
anti-dopamine-β-hydroxylase conjugated with saporin toxin
end-diastolic pressure-volume relationship
end-systolic pressure-volume relationship
LV end-diastolic diameter
LV end-systolic diameter
LV end-diastolic volume
LV end-diastolic pressure
LV end-systolic pressure
LV end-systolic volume
mean arterial BP
nucleus of the solitary tract
rapid eye movement
rostral ventrolateral medulla
These authors contributed equally to this work.