Abstract
Renal nerves play a critical role in cardiorenal interactions. Renal denervation (RDN) improved survival in some experimental heart failure (HF) models. It is not known whether these favorable effects are indirect, explainable by a decrease in vascular afterload, or diminished neurohumoral response in the kidneys, or whether RDN procedure per se has direct myocardial effects in the failing heart. To elucidate mechanisms how RDN affects failing heart, we studied load-independent indexes of ventricular function, gene markers of myocardial remodeling, and cardiac sympathetic signaling in HF, induced by chronic volume overload (aorto-caval fistula, ACF) of Ren2 transgenic rats. Volume overload by ACF led to left ventricular (LV) hypertrophy and dysfunction, myocardial remodeling (upregulated Nppa, MYH 7/6 genes), increased renal and circulating norepinephrine (NE), reduced myocardial NE content, increased monoaminoxidase A (MAO-A), ROS production and decreased tyrosine hydroxylase (+) nerve staining. RDN in HF animals decreased congestion in the lungs and the liver, improved load-independent cardiac function (Ees, PRSW, Ees/Ea ratio), without affecting arterial elastance or LV pressure, reduced adverse myocardial remodeling (Myh 7/6, collagen I/III ratio), decreased myocardial MAO-A and inhibited renal neprilysin activity. RDN increased myocardial expression of acetylcholinesterase (Ache) and muscarinic receptors (Chrm2), decreased circulating and renal NE, but increased myocardial NE content, restoring so autonomic control of the heart. These changes likely explain improvements in survival after RDN in this model. The results suggest that RDN has remote, load-independent and favorable intrinsic myocardial effects in the failing heart. RDN therefore could be a useful therapeutic strategy in HF.
Results
Weights, cardiac dimensions and principal LV hemodynamics
Table 1 shows organ weights and hemodynamics in the sham-operated control group, in a group with HF induced by ACF, and in a group with ACF and RDN. The sham/RDN group compared to the sham/intact group displayed no significant changes in organ weight parameters but significantly decreased end-systolic pressure (149 ± 3.8 vs. 169 ± 3.7 mmHg, p < 0.05) and mean LV pressure (65 ± 2.2 vs. 75.4 ± 3.7 mmHg, p < 0.05).
ACF had an impact on multiple organ weight parameters that are typically changed in HF, with no effect on the body weight or tibia length (not shown). ACF/intact rats had increased heart weight and LV weight. Similarly, compared to the sham group, ACF/intact rats had significantly increased weight of the left atrium (LA) and weight of the lungs, reflecting HF-related congestion. Compared to the sham/intact group, ACF/intact rats had significantly increased stroke volume, stroke work, and cardiac output. End-systolic and end-diastolic pressure (EDP) measured by PV analysis were increased in ACF/intact group compared to the sham/intact group, similar to echocardiographic measurements (Fig. 1b, c). Moreover, ACF rats had also a significant decrease in end-systolic pressure (143 ± 3.1 vs. 169 ± 3.7 mmHg, p < 0.05) compared to sham rats.
Fig. 1
figure 1
In vivo measurement of LV contractility and dimensions. a Representative pressure-volume loops from invasive pressure-volume analysis. Red line—end-systolic elastance (Ees), blue line—end-diastolic pressure-volume relationship (EDPVR). b Echocardiographic M mode images of parasternal long axis view. LV AWd left ventricular anterior wall thickness in diastole, LV AWs left ventricular anterior wall thickness in systole, LVIDd left ventricular internal diameter in diastole, LVIDs left ventricular internal diameter in systole, LV PWd left ventricular posterior wall thickness in diastole, LV PWs left ventricular posterior wall thickness in systole. c Diameter of left ventricle in systole (LVIDs) and diastole (LVIDd) measured during each week of experiment (3 weeks); FS fractional shortening. N = 10 in sham/intact, N = 19 in ACF/intact, N = 13 in ACF/RDN. ###p < 0.001; ##p < 0.01; #p < 0.05, ACF/intact vs. ACF/RDN group, compared to the day 14
Compared to intact ACF, RDN significantly decreased heart weight, LA, LV weight, and congestion of the lungs and liver. RDN in ACF rats significantly decreased stroke work and normalized stroke volume and cardiac output. RDN in ACF rats also decreased dilatation of LV (Fig. 1b, c), which was shown as reduced LV end-systolic and end-diastolic volumes. We observed that ACF/RDN group had also reduced EDP (8.2 ± 0.69 vs. 12.7 ± 1.63 mmHg, p < 0.05) compared to ACF intact group. Heart rate was not affected by ACF or RDN in any groups.
LV function and HF markers: the impact of ACF
ACF/intact group had significantly decreased systolic function compared to the sham/intact group. ACF/intact group had also decreased end-systolic elastance (Ees) and preload recruitable stroke work (PRSW) compared to the sham/intact group. ACF/intact had also decreased ventricular-arterial coupling compared to sham/intact (Ees/Ea ratio, Fig. 2a).
Fig. 2
figure 2
LV function and gene expression of selected HF markers and the impact of RDN. a Systolic function parameters measured by invasive PV analysis, Ees/Ea, ventricular-arterial coupling ratio. b Gene expression of markers of fibrosis—collagen I/III (Col1a1/Col3a1) ratio, myocardial stress—Myosin heavy chain 7/6 (Myh 7/6) ratio, natriuretic peptide A (Nppa) and mitochondrial fatty acid beta-oxidation pathway, acyl-CoA dehydrogenase medium chain (Acadm). N = 9 in sham/intact, N = 9 in ACF/intact, N = 10 in ACF/RDN
ACF/intact group had extensive upregulation of markers of myocardial damage/remodeling compared to the sham/intact group. ACF/intact group had increased fibrotic marker collagen I/III (Col1a1/Col3a1) gene expression ratio. Similarly, maladaptive hypertrophy markers myosin heavy chain isotype ratio (Myh 7/6) and myocardial stress gene natriuretic peptide A (Nppa) were increased in ACF/intact group compared to the sham/intact group. ACF/intact group had a significantly decreased (p = 0.05) medium-chain acyl-Coa dehydrogenase (Acadm, Fig. 2b).
Cardiac autonomic nervous system: the impact of ACF
ACF rats had significantly increased NE levels in plasma and kidney (Fig. 3a, b), but depleted LV content of NE compared to the sham group (Fig. 3c). Correspondingly, we observed decreased LV protein expression of the key NE-synthetizing enzyme tyrosine hydroxylase (TH) in the LV (Fig. 3d) and diminished LV myocardial density of TH-positive sympathetic nerves (Fig. 4a, c, d). From proteins involved in the myocardial fate of NE, we observed an increased expression (p = 0.03) of presynaptic norepinephrine transporter (NET, responsible for synaptic NE reuptake, Fig. 3e) and significant decrease of organic cation transporter (OCT3, responsible for myocardial uptake of NE) in ACF compared to the sham/control group (Fig. 3f). MAO-A, NE-degrading enzyme was upregulated (Fig. 3g) and correspondingly, ROS generated by MAO-A (Fig. 3h) were increased in ACF LV, while gene expression of Adrb1 was downregulated compared to sham/intact group (Fig. 3i).
Fig. 3
figure 3
Impact of ACF and effects of RDN on selected parameters of sympathetic nervous system in left ventricle. a Plasma norepinephrine (NE). b NE content in kidney. c NE content in left ventricle (LV). d Biosynthesis of NE—protein expression of tyrosine hydroxylase (TH). e Preganglionic NE transport—protein expression of NE transporter (NET). f NE transport to cardiomyocyte—protein expression of organic cation transporter 3 (OCT3). g Degradation of NE—protein expression of monoamine oxidase A (MAO-A). h Production of reactive oxygen species (ROS) by MAO-A. i Gene expression of beta-1 adrenergic receptor (Adrb1). j Gene expression of choline muscarinic receptor type 2 (Chrm2). k Gene expression of acetylcholinesterase (Ache). l Neprilysin activity measured in kidney. N = 8 in sham/intact, N = 8 in ACF/intact, N = 8 in ACF/RDN
Fig. 4
figure 4
Results of immunohistochemical staining of tyrosine hydroxylase (TH, red color) in left ventricle. Zoom 25x in a smaller square embedded in an illustrative zoom 2x in a larger square. a Ratio of sympathetic nerves immunostained with TH antibody to the total area. b sham/intact. c ACF/intact. d ACF/RDN. N = 4 in sham/intact, N = 5 in ACF/intact, N = 4 in ACF/RDN
In parasympathetic cardiac signalization, ACF/intact rats had decreased acetylcholinesterase (Ache) and an unsignificant trend to decreased choline muscarinic receptor type 2 (Chrm2, Fig. 3j, k) in the LV compared to sham/intact.
ACF/intact rats displayed increased neprilysin activity in the kidney, compared to the sham/intact group (Fig. 3l).
LV function and HF markers: the impact of RDN
RDN procedure significantly improved LV systolic function in ACF/RDN animals compared to the ACF/intact group. ACF/RDN group had increased Ees, PRSW, and Ees/Ea ratio compared to ACF/intact group (Fig. 2a). Peak LV pressure or effective arterial elastance (Ea) was not affected by RDN (2 ± 0.23 vs. 2.2 ± 0.22, p = 0.5, see Supplementary Information).
ACF/RDN group had less elevated markers of adverse myocardial remodeling compared to ACF/intact group—reduced gene expression of fibrotic markers (Col1a1/Col3a1 ratio), decreased the Myh 7/6 ratio compared to ACF/intact group, while Nppa gene expression was not significantly reduced. After RDN in the ACF group, we did not observe any changes in medium-chain fatty acids in gene expression of Acadm compared to ACF/intact rats (Fig. 2b).
Cardiac autonomic nervous system in HF: the impact of RDN
RDN in ACF rats significantly reduced NE in the plasma and in the kidney (Fig. 3a, b). Despite we targeted the sympathetic nervous system in the kidney, we observed profound changes of sympathetic nerves in the heart—RDN led to increased NE levels in LV compared to ACF/intact rats (Fig. 3c).
In ACF/RDN group, we observed a numerically higher, but not significant increase in protein expression of TH (Fig. 3d). Sympathetic nerve density measured by TH staining was not significantly changed in ACF/RDN compared to ACF/intact (Fig. 4a, d). There was no difference in OCT3 (Fig. 3f), but strong trend to reduce protein expression of NET (presynaptic NE reuptake, p = 0.06, Fig. 3e) and significantly reduced MAO-A in ACF/RDN group, compared to ACF/intact rats (Fig. 3g). RDN/ACF rats displayed a trend to (p = 0.07) to higher gene expression of Adrb1 (Fig. 3i), and significantly increased gene expression of Chrm2 and Ache (Fig. 3j, k). We observed a significant positive correlation between gene expression of Adrb1 and TH among all groups (see Supplementary Information). RDN in ACF rats significantly reduced the activity of neprilysin in the kidney, compared to ACF/intact rats (Fig. 3l).
DOI
https://doi.org/10.1038/s41440-024-01580-3
Hypertension Research volume 47, pages2718–2730 (2024)
Published
02 February 2024
