Beneficial effects of paricalcitol on cardiac dysfunction and remodelling in a model of established heart failure

Abstract

Background and Purpose

The synthetic vitamin D3 analogue paricalcitol acts as a selective activator of the vitamin D receptor (VDR). While there is evidence for cardioprotective effects of paricalcitol associated with the VDR pathway, less information is available about the structural and functional cardiac effects of paricalcitol on established heart failure (HF) and particularly its effects on associated electrophysiological or Ca²⁺ handling remodelling.

Experimental Approach

We used a murine model of transverse aortic constriction (TAC) to study the effect of paricalcitol on established HF. Treatment was initiated 4 weeks after surgery over five consecutive weeks, and mice were sacrificed 9 weeks after surgery. Cardiac MRI (CMRI) was performed 4 and 9 weeks after surgery. Hearts were used for biochemical and histological studies and to isolate ventricular myocytes for electrophysiological and calcium imaging studies.

Key Results

CMRI analysis revealed that, compared with vehicle, paricalcitol treatment prevented the progression of ventricular dilation and hypertrophy after TAC and halted the corresponding decline in ejection fraction. These beneficial effects were related to the attenuation of intracellular Ca²⁺ mishandling remodelling, antifibrotic and antihypertrophic effects and potentially antiarrhythmic effects by preventing the reduction of K⁺ current density and the long QT, JT and TpTe intervals observed in HF animals.

Conclusion and Implications

The results suggest that paricalcitol treatment in established HF hampers disease progression and improves adverse electrophysiological and Ca²⁺ handling remodelling, attenuating the vulnerability to HF‐associated ventricular arrhythmias. Paricalcitol may emerge as a potential therapeutic option in the treatment of HF.

Abbreviations

CMRI

cardiac MRI

EF

ejection fraction

HF

heart failure

HW

heart weight

I_kur

the ultrarapid delayed rectifier K⁺ current

I_ss

the non‐inactivating steady‐state outward current

I_tof

the fast transient outward current

JT interval

interval between QRS end, J point, and the end of the T wave

LV

left ventricle

LVEDV

left ventricular end‐diastolic volume

LVEF

left ventricular ejection fraction

LVESV

left ventricular end‐systolic volume

LVH

left ventricular hypertrophy

LVM

left ventricular mass

PTH

parathyroid hormone

SR

sarcoplasmic reticulum

TAC

transverse aortic constriction

TL

tibia length

TpTe

the interval from the peak to the end of the T wave

VDR

vitamin D receptor

What is already known

• Depressed cardiac function in heart failure is commonly associated with impairment of Ca_i ²⁺ handling.

• Heart failure is associated with an increased incidence of malignant arrhythmias and sudden death.

What this study adds

• Paricalcitol treatment of established heart failure prevents Ca²⁺ mishandling

• Paricalcitol treatment counteracts the down‐regulation of K⁺ currents associated with heart failure .

What is the clinical significance

• Paricalcitol treatment reduces left ventricular hypertrophy, fibrosis and adverse electrophysiological and Ca_i ²⁺ handling remodelling

• Paricalcitol might emerge as a potential therapeutic option in the treatment of heart failure.

1. INTRODUCTION

Chronic heart failure (HF) is a major public health concern in ageing societies and a leading cause of hospitalisation and mortality (Ponikowski et al., 2014; Savarese & Lund, 2017). The disorder is usually accompanied by progression from adaptive to maladaptive hypertrophy and also cardiac dilation and diminished left ventricular ejection fraction (LVEF). Adverse cardiac remodelling is an important determinant of the clinical outcome of HF and is linked to disease progression and poor prognosis (Heusch et al., 2014). Ventricular remodelling encompasses cardiomyocyte hypertrophy, pro‐fibrotic responses and down‐regulation of K⁺ currents, which alter the electrotonic coupling between cells and prolong QT intervals, overall increasing the risk of ventricular arrhythmias and sudden cardiac death (Nass, Aiba, Tomaselli, & Akar, 2008; Tomaselli et al., 1994). Depressed cardiac function in HF is also commonly associated with impairment of intracellular Ca²⁺ handling. In this regard, failing hearts often show depressed systolic Ca²⁺ release and lower sarcoplasmic reticulum (SR)‐Ca²⁺ load, which compromises cell contractility (Gómez‐Hurtado et al., 2017; Ruiz‐Hurtado et al., 2015; Val‐Blasco et al., 2017).

Over the past 25 years, considerable progress has been made in the treatment of chronic HF using angiotensin‐converting enzyme inhibitors and angiotensin AT₁ receptor antagonists, and also mineralocorticoid receptor antagonists, β‐adrenoceptor blockers and resynchronisation therapy (Owens, Brozena, & Jessup, 2016). Although the pharmacological treatment of HF aims to halt progressive cardiac hypertrophy and LVEF decline, as many as 25%–40% of all patients die from chronic HF within 1 year of diagnosis. New therapies are entering clinical trials yearly (Nabeebaccus, Zheng, & Shah, 2016; Tamargo, Caballero, & Delpón, 2018), but the identification of new targets with therapeutic potential in HF continues to be an area of great interest.

Paricalcitol (19‐nor‐1α, 25‐dihydroxyvitamin D2) is a synthetic vitamin D analogue that acts as a selective activator of the vitamin D receptor (VDR) and is indicated for the prevention and treatment of secondary hyperparathyroidism associated with chronic kidney disease (Robinson & Scott, 2005). There is a growing body of work demonstrating beneficial cardioprotective properties associated with the VDR pathway (Gardner, Chen, & Glenn, 2013; Meredith & McManus, 2013; Norman & Powell, 2014). Yet little information is available on the cardiac structural and functional effects of paricalcitol on established HF. Likewise, there is a paucity of evidence on the effect of paricalcitol on arrhythmogenic disorders, or in the remodelling of Ca²⁺ handling, associated with established HF.

In the present study, we used a murine model of HF induced by pressure overload (transverse aortic constriction [TAC]) to test the hypothesis that paricalcitol treatment blocks the progression of the disease and has cardioprotective effects on adverse Ca²⁺ handling and electrical remodelling.

2. METHODS

2.1. Animals

All animal care and experimental procedures followed the guidelines for ethical care of experimental animals of the European Union (2012/63/EU) and were approved by the Bioethical Committeede of the Consejo Superior de Investigaciones Científicas (Proex 035‐15). Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010; McGrath, Drummond, McLachlan, Kilkenny, & Wainwright, 2010) and with the recommendations made by the British Journal of Pharmacology.

Male C57BL/6J mice (24–29 g, 10 weeks of age) were used in all experiments. Mice were bred and housed under specific pathogen‐free conditions in the Experimental Animal Centre of the Biomedical Research Institute “Alberto Sols” CSIC‐UAM/CIBER‐CV, Madrid, Spain. Animals were maintained at controlled temperature (23–25°C) on a 12‐h light/dark cycle with ad libitum access to water and a standard diet (Teklad Global 14% Protein Rodent Maintenance Diet, Harlan Laboratories Inc., Indianapolis, IN) that contained 600 IU·g⁻¹ of vitamin D₃ (cholecalciferol). The animal cages (Polysulfone type SII, Techniplast, Italy) were 553 cm² by 20.8 cm depth, and animals were housed with a maximum of four mice per cage.

2.2. Transverse aortic constriction

Animals were anaesthetised by i.p. injection of a mixture of ketamine (100 mg·kg⁻¹) and xylazine (10 mg·kg⁻¹). Adequacy of anaesthesia was determined by assessing loss of withdrawal reflex. Mice were then intubated and underwent trans‐sternal thoracotomy. The transverse aorta was constricted with a 6–0 black braided silk suture tied against a 27‐gauge needle, as described by Rockman et al. (1991). A similar procedure was followed for sham‐operated mice, but no suture was tied. Mice were given subcutaneous buprenorphine (1 mg·kg⁻¹) for pain relief before and after the surgery. A heated pad was used during surgery to minimise discomfort and maintain body temperature.

2.3. Study design

A scheme of the experimental approach is shown in Figure 1. Animals were distributed into two experimental groups: sham operated and TAC operated. Cardiac structure and function were analysed by cardiac MRI (CMRI) 4 weeks after surgery. At this point, only TAC‐operated mice with ejection fraction (EF) <58% and with significant left ventricle (LV) dilation were included in the TAC group. The mice were then blindly randomised to the vehicle or paricalcitol treatment groups, with each group originally consisting of 16 mice. Treatment with vehicle (water for injection, 3.9% propylene glycol and 1.3% ethyl alcohol) or with 300 ng·kg⁻¹ paricalcitol (Normon S.A., Madrid, Spain) was delivered by i.p. injection three times a week (Monday, Wednesday and Friday) for five consecutive weeks. CMRI was repeated after the 5 weeks of treatment (9 weeks after the surgery). Also, a standard ECG was performed, and blood samples were drawn from the retro‐orbital plexus. Finally, mice were killed, and hearts were excised, weighed and prepared for biochemical and histological studies. In some experiments, hearts were retrogradely perfused through the aorta using a modified Langendorff apparatus to isolate ventricular myocytes (see below).

FIGURE 1.

Diagram of experimental design

CMRI assessments before surgery were not performed because the aim of the study was to analyse the effect of paricalcitol on established HF. Therefore, control values were those obtained in the sham group 4 and 9 weeks after the surgery.

2.4. Macroscopic parameters, cell capacitance and serum analysis

The heart weight (HW) to tibia length (TL) ratio was measured as an index of cardiac hypertrophy (Table 1). Cardiomyocyte surface area was quantified using LSM Zeiss Image Browser 4.2 software (Carl Zeiss, Germany) (Table 1), and the membrane capacitance of the cardiomyocytes was evaluated using the whole‐cell configuration of the patch‐clamp technique (see below) (Table 1). Blood samples were centrifuged at 1,600x g for 5 min to obtain blood serum. A parathyroid hormone (PTH) assay was performed using the Mouse PTH 1–84 elisa Kit (Immutopics, San Clemente, CA). Calcium and phosphorus assays were performed using colorimetric assay kits (Calcium Assay Kit and Phosphate Assay Kit, respectively; Abcam, Cambridge, UK) on an EnSpire™ Multimode Plate Reader (Perkin Elmer, Waltham, MA).

TABLE 1.

Macroscopic parameters, cell capacitance and serum analysis from each experimental group 9 weeks after surgery

	Sham		TAC
	Vehicle	Paricalcitol	Vehicle	Paricalcitol
Ca2+ (mg·dl−1)	12.5 ± 0.6 (N = 16)	11.4 ± 0.6 (N = 8)	11.0 ± 0.6 (N = 15)	12.4 ± 0.5 (N = 14)
PTH (pg·ml−1)	145.2 ± 17.4 (N = 12)	56.0 ± 6.9 # (N = 7)	123.0 ± 12.4 (N = 14)	52.8 ± 3.4 & (N = 11)
Pi (mg·dl−1)	8.6 ± 0.5 (N = 15)	6.8 ± 0.7 (N = 9)	7.1 ± 0.7 (N = 15)	8.0 ± 0.5 (N = 15)
Ratio (HW·TL−1)	9.9 ± 0.3 (N = 13)	9.9 ± 0.3 (N = 12)	15.2 ± 1.0 # (N = 15)	12.1 ± 0.6 & (N = 14)
HW (mg)	174.6 ± 5.9 (N = 13)	180.3 ± 7.1 (N = 12)	273.6 ± 18.3 # (N = 15)	187.7 ± 21.9 & (N = 14)
TL (mm)	17.6 ± 0.2 (N = 13)	18.2 ± 0.2 (N = 12)	17.9 ± 0.2 (N = 15)	17.7 ± 0.2 (N = 14)
Cell area (μm2)	3,251 ± 121.0 (N = 6; n = 72)	3,422 ± 159.1 (N = 5; n = 45)	4,555 ± 167.4 # (N = 6; n = 78)	3,941 ± 177.1 & (N = 6; n = 56)
Capacitance (pF)	186.3 ± 16.8 (N = 5; n = 13)	189.7 ± 15.8 (N = 5; n = 10)	375.9 ± 39.6 # (N = 5; n = 12)	236.2 ± 16.8 & (N = 5; n = 15)

Note: Values are expressed as mean ± SEM.

Abbreviations: HW, heart weight; n, number of independent cardiomyocytes; N, number of mice; PTH, parathyroid hormone; TL, tibia length.

^# P < .05, significantly different from sham–vehicle.

^& P < .05 significantly different from TAC–vehicle.

2.5. Cardiac MRI

CMRI was carried out on a 7.0‐Tesla MR system (Bruker Pharmascan, Bruker, Ettlingen, Germany). Analysis was performed in the Biomedical Research Institute “Alberto Sols” CSIC‐UAM. Mice were anaesthetised with an isoflurane and oxygen mixture (2% in 1 L·min⁻¹ for induction and 1.5% during the acquisition). The temperature of the animals was monitored and maintained at 35–36°C during the experiments. Heart and respiratory rates were recorded using the 1025 SAM monitoring and gating system (SA Instruments, Inc., New York, NY). Several images were acquired to localise the short‐axis planes. After position adjustment, 6 to 12 slices were acquired to cover the entire heart. Each slice consisted of 20 gated frames synchronised with the cardiac cycle. For image acquisition, we used the Intragate FLASH sequence with the following parameters: echo time (TE) = 1.262 ms; field of view = 3 × 3 cm²; slice thickness = 0.8 mm; matrix size = 256 × 256. The acquired data were zero‐filled to achieve a reconstructed matrix size of 256 × 256. Images were analysed using SEGMENT v2.0 R5642 (http://segment.heiberg.se). Six slices were selected from each heart and analysed by manual segmentation of left ventricular endocardial and epicardial borders in all the image frames. After segmentation of the images, SEGMENT was used to automatically calculate the following functional parameters: left ventricular end‐diastolic volume (LVEDV) (μl) and left ventricular end‐systolic volume (LVESV) (μl), EF (%) and left ventricular mass (LVM) (mg). LVM was estimated by LV wall volume × the specific gravity of the myocardium (1.05 g·cm⁻³).

2.6. ECG recordings

ECG recordings were obtained on the Small Animal Physiological Monitoring System (Harvard Apparatus, Holliston, MA). Mice were located in prone position on a warm pad at 37°C during the recordings and were lightly anaesthetised with an isoflurane and oxygen mixture (1.5%). ECG recordings were obtained under basal conditions for 5 min. ECG registries were converted into LabChart binary files using a cross‐platform Java program. Files were analysed in a blinded fashion using LabChart 7.0 software (AD Instruments, Sydney, Australia) (RRID:SCR_017551).

2.7. Histology

Sirius red staining was performed on paraffin sections to measure fibrosis. Tissue samples were dehydrated, embedded in paraffin and cut into sections (5‐μm thickness). Slices were stained with Sirius red (Direct Red 80) (25 ml of 10% aqueous Direct Red in 225 ml of 1.3% aqueous picric acid; Sigma‐Aldrich Chemical Company, Madrid, Spain). Quantification of collagen content was performed using ImageJ software (NIH) (RRID:SCR_003070) and expressed as the perivascular fibrosis area divided by vessel lumen area, and as a percentage of interstitial fibrosis over the total amount of tissue, excluding lumen and perivascular fibrosis. A single investigator blinded to the experimental groups performed the analysis.

2.8. Total RNA isolation and real‐time PCR

Total RNA was extracted from a portion of the LV of representative mice of each experimental group using the RNeasy Mini Kit on a QIAcube robotic workstation (both from Qiagen, Hilden, Germany) and was quantified using a NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, DE). In total, 250 ng of RNA was retrotranscribed using the High‐Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA), and real‐time PCR was performed on an ABI 7900HT Fast Real‐Time PCR platform (Applied Biosystems). The endogenous Rplp0 gene (36B4) was used as reference for fold‐induction calculations using the ^ΔΔCt method.

The following primers were purchased from Invitrogen (Carlsbad, CA):

• m‐Nppa‐F: ATTGACAGGATTGGAGCCCAGAGT

• m‐Nppa‐R: TGACACACCACAAGGGCTTAGGAT

• m‐Rcan1.4‐F: GAGCGAGTCGTTCGTTAAGC

• m‐Rcan1.4‐R: GCCACACAAGCAATCAGGGA

• m‐Serpine1‐F: CGGCAGATCCAAGATGCTATG

• m‐Serpine1‐R: GACCAGCTCTAGGTCCCGCT

• m‐Col1a1‐F: AATGGCACGGCTGTCTGCGA

• m‐Col1a1‐R: AGCACTCGCCCTCCCGTCTT

• m‐Col3a1‐F: CTGTAACATGGAAACTGGGGAAA

• m‐Col3a1‐R: CCATAGCTGAACTGAAAACCACC

• m‐Atp2a2‐F: TAAATGCCCGCTGTTTTGCT

• m‐Atp2a2‐R: TTGTCATCTGCCAGGACCAT

• m‐Pln‐F: ACCGAAGCCAAGGTCTCCTA

• m‐Pln‐R: TCCATTATGCCAGGAAGGCAA

• m‐Rplp0‐F: AGATGCAGCAGATCCGCAT

• m‐Rplp0‐R: GTTCTTGCCCATCAGCACC

2.9. Cardiomyocyte isolation

Adult single ventricular cardiomyocytes were isolated as reported previously (Delgado et al., 2015). Briefly, mice were heparinised and anaesthetised with ketamine (100 mg·kg⁻¹) and xylazine (10 mg·kg⁻¹) by i.p. injection. Adequacy of anaesthesia was determined by assessing loss of withdrawal reflex. Hearts were then rapidly removed and retrogradely perfused through the aorta using a modified Langendorff apparatus. Hearts were perfused for 2–3 min at 36–37°C with a standard calcium‐free Tyrode's solution containing 0.2‐mM EGTA and then for 3–4 min with the same Tyrode's solution containing collagenase type II (251 IU·ml⁻¹, Worthington Biochemical, Lakewood, NJ) and 0.1‐mM CaCl₂. After perfusion, the hearts were removed from the Langendorff apparatus, and the ventricles were chopped into small pieces and gently stirred for 2–5 min in a standard Tyrode's solution containing 0.1 mmol·L⁻¹ CaCl₂, collagenase (251 IU·ml⁻¹) and BSA (Sigma‐Aldrich) to disperse the isolated ventricular myocytes. Cell suspensions were filtered through a 250‐μm nylon mesh, pelleted by centrifugation for 3 min at 20× g and suspended in Tyrode's solution containing 0.5 mmol·L⁻¹ CaCl₂ and BSA. Cells were centrifuged as before and suspended in a storage solution containing 1 mmol·L⁻¹ CaCl₂ and BSA. Standard calcium‐free Tyrode's solution contained in mmol·L⁻¹: 130 NaCl, 5.4 KCl, 0.4 NaH₂PO₄, 0.5 MgCl₂, 25 HEPES, 22 glucose; the pH was adjusted to 7.4 with NaOH. Single rod‐shaped and Ca²⁺‐tolerant cells with clear cross‐striations were used for electrophysiological and intracellular Ca²⁺ imaging studies.

2.10. Electrophysiological studies

Isolated ventricular myocytes were placed in a chamber mounted on the stage of an inverted microscope and allowed to adhere for 5 min before being superfused with external solution. Whole‐cell voltage‐clamp recordings were obtained in the ruptured patch configuration using an Axopatch 200B patch‐clamp amplifier (Molecular Devices, Sunnyvale, CA). The patch pipette resistance for K⁺ current recordings was 1–2 MΩ and was filled with a solution containing in mmol·L⁻¹: 135 KCl, 4 MgCl₂, 5 EGTA, 10 HEPES, 10 glucose, 5 Na₂ATP, and 3 disodium creatine phosphate; the pH was adjusted to 7.2 with KOH. Whole‐cell voltage‐clamp experiments were performed at room temperature (24–26°C).

In the adult mouse ventricle, the repolarising K⁺ currents involve three components that can be identified by their specific voltage dependence and their sensitivity to pharmacological agents. First, total K⁺ currents (I_K+) were elicited by applying 300‐ms depolarising steps from −50 to +50 mV (with 10‐mV steps) from a holding potential of −80 mV at a frequency of 0.2 Hz. Then the fast transient outward current (I_tof), the ultrarapid delayed rectifier K⁺ current (I_kur) and the non‐inactivating steady‐state outward current (I_ss) were obtained as previously reported (Tamayo et al., 2018). The external solution for I_K+ recordings contained, in mmol·L⁻¹: 135 NaCl, 10 glucose, 10 HEPES, 1 MgCl₂, 1 CaCl₂, 4 KCl, 2 CoCl₂; the pH was adjusted to 7.4 with NaOH.

L‐type Ca²⁺ currents (I_CaL) were elicited by applying 300‐ms depolarising voltage pulses from a holding potential of −50 mV, between −40 and +60 mV (with 10‐mV steps) at a frequency of 0.2 Hz. Ca²⁺ currents were normalised to cell capacitance to obtain current density. The external Tyrode's solution for I_CaL recordings contained, in mmol·L⁻¹: 140 NaCl, 1.1 MgCl₂, 5.4 CsCl, 10 glucose, 5 HEPES, 1.8 CaCl₂; the pH was adjusted to 7.4 with CsOH. The intracellular recording pipette solution for whole‐cell experiments contained, in mmol·L⁻¹: 100 CsCl, 20 TEACl, 5 EGTA, 10 HEPES, 5 Na₂ATP, 0.4 Na₂GTP, 5 Na₂ creatine phosphate, 0.06 CaCl₂; the pH was adjusted to 7.2 with CsOH.

Current density was calculated from peak K⁺ currents or current amplitude (I_CaL) normalised to the membrane capacitance. Membrane capacitance (C _m) (Table 1) was elicited by applying ±10‐mV voltage steps from −60 mV, and C_m was calculated according to the following equation:

$$C_{\mathrm{m}}={\tau }_{\mathrm{c}}{I}_0/V_{\mathrm{m}}\left[1-\left(I\infty /I_0\right)\right],$$

where τ _c is the time constant of the membrane capacitance, I ₀ the maximum capacitance current value, V _m the amplitude of the voltage step and I∞ the amplitude of the steady‐state current.

2.11. Intracellular Ca²⁺ dynamics

Changes in intracellular Ca²⁺ concentration ([Ca²⁺]_i) were recorded in intact isolated cardiomyocytes. Cardiomyocytes were loaded with the fluorescent Ca²⁺ dye Fluo‐3 acetomethyl ester (Fluo‐3 AM, Invitrogen) (5 μmol·L⁻¹). All recordings were carried out at room temperature (20–23°C). Images were recorded on a Meta Zeiss LSM 710 confocal microscope (40× oil inversion objective with a 1.2 NA), by scanning cells with an Argon laser every 1.54 s. Fluo‐3 AM was excited at 488 nm, and the emitted fluorescence was collected at >505 nm. [Ca²⁺]_i transients were recording in Fluo‐3 AM‐loaded cardiomyocytes electrically stimulated at 2 Hz. The amplitude of [Ca²⁺]_i transients (F/F ₀) was calculated by normalising the maximal fluorescence by the basal fluorescence. Subtraction of the background fluorescence was carried out beforehand in both cases. Therefore, we calculated F/F ₀ as (F‐background)/(F ₀‐background). The decay time constant of Ca²⁺ transients (τ) was measured by fitting the decay trace. Ca²⁺ sparks were recorded in quiescent cells and identified using an automated detection system and by employing a criterion that discriminated the detection of false events, as previously reported (Gómez‐Hurtado et al., 2017). SR‐Ca²⁺ load was determined by rapid caffeine (10 mmol·L⁻¹) administration to deplete the SR of Ca²⁺ stores, after field stimulation to reach the steady state of Ca²⁺ load.

SERCA2a activity (k SERCA2a) was indirectly determined by subtracting the decay rate constant of caffeine‐evoked [Ca²⁺]_i transients from the decay rate constant of systolic [Ca²⁺]_i transients (Bode et al., 2011; Delgado et al., 2015).

Data analysis was performed with “in‐house” routines using IDL 8 software (Research System Inc. Boulder, CO) and designed by AM Gómez (UMR‐S 1180, INSERM). Images were corrected for background fluorescence. Cardiomyocyte surface area was quantified with LSM Zeiss Image Browser 4.2 software (Carl Zeiss) (Table 1).

2.12. Data and statistical analysis

The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology (Curtis, Ashton, Moon, & Ahluwalia, 2018). Data are expressed as mean ± SEM or mean ± SD when individual values were included in the plot. Statistical analysis was performed using GraphPad Prism v.6.0 (GraphPad Software Inc., La Jolla, CA) (RRID:SCR_002798). One‐way ANOVA was used to compare significance among groups. If ANOVA produced a significant value of F (P < 0.05) and there was no significant variance inhomogeneity, Bonferroni's post hoc multicomparison analysis was applied. Paired Student's t test was used for comparison of resonance magnetic imaging data obtained in the same mouse 4 and 9 weeks after surgery. All P values were two‐tailed, and P values <0.05 were considered as statistically significant. Statistical analysis was undertaken only for studies where each group size was at least 5. N = number of mice and n = number of independent cardiomyocytes. The data that support the findings of this study are available from the corresponding authors upon reasonable request.

2.13. Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to Pharmacology (Harding et al., 2018), and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20 (Alexander, Cidlowski et al., 2019; Alexander, Fabbro et al., 2019; Alexander, Kelly et al., 2019; Alexander, Mathie et al., 2019).

3. RESULTS

3.1. Paricalcitol treatment prevents the progression of established cardiac hypertrophy and HF

Cardiac structure and function were evaluated by CMRI in the same mouse 4 and 9 weeks after the surgery. Figure 2a–d shows individual values of LVM, LVEDV, LVESV and EF, 4 weeks after the surgery, just prior to vehicle or paricalcitol treatment. LVM, LVEDV and LVESV were significantly increased, and EF was decreased in both TAC groups compared with sham groups. Moreover, there were no differences between the TAC and TAC+paricalcitol groups in any parameter before starting the treatment. Figure 3a shows CMR images of representative hearts 9 weeks after surgery, specifically, four‐chamber long‐axis (upper panel) and two‐chamber short‐axis (lower panel) views from the four experimental groups (sham, sham+paricalcitol, TAC and TAC+paricalcitol). The results of this analysis are shown in Figure 3b–e. Evaluation of LVM is represented in Figure 3b and showed that LVM was maintained in both sham groups from 4 to 9 weeks. As mentioned, LVM was significantly higher in the TAC groups than in the sham groups 4 weeks after surgery. Notably, whereas LVM continued to increase over the next 5 weeks in the TAC–vehicle group, it remained stable in the TAC+paricalcitol group. These data strongly suggest that paricalcitol has a significant effect on preventing increases in cardiac mass in the setting of established cardiac hypertrophy. Quantification of the CMRI parameters LVEDV and LVESV is shown in Figure 3c,d, respectively. Both parameters significantly increased 4 weeks after the TAC surgery, compared with the sham group, indicating a dilation of the LV that was significantly greater 9 weeks after the surgery. Treatment with paricalcitol for 5 weeks once the LV dilation was present, significantly prevented the progression of both parameters. Consistent with these changes, the EF was significantly lower in the TAC groups than in equivalent sham groups at 4 weeks after surgery (Figure 3e). The EF further decreased in the TAC–vehicle group at 9 weeks after surgery, whereas treatment with paricalcitol during 5 weeks inhibited the decline in EF. Taken together, these data suggest that paricalcitol blocks the functional and structural deleterious remodelling in this model of established HF. In addition, the beneficial effects of paricalcitol occurred in the absence of an increase in the serum concentration of Ca²⁺ or phosphate but with a significant decrease in PTH indicating that the drug effectively activates VDRs (Table 1).

FIGURE 2.

Stage of the disease before paricalcitol treatment analysed by cardiac MRI (CMRI). (a) Left ventricular mass (LVM), (b) left ventricular end‐diastolic volume (LVEDV), (c) left ventricular end‐systolic volume (LVESV) and (d) ejection fraction (EF). There were no differences between the TAC and TAC PC groups in any parameter before the treatment was initiated. Groups are sham (N = 16), sham PC (N = 15), TAC (N = 14) and TAC PC (N = 16). Data are expressed as mean ± SD. ⁺ P < .05, significantly different as indicated. N, number of mice; n.s., non‐significant; PC, paricalcitol

FIGURE 3.

Paricalcitol blocks pathological progression of heart failure. (a) Representative cardiac magnetic resonance (CMR) images of mouse hearts at the end of diastole. Top, four‐chamber long‐axis views; bottom, two‐chamber short‐axis views. (b) CMR analysis of left ventricular mass (LVM), (c) left ventricular end‐diastolic volume (LVEDV), (d) left ventricular end‐systolic volume (LVESV) and (e) ejection fraction (EF). Groups are sham (N = 16), sham PC (N = 15), TAC (N = 14) and TAC PC (N = 16). Data are expressed as mean ± SEM. *P < .05, significantly different from TAC 4 weeks. ^& P < .05, significantly different as indicated. Each TAC group was significantly different from each sham group. Scale bar represents 5 mm. N, number of mice; PC, paricalcitol

3.2. Paricalcitol improves [Ca²⁺]_i mishandling remodelling associated with established HF

Given that cardiac dysfunction and Ca²⁺ mishandling are closely related, we next determined whether the beneficial effects of the paricalcitol treatment regime were associated with changes to intracellular Ca²⁺ dynamics.

First, we analysed the I_CaL in cardiomyocytes using the patch‐clamp technique in the whole‐cell configuration. As shown in Figure 4a, the mean values of current–voltage (IV) density curves for I_CaL were similar between the four groups. We examined systolic Ca²⁺ release by the analysis of [Ca²⁺]_i transients in isolated cardiomyocytes on field stimulation at 2 Hz. Figure 4b shows representative line‐scan confocal images of [Ca²⁺]_i transients in cells from the four groups. Results showed that the amplitude of [Ca²⁺]_i transients was significantly lower in the TAC–vehicle group than in the sham groups, which was accompanied by a slower decay time constant (τ) (Figure 4c,d). These changes were also associated with a decrease in cell contractility in the TAC–vehicle group (Figure 4e). By contrast, cardiomyocytes obtained from TAC+paricalcitol mice showed values of τ and amplitude of [Ca²⁺]_i transients that were similar to those of the sham groups. Paricalcitol treatment also prevented the TAC‐induced decrease in cell shortening (Figure 4e). To determine whether the impairment in the systolic Ca²⁺ release was related to changes in SR‐Ca²⁺ load, we used caffeine administration to estimate the SR‐Ca²⁺ load in representative cardiomyocytes of the four groups. As shown in Figure 4f, the amplitude of caffeine‐evoked [Ca²⁺]_i transients was lower in the TAC group than in the sham group, and this was significantly prevented after 5 weeks of paricalcitol treatment. Decreased SR‐Ca²⁺ load, together with slower time decay of [Ca²⁺]_i transients, would indicate a fault in SR‐Ca²⁺ reuptake. Given that SR‐Ca²⁺ ATPase (SERCA2a) plays a key role in SR‐Ca²⁺ uptake, we next analysed SERCA2a activity (k SERCA2a). Results showed that k SERCA2a was significantly lower in the TAC group (4.3 ± 0.6 s⁻¹; n = 21) than in the sham group (5.6 ± 1.3 s⁻¹; n = 16) or the sham+paricalcitol group (5.8 ± 1.8 s⁻¹; n = 11) (P < 0.05), clearly indicating an impairment of SERCA2a function in cardiomyocytes from TAC hearts. Treatment of TAC mice with paricalcitol prevented this impairment (5.4 ± 1.7 s⁻¹; n = 17). We next used RT‐PCR to look for possible changes in the expression levels of Atp2a2 (SERCA2a) and Pln (phospholamban), which could account for the functional impairment in SR‐Ca²⁺ load and reuptake and the modulation by paricalcitol treatment. As shown in Figure 4g,h, expression of Atp2a2 and Pln was significantly lower in the TAC group than in the sham group, and paricalcitol treatment prevented the decrease in expression of both genes. In addition, Atp2a2/Pln ratio was calculated, and similar values between groups were obtained (Figure S1). These results indicate that paricalcitol prevents Ca²⁺ mishandling in HF by maintaining both systolic Ca²⁺ release and physiological SR‐Ca²⁺ load. As depressed SR‐Ca²⁺ load in HF is usually associated with an increase in diastolic Ca²⁺ leak, we next analysed the frequency of Ca²⁺ sparks in all experimental groups. As expected, the analysis of the frequency of Ca²⁺ sparks normalised to SR‐Ca²⁺ load in quiescent cardiomyocytes showed a significant increase in the number of these events in the TAC group, compared with the sham‐operated group. By contrast, the frequency of Ca²⁺ sparks was significantly lower in the TAC group treated with paricalcitol and was similar to that found in the sham+paricalcitol group (Figure S2). These data demonstrate that paricalcitol treatment significantly prevents the increase in diastolic Ca²⁺ leak observed in isolated cardiomyocytes from TAC hearts mice.

FIGURE 4.

Paricalcitol treatment attenuates [Ca²⁺]_i mishandling in isolated myocytes from hearts after 9 weeks of TAC. (a) Current–voltage curves for I_CaL density obtained in myocytes isolated from the sham (n = 13, N = 4), sham PC (n = 10, N = 4), TAC (n = 12, N = 4) and TAC PC (n = 15, N = 4) groups. The upper panel shows the protocol used to elicit I_CaL. (b) Representative line‐scan confocal images of transients stimulated to 2 Hz in cells from the sham, sham PC, TAC and TAC PC groups. (c) Mean values of peak of fluorescence [Ca²⁺]_i transients (peak F/F ₀), (d) decay time constant (τ) and (e) cell shortening in the sham (n = 42, N = 6), sham PC (n = 42, N = 5), TAC (n = 62, N = 6) and TAC PC (n = 58, N = 6) groups. (f) Mean values of amplitude SR‐Ca²⁺ load (peak F/F ₀) are shown in the sham (n = 24, N = 6), sham PC (n = 26, N = 5), TAC (n = 34, N = 6) and TAC PC (n = 29, N = 6) groups. (g) Paricalcitol treatment increases mRNA expression of Atp2a2 (SERCA2a) and (h) Pln (phospholamban) 9 weeks after TAC surgery. Groups are sham (N = 7), sham PC (N = 5), TAC (N = 7) and TAC PC (N = 6). Data are expressed as mean ± SEM in panel (a) and mean ± SD in panels (c)–(h). ^# P < .05, sham significantly different from TAC. ^& P < .05, TAC significantly different from TAC PC. The sham groups were not significantly different from sham PC groups. N, number of mice; n, number of independent cardiomyocytes; PC, paricalcitol

3.3. Paricalcitol prevents the transcriptional and structural development of fibrosis in established HF

Chronic pressure overload is commonly associated with myocardial fibrosis (Shimizu & Minamino, 2016). Accordingly, we next examined for fibrosis in heart tissue of the four experiment groups 9 weeks after surgery. Figure 5a,b shows representative sirius red‐stained images of perivascular (Figure 5a) and interstitial (Figure 5b) collagen. Histological and quantitative analysis showed that both parameters were significantly greater 9 weeks after TAC surgery (Figure 5c,d, respectively). Comparison of hearts from the TAC–vehicle and TAC+paricalcitol groups revealed that the development of perivascular (Figure 5c) and interstitial (Figure 5d) fibrosis was significantly attenuated in the latter.

FIGURE 5.

Paricalcitol treatment inhibits the progression of both perivascular and interstitial cardiac fibrosis by preventing the increased expression of serpine1, Col1a1 and Col3a1. (a, b) Representative images of Sirius red staining showing perivascular (a) and interstitial (b) fibrosis from each experimental group 9 weeks after the surgery. (c) Paricalcitol treatment attenuates the increase of perivascular fibrosis (normalised by luminal area). Groups are sham (n = 49, N = 9), sham PC (n = 57, N = 9), TAC (n = 43, N = 9) and TAC PC (n = 60, N = 10). (d) Interstitial fibrosis is also diminished by paricalcitol treatment. Groups are sham (N = 9), sham PC (N = 9), TAC (N = 9) and TAC PC (N = 10). (e–g) Real‐time PCR analysis of mRNA expression of serpine1, col1a1 and col3a1. Groups are sham (N = 7), sham PC (N = 5), TAC (N = 7) and TAC PC (N = 6). Data are expressed as mean ± SD. ^# P < .05, sham significantly different from TAC. ^& P < .05 TAC significantly different from TAC PC. ^$ P < .05 sham PC significantly different from TAC PC. The sham groups were not significantly different from the sham PC groups. LA, luminal area; n, number of independent cardiomyocytes; N, number of mice; PC, paricalcitol; PVCA, perivascular collagen area

Cardiac fibrosis is related to excessive synthesis and accumulation of the matrix protein collagen and to the increased expression of plasminogen activator inhibitor (PAI‐1) (Takeshita et al., 2004). Analysis of the expression of the pro‐fibrotic genes Serpine‐1 (PAI‐1), Col1a1 (Collagen‐1) and Col3a1 (Collagen‐3) revealed that the expression of all was significantly higher in the TAC–vehicle group than in the sham groups (Figure 5e–g). Of note, cardiomyocytes from the TAC+paricalcitol group displayed mRNA values of the pro‐fibrotic genes very similar to those of the sham groups. These results clearly indicate that paricalcitol treatment hinders, at the transcriptional and structural level, the development of cardiac fibrosis in an established model of HF.

3.4. Antihypertrophic effects of paricalcitol are related to inhibition of the calcineurin/NFAT pathway

Histological examination of hearts after 9 weeks of TAC corroborated the existence of left ventricular hypertrophy (LVH), with an increase in the size of the LV cavity (Figure 6a) and evident cardiomyocyte hypertrophy (Figure 6b). Paricalcitol treatment significantly attenuated TAC‐induced LVH, at both the macroscopic and cellular levels. As shown in Figure 6c and Table 1, the HW/TL ratio was significantly greater in the TAC–vehicle group than in the sham groups, whereas this parameter was not affected in the TAC+paricalcitol group. In addition, quantitative analysis of cardiomyocyte surface area in the four groups showed that cells from the TAC–vehicle group were significantly larger than those from the sham groups (Figure 6d and Table 1) and the TAC‐induced increase in size was blocked by paricalcitol. The results from membrane capacitance analysis mirrored these findings (Table 1). To validate these results, we analysed the mRNA expression of atrial natriuretic peptide (ANP) (Nppa), a classic hypertrophic marker gene, in all groups, finding a pattern similar to that for the fibrosis marker genes (Figure 6e).

FIGURE 6.

Paricalcitol treatment blunts hypertrophic development by attenuating the calcineurin/Rcan1.4 pathway. (a) Representative examples of histological images of whole hearts in axial views from sham, sham PC, TAC and TAC PC mice. (b) Representative confocal microscopy images of adult cardiomyocytes. (c) The TAC PC‐treated group has a ratio of HW (mg) to TL (mm) lower than that of the TAC group. Groups are sham (N = 13), sham PC (N = 12), TAC (N = 15) and TAC PC (N = 15). (d) Paricalcitol prevents cardiomyocyte hypertrophy. Groups are sham (n = 72, N = 6), sham PC (n = 45, N = 5), TAC (n = 78, N = 6) and TAC PC (n = 56, N = 6). (e, f) Real‐time PCR analysis of mRNA expression of Nppa and Rcan1.4. Groups are sham (N = 7), sham PC (N = 5), TAC (N = 7) and TAC PC (N = 6). Data are expressed as mean ± SD. ^# P < .05, sham significantly different from TAC. ^& P < .05, TAC significantly different from TAC PC. HW, heart weight; n, number of independent cardiomyocytes; N, number of mice; PC, paricalcitol; TL, tibia length

Finally, because the calcineurin/NFAT pathway is recognised as one of the most important signal transduction cascades underlying cardiac hypertrophy (Molkentin et al., 1998), we analysed the expression of the calcineurin/NFAT target gene Rcan1.4 (regulator of calcineurin). As shown in Figure 6f, Rcan1.4 expression was significantly higher in the TAC–vehicle group than in the sham groups, and, as anticipated, paricalcitol significantly attenuated its expression. This result suggests that the calcineurin/NFAT pathway is likely to be a target for the antihypertrophic action of paricalcitol.

3.5. Long QT, JT and TpTe intervals and reduced repolarising I_K+ densities in mice with established HF are prevented by chronic paricalcitol administration

LVH and HF are strongly associated with long QT intervals and repolarisation‐related arrhythmogenic disorders. Accordingly, we measured ECG signals in mice 9 weeks after TAC surgery to study the effect of pressure overload on QT interval and the possible influence of paricalcitol on this parameter. Representative ECG recordings of QT interval measurements are shown in Figure 7a, and mean QT intervals in the four groups are represented in Figure 7b. The results revealed that mean QT intervals were significantly longer in the TAC–vehicle group than in the sham groups. Notably, the mean QT intervals in the TAC+paricalcitol group were significantly shorter than in the TAC group and were similar to those in the sham groups. All of these changes occurred in the absence of alterations to the heart rate (Figure 7c). In addition to the QT interval, we analysed other parameters associated with ventricular electrocardiographic repolarisation, namely, the JT interval (the interval between QRS end, J point, and the end of the T wave) and the TpTe interval (the interval from the peak to the end of the T wave). Both ECG parameters were increased in the TAC group, and treatment with paricalcitol prevented their prolongation (Figure 7d,e).

FIGURE 7.

Long QT, JT and TpTe intervals associated with TAC 9 weeks after the surgery are prevented by paricalcitol treatment. (a) Representative examples of baseline ECG and QT interval measurement, (b) QT intervals, (c) HR, (d) JT intervals and (e) TpTe intervals values obtained in the sham, sham PC, TAC and TAC PC groups 9 weeks after the surgery. Groups are sham (N = 6), sham PC (N = 5), TAC (N = 7) and TAC PC (N = 8). Data are expressed as mean ± SD. ^# P < .05, sham significantly different from TAC. ^& P < .05, TAC significantly different from TAC PC. HR, heart rate; PC, paricalcitol

K⁺ currents (I_K+) are important contributors to repolarisation in healthy myocardium, and a reduction in the densities of I_K+ have been associated with the acquired long QT syndrome seen in LVH and HF (Choy et al., 1997; Marionneau et al., 2008). Figure 8a shows representative recordings of total I_K+ density in cardiomyocytes isolated from the four different groups of mice, and Figure 8b shows mean IV curves for I_K+ density. I_K+ densities were significantly lower in the TAC–vehicle group than in the sham groups. By contrast, I_K+ densities in the TAC+paricalcitol group were similar to those of the sham+paricalcitol group. In the adult mouse ventricle, outward K⁺ currents involved in the repolarisation phase are composed of three components. Figure 8c–e shows mean IV curves for I_tof, I_kur, and I_ss densities, respectively, obtained in myocytes isolated from the four experimental groups. Mean values of all three components were significantly decreased in the TAC group for the three components as compared with the sham or sham+paricalcitol groups. One‐way ANOVA of I_tof, I_kur and I_ss densities followed by Bonferroni's multiple comparisons test at +50 mV in the four experimental groups showed that the treatment with paricalcitol significantly prevented the decrease of I_tof but not that of I_kur or I_ss (Figure 8f–h, respectively).

FIGURE 8.

Paricalcitol treatment blocks the reduction of I_K+ density and I_tof in myocytes isolated from mice 9 weeks after TAC surgery. (a) Representative records of I_K+ density obtained in four different myocytes isolated from sham, sham PC, TAC and TAC PC mice. The upper panel shows the protocol used to elicit I_K+. (b–e) Current density–voltage curves for I_K+ (b), I_tof (c), I_kur (d) and I_ss (e), obtained on myocytes isolated from sham, sham PC, TAC and TAC PC. Panels (f) and (g) shows individual values of I_tof, I_kur and I_ss densities recorded at +50 mV. Data are expressed as mean ± SEM for IV curves and mean ± SD for data at +50 mV. Groups are sham (n = 13 for I_K+, n = 13 for I_tof, n = 10 for I_kur and n = 14 for I_ss), sham PC (n = 10 for I_K+, n = 9 for I_tof, n = 10 for I_kur and n = 11 for I_ss), TAC (n = 12 for I_K+, n = 19 for I_tof, n = 15 for I_kur and n = 15 for I_ss) and TAC PC (n = 15 for I_K+, n = 16 for I_tof, n = 11 for I_kur and n = 13 for I_ss). ^# P < .05, sham significantly different from TAC. ^& P < .05, TAC significantly different from TAC PC. n, number of independent cardiomyocytes; n.s., non‐significant; PC, paricalcitol

These results indicate that paricalcitol prevents QT, JT and TpTe interval prolongation induced by TAC, in the main part by preserving physiological I_K+ density, mostly I_tof , in failing hearts.

4. DISCUSSION

Paricalcitol is a synthetic vitamin D3 analogue that functions as a selective activator of VDR and has been shown to reduce PTH levels with very low hypercalcaemic and hyperphosphataemic secondary effects (Hervás Sánchez, Prados Garrido, Polo Moyano, & Cerezo Morales, 2011; Martin et al., 1998). The recommended dosage of paricalcitol in humans ranges from 0.04 to 0.24 μg·kg⁻¹ administered as a bolus on alternate days (Robinson & Scott, 2005). In the present study, we used 0.3 μg·kg⁻¹ paricalcitol by i.p. injection three times a week (Monday, Wednesday and Friday) for five consecutive weeks. A large body of work demonstrates that the selective activation of VDRs modulates, directly or indirectly, the transcriptional activity of hundreds of genes and modifies several enzymes and signalling pathways. Many of these pathways are involved in regulatory processes related to the parathyroid gland, gut and bone, but others are involved in non‐classical actions of VDR, including those of potential relevance to cardiovascular diseases such as contractile function, hypertrophy, fibrosis, neurohormonal activation and inflammation (Andress, 2007; Lavie, Lee, & Milani, 2011; Meredith & McManus, 2013; Norman & Powell, 2014).

Our results showed that 5‐week paricalcitol treatment of mice with established HF induced by TAC, had a significant cardioprotective effect by halting the decline of LVEF and the progression of pre‐existing cardiac hypertrophy. Supporting our data, other authors have reported cardioprotective effects of paricalcitol on cardiac hypertrophy and development of HF in an experimental model of pressure overload in rats induced by a high‐salt diet (Bae et al., 2011; Bodyak et al., 2007). Importantly, in the present study, we comprehensively analysed the mechanisms involved in the beneficial effect of paricalcitol on the deleterious remodelling that occurs during HF development and specifically focused on Ca²⁺ handling, myocardial fibrosis, cardiac hypertrophy and remodelling of repolarising K⁺ currents.

Cardiac pump function depends on cardiomyocyte contraction, which is activated by Ca²⁺. Contraction is initiated by an action potential. The initial depolarisation (mediated by Na⁺ channels) activates I_CaL, which triggers Ca²⁺ release from the SR, resulting in an increased [Ca²⁺]_i that activates contraction. [Ca²⁺]_i must be reduced for relaxation to take place, and this occurs mainly by pumping Ca²⁺ to the SR via SERCA2a and across the plasma membrane via the sarcolemmal Na⁺–Ca²⁺ exchanger (Bers, 2002). A decline in systolic Ca²⁺ release and SR‐Ca²⁺ load is frequently reported in HF models, which also document a reduced expression and/or function of SERCA2a (Bers, 2006; Gómez‐Hurtado et al., 2017). In our study, I_CaL density was not modified in any of the groups studied; however, ventricular cardiomyocytes isolated from failing hearts 9 weeks after TAC surgery showed diminished [Ca²⁺]_i transients with a slower time decay rate (τ) and a significant decrease in cell shortening. Furthermore, the caffeine‐evoked [Ca²⁺]_i transients showed a decrease in SR‐Ca²⁺ load. In this setting, paricalcitol treatment maintained the levels of [Ca²⁺]_i transient amplitude and cell shortening close to that observed in the sham group. In addition, SR‐Ca²⁺ load was preserved. Moreover, paricalcitol treatment suppressed the decrease in the expression of both Atp2a2 (SERCA2a) and Pln (phospholamban) observed in TAC hearts, which is likely to contribute to sustaining SR‐Ca²⁺ content, improving [Ca²⁺]_i transient amplitude and cell contractility. These data do not discount the possibility that post‐translational changes in phospholamban and/or SERCA2a proteins might be involved in the observed effects.

Importantly, the prevention of the impairment in both systolic Ca²⁺ release and SR‐Ca²⁺ load induced by paricalcitol administration in the TAC group can also be due to a prevention of the increased diastolic Ca²⁺ leak associated with HF development. Indeed, our results support this idea, as paricalcitol administration significantly prevented the increased frequency of Ca²⁺ sparks in isolated cardiomyocytes obtained from hearts of TAC‐induced mice. Along this line, paricalcitol significantly attenuated the impairment of Ca²⁺ handling and cardiac dysfunction in mice deleted for 1α‐hydroxylase (Choudhury et al., 2014), which converts the inactive form of vitamin D3 (25(OH)D₃) to the biological active form (1,25(OH)₂D₃).

Cardiac fibrosis is a distinguishing feature of pathological ventricular remodelling and can contribute to both systolic and diastolic cardiac dysfunctions. It is characterised by myofibroblast differentiation and excessive synthesis and accumulation of matrix proteins, together with the induction of protease inhibitors, PAI‐1 and tissue inhibitors of metalloproteinases (Kong, Christia, & Frangogiannis, 2014; Takeshita et al., 2004; Weber, Brilla, & Janicki, 1993). Indeed, it is well recognised that synthesis of type I and III collagens is markedly increased in the remodelling fibrotic heart, regardless of the aetiology of fibrosis (Mukherjee & Sen, 1993; Weber et al., 1993). Our histological and molecular analysis showed that paricalcitol treatment depressed the formation of perivascular and interstitial fibrosis 9 weeks after TAC surgery. In accordance with our data, other authors have reported a role for paricalcitol in reducing fibrosis and remodelling in the myocardium (Gardner et al., 2013; Meems et al., 2012; Meredith & McManus, 2013).

The CMRI study also showed that paricalcitol had a significant effect in preventing the progression of pre‐existing cardiac hypertrophy, as revealed by analysis of the HW/TL ratio and ventricular cardiomyocyte size. Similarly, the TAC‐induced increase in the expression of the hypertrophic gene marker Nppa (ANP) was significantly blunted by paricalcitol, further supporting its antihypertrophic activity. Indeed, antihypertrophic effects of paricalcitol have been described in other experimental models of HF (Bae et al., 2011; Bodyak et al., 2007; Kong et al., 2014) and in uraemic rats (Freundlich et al., 2014). The mechanisms underlying the antihypertrophic effects of the VDR pathway have been investigated in cardiomyocyte‐specific VDR knockout mice, which exhibit significant cardiac hypertrophy and activation of the fetal gene program (Chen et al., 2011). Importantly, the antihypertrophic activity of paricalcitol was lost in these mice after treatment with angiotensin II, supporting the possibility that the direct action of paricalcitol on cardiomyocytes is independent of haemodynamic effects (Chen & Gardner, 2013; Gardner et al., 2013). Our results clearly demonstrate an increase of diastolic Ca²⁺ leak in failing (TAC) mice. Moreover, the sustained elevation of the cytosolic calcium that occurs in HF is known to activate calcineurin‐NFAT signalling. Dephosphorylation of NFAT by calcineurin allows for NFAT to enter the nucleus, to promote the expression of genes involved in structural cardiac remodelling (Olson & Williams, 2000; Wilkins & Molkentin, 2004). Paricalcitol treatment prevented the impairment in diastolic Ca²⁺ release and SR‐Ca²⁺ load, thus reducing the cytosolic Ca²⁺ concentration, which counteracts the activation of the calcineurin/NFAT pathway. Moreover, results from gene expression array analysis of isolated ventricular myocytes of cardiomyocyte‐specific VDR knockout mice showed that the expression of modulatory calcineurin inhibitory protein 1 (or Rcan1.4) was up‐regulated twofold as compared with control cells (Chen et al., 2011). RCANs are a family of proteins that can bind directly to the catalytic subunit of calcineurin and inhibit its activity. RCAN 1.4 works as an endogenous feedback inhibitor, protecting cells from uncontrolled calcineurin activity. The calcineurin/NFAT/Rcan1.4 cascade has been linked to the development of pathological hypertrophy in several experimental models of pressure overload (Diedrichs et al., 2004; Wilkins et al., 2004; Zou et al., 2001) and also in human HF (Diedrichs et al., 2004; Lim & Molkentin, 1999). In the present study, we found that the increase in Rcan1.4 mRNA expression in hypertrophied hearts from TAC mice was blunted by paricalcitol treatment, suggesting that the VDR‐dependent antihypertrophic activity of paricalcitol is linked to the negative modulation of the calcineurin/NFAT/Rcan1.4 pathway. A similar inhibition of the NFAT target gene Rcan1.4 associated with reduced cardiac hypertrophy was recently reported in paricalcitol‐treated uraemic rats (Czaya et al., 2019).

LVH and HF are associated with QT prolongation and increased susceptibility for ventricular arrhythmias (Nass et al., 2008; Tomaselli & Marbán, 1999). Indeed, about 50% of the mortality in patients with HF is reported to be associated with sudden cardiac death, which is often related to malignant ventricular arrhythmias (Nass et al., 2008; Tomaselli et al., 1994). In addition, prolonged TpTe intervals have been associated with increased risk of sudden cardiac death in patients (Panikkath et al., 2011). Therefore, in addition to the QT interval, we examined other electrocardiographic parameters associated with ventricular repolarisation, finding that the QT, JT and TpTe intervals were significantly prolonged in TAC‐operated mice, whereas TAC‐induced mice receiving paricalcitol showed a significant decrease in the duration of the three intervals. The mechanism for prolongation of ventricular repolarisation in HF involves the development of adverse electrophysiological remodelling including down‐regulation of K⁺ currents responsible for repolarisation of the action potential (Nattel, Maguy, Le Bouter, & Yeh, 2007). Using patch‐clamp experiments, we found that the density of I_K+, I_tof, I_kur and I_ss was significantly lower in myocytes isolated from TAC mice than in the sham or sham+paricalcitol groups. Treatment of TAC mice with paricalcitol prevented this decrease mainly for I_K+ and I_tof. A reduction in I_tof density has been reported in animal models of HF (Gómez‐Hurtado et al., 2017; Wang et al., 2007) and in humans (Beuckelmann, Näbauer, & Erdmann, 1993; Näbauer, Beuckelmann, & Erdmann, 1993). Our results thus establish that paricalcitol reduces QT, JT and TpTe interval prolongation associated with LVH and HF by preventing deleterious electrophysiological remodelling. Therefore, through this mechanism, paricalcitol might attenuate the vulnerability to LVH/HF‐associated ventricular arrhythmias.

In conclusion, we provide the first demonstration that paricalcitol, a drug used for the treatment of secondary hyperparathyroidism, prevents the progression of established HF by alleviating LVH, cardiac fibrosis and adverse electrophysiological and Ca²⁺ handling remodelling in mice. Our results suggest that paricalcitol might emerge as a potential therapeutic option in the treatment of HF.

AUTHOR CONTRIBUTIONS

M.T. and L.M.‐N. performed most of the experiments, A.V.‐B. and M.F.‐V. helped with Ca²⁺ handling experiments, E.L. contributed to CMRI analysis, J.A.N.‐G. and G.R.‐H. performed ECG studies, P.P. helped with RT‐PCR studies, M.J.G.‐P. helped with TAC surgery, G.R.‐H. and M.F.‐V. contributed to the critical interpretation of data and M.F.‐V. and C.D. performed some electrophysiological and Ca²⁺ handling experiments, conceived the hypothesis, supervised the project, reviewed the data and wrote the manuscript. All the authors critically analysed the manuscript and approved its final version for publication.

CONFLICT OF INTEREST

The authors have no conflicts of interest to disclose.

DECLARATION OF TRANSPARENCY AND SCIENTIFIC RIGOUR

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the BJP guidelines for Design & Analysis, and Animal Experimentation, and as recommended by funding agencies, publishers and other organisations engaged with supporting research.

Supporting information

Figure S1: Atp2a2/Pln ratio. Similar values between groups were obtained. Sham (N = 6), Sham PC (N = 5), TAC (N = 6) and TAC PC (N = 5). Data expressed as mean ± SD. N = number of mice. PC = paricalcitol.

Figure S2: Ca²⁺ sparks frequency normalised by SR‐Ca²⁺ load. TAC surgery increases the number of sparks compared with the sham group whereas paricalcitol treatment prevents this increase. Sham (n = 30, N = 6), sham PC (n = 37, N = 5), TAC (n = 58, N = 6) and TAC PC (n = 51, N = 6). ^# p < 0.05 Sham vs. TAC. ^&p < 0.05 TAC vs TAC PC. Ca²⁺ sparks frequency in the sham and sham PC groups was not significantly different. Data expressed as mean ± SD. N = number of mice; n = number of independent cardiomyocytes. PC = paricalcitol.

Tamayo M, Martín‐Nunes L, Val‐Blasco A, et al. Beneficial effects of paricalcitol on cardiac dysfunction and remodelling in a model of established heart failure. Br J Pharmacol. 2020;177:3273–3290. 10.1111/bph.15048