Experimental animals, ACF-induced HF and design of the experiment
Rats were bred, maintained, and handled at the Centre of Experimental Medicine of Institute of Clinical and Experimental Medicine (IKEM), Prague and approved on 26 June 2017 by the Animal Care and Use Committee of the IKEM, Prague, project number 50/2017, in accordance with guidelines and practices established by the Directive 2010/63/EU of the European Parliament on the Protection of Animals Used for Scientific Purposes. This study was conducted in compliance with the ARRIVE 2.0 Guidelines for Reporting Animal Research 45,46.
Rats were housed under standard conditions, at 22±1 °C, 12-hours light/dark cycles, with ad libitum access of rat chows and tap water. In the experiment, 80 males of the hypertensive heterozygous Ren-2 transgenic (mREN2)27 rats (TGR, n = 40) and normotensive Hannover Sprague-Dawley rats (HSD, n= 40) were used.
As previously reported 25, HF due volume overload was induced by creation of an aortocaval fistula (ACF) between the abdominal aorta and inferior vena cava 47. Operation procedures were performed under general anaesthesia by an intraperitoneal administration of ketamine (Calypsol, Gedeon Richter, Hungary, 160 mg/kg) and midazolam (Dormicum, Roche, France, 160 mg/kg). In the 8 weeks old HSD and TGR rats, shunt was created by needle technique between vessels (18-gauge needle,1.2 mm in diameter) and subsequently sealed with cyanoacrylate glue (Histoacryl, B.Braun AG, Germany). ACF induction was confirmed by recording of pulsatile flow in the vena cava. Control HSD and TGR rats underwent a sham operation. After five weeks from successful ACF induction, while heart failure was fully developed in rats, 15 weeks of treatment has begun. HSD and TGR rats with ACF were treated with an AT1 receptor blocker (200 mg/l, losartan, Lozap, Zentiva, Prague, CZ) or an ACE inhibitor trandolapril (6 mg/l, Gopten, Mylan, Canonsburg, Pennsylvania, USA), drugs dissolved in drinking water. Age matched control without treatment were used. At the end of experiment, were rats in the age 28 weeks old decapitated, blood and heart tissue (left and right ventricles) collected for further analyses (Figure 1).
Assessment of circulating levels of atrial natriuretic peptide (ANP)
ANP as a marker of HF was estimated in blood serum samples according to the manufacturer's recommended protocol (Atrial Natriuretic Peptide EIA Kit, RAB00385, Sigma-Aldrich, St. Louis, USA). Briefly, 100 μl of anti-ANP antibody was pipetted into the included secondary antibody-coated microplate. After incubation and anti-ANP antibody removal, 100 μl of blood serum/ standards was added. After these steps followed incubations with 100 μl of HRP-streptavidin solution, and 100 μl of TMB solution (3,3,5,5-tetramethylbenzidine). Finally, 50 μl of STOP solution was added and the absorbance was measured with a spectrophotometer (Synergy H1, BioTek, USA) at 450 nm. The amount of ANP (pg/ml) in the samples was calculated using a regression equation from the values of the standards.
Determination of thiobarbituric acid reactive substances (TBARS)
TBARS levels, as a marker for lipid peroxidation, were analysed as was described previously Shlafer and Shepard (1984) with modifications by spectrophotometric method. Briefly, 40 µl of standards, blood serum samples or tissue homogenates (same as in SDS-PAGE method) from RV and LV together with 40 µl of 20% trichloroacetic acid solution were mixed with a quadruple volume of TBARS reagent (37 mmol/L C4H4N2O2S; 500 mmol/L NaOH; 15% v/v CH3COOH) and incubated for 70 min at 100 °C. After cooling the samples for 20 minutes, were samples pipetted into another tube with prepared mixture of n-butanol and pyridine (14:1, v/v), and subsequently centrifuged at 5000x g for 10 minutes. The resulting supernatant (organic phase) was used for measurement of absorbance at 535 nm by Synergy H1 Hybrid Multi-Mode Microplate Reader (Biotek, Vermont, USA). The concentration of malondialdehyde (MDA) in the samples was calculated from the calibration curve formed from a tetrabutylammonium malondialdehyde salt 48,49.
SDS-PAGE and Western Blotting
As was described in detailed in our previous works 50, approximately 100 mg of frozen heart tissue (RV, LV) was homogenized in lysis buffer (20%, 10 mmol/L EDTA, 100 mmol/L Tris, pH 6.8) and diluted in Laemmli sample buffer under reducing and non-reducing conditions. Loading equal amounts of protein (6 – 60 µg) per lane were separated in 10% SDS-polyacrylamide gels at a constant voltage of 90 V (Mini-Protean TetraCell, Bio-Rad,Hercules, CA, USA) and electrically transferred to a nitrocellulose membrane (0.2 m pore size, Advantec, Tokyo, Japan). Membranes were blocked with 5% low-fat milk and incubated with the appropriate primary and secondary antibodies (Table 1). Proteins visualization was carried out by enhanced chemiluminescence method and quantification by densitometric analysis using Carestream Molecular Imaging Software (version 5.0, Carestream Health, New Haven, CT, USA).
Immunofluorescence methods and Quantitative Image Analysis (QIA)
As was described in our previous works, we used for immunodetection of Cx43 and collagen-1 a 10 µm thick frozen tissue sections from the apex of the heart. Cryosections were fixed in ice-cold methanol, permeabilized in 0.3% Triton X-100, and blocked with the solution of 1% bovine serum albumin. Tissue sections were incubated with primary antibody and then with secondary antibody (Table 1). For visualization of actin filaments, phalloidin (Sigma-Aldrich, St.Louis, MO, USA, #P2141) was applied to the sections. At the end were tissue sections mounted in the Vectashield medium (H-1200, Vector Laboratories-Inc., Burlingame, CA, USA) and detected by Zeiss Apotome 2 microscope (Carl Zeiss, Jena, Germany).
For QIA ten randomly acquired images per heart were examined. Immunofluorescence signal was defined as a number of pixels with the protein signal intensity exceeding a threshold of 30 on the 0–255 Gy scale. The total number of Cx43 positive pixels was indicated as a total integral optical density per area (IOD). Quantification of Cx43 lateralization was manually performed by selection and delineation of terminal intercalated disc-related areas of Cx43. IOD of lateral Cx43 immunofluorescence signal (Image-Pro Plus) corresponds to the difference between total IOD per area and IOD of terminal areas. Lateralization of Cx43 was calculated from the ratio of IOD of lateral topology divided by the total IOD and expressed in percentage 34,51.
Determination of Collagen Content by Hydroxyproline Measurement
The hydroxyproline content considered a marker of fibrosis was measured by a spectrophotometric method as described previously. Approximately 100 mg of myocardial tissue from LV and RV was dried overnight and hydrolysed in 6 M HCl. Samples were dried again and incubated with 4 M NaOH, Chloramine T in the acetate-citrate buffer. After oxidation reaction were samples incubated with Ehrlich’s reagent solution. Samples were pipetted onto a microplate and final concentration of hydroxyproline was measured spectrophotometrically at 550 nm. The hydroxyproline content was then expressed in mg per total weight of the LV or RV 17,48.
Histology and Enzyme Histochemistry of myocardial tissue
10 µm thick tissue cryosections from the apex were used for conventional hematoxylin–eosin staining and catalytic enzyme histochemistry performed according to Lojda (1979) with modifications. For hematoxylin–eosin reaction to viewing tissue structure were tissues fixed in 4% buffered formaldehyde, stained with hematoxylin - eosin solutions, poured with gelatin and covered with a coverslip.
Endothelium-related alkaline phosphatase (AP, E.C.3.1.3.1) with naphthol AS-MX phosphate as a substrate and dipeptidyl peptidase-4 (DPP4, E.C.3.4.15.4) with glycyl-L-proline-4-methoxy-beta naphthylamide as a substrate were used to detect activities arteriolar and venular capillary network. Cryosections were incubated in solution (1.2 mM L-Leucine 4-methoxy-β-naptylamide hydrochloride; 5% dimethylformamide; 2.4 mM Fast blue BB; 0.1 M Na2 HPO4 x 2H2O; 1 M KH2PO4), poured with gelatin and covered with a coverslip.
Examination of stained sections were observed under light microscope (Zeiss Apotome 2 microscope Carl Zeiss, Jena, Germany). Ten randomly acquired microscopic images per heart were used for quantification 34,52.
Statistical analysis
All data are expressed as mean ± SD. Kolmogorov–Smirnov normality test to examine if variables are normally distributed was used. Differences between groups were detected by one-way ANOVA, followed by Bonferroni’s posthoc test. p<0.05 was considered statistically significant.