*2.7. ELISA*

To determine the protein content of TGF-β<sup>1</sup> and TGF-β<sup>2</sup> in the LV, the immunoassay was performed according to the Quantikine® ELISA Mouse/Rat/Porcine/Canine TGF-β<sup>1</sup> Immunoassay Protocol (R&D Systems, Inc., Minneapolis, MN, USA) and the Quantikine® Human TGF-β<sup>2</sup> Immunoassay Protocol (R&D Systems, Inc., Minneapolis, MN, USA). Polyclonal secondary rabbit-antibody (Dianova, Hamburg, Germany) conjugated to horseradish peroxidase specific for TGF-β<sup>1</sup> was used. Absorbance was measured at a wavelength of 450 nm. The TGF-β<sup>1</sup> and TGF-β<sup>2</sup> concentrations were calculated by the I-smart program (Packard Instruments Company, Inc., Downers Grove, IL, USA). The protein concentrations were related to the median of the CTRL animals (Ctrl = 1.0).

#### *2.8. Histological Investigation of the Heart*

The apex of the heart was cut off, fixated in formalin and embedded in paraffin. Sections of 8–9 μm thickness were stained with hematoxylin—eosin and Masson's trichrome. For the evaluation of cardiac fibrosis, sections were examined with an Axioscope microscope (Carl Zeiss, Oberkochen, Germany), digitized with AxioCam MRc 5 (Carl Zeiss, Oberkochen, Germany) and evaluated using the Photoshop CS6 image processing program. A score ranging from 0 to 3 was applied to assess the histological degree of fibrosis: 0 = no signs of fibrosis; 0.5 = marginal perivascular fibrosis and/or marginal interstitial fibrosis; 1 = perivascular fibrosis + mild interstitial fibrosis; 2 = perivascular fibrosis + moderate interstitial fibrosis; 3 = fibrosis of the entire heart [14].

#### *2.9. Statistical Analysis*

Statistical analyses were carried out with the software package SigmaPlot Version 14.0 (Systat Software GmbH, Erkrath, Germany) for Windows. The groups were statistically compared using Analysis of Variance (ANOVA) procedures. Firstly, a Shapiro-Wilk test of normality was performed. If data were normally distributed, we used a One Way ANOVA with a post-hoc test according to Fisher's LSD method. This applied to SBP, HW and quantitative evaluations from histology. These data are presented as means ± SEM. If the data were not normally distributed, a Kruskal-Wallis ANOVA on ranks with a post-hoc test according to Dunn's method was applied. These data (the results of the biochemical analyses) are given as medians [25th/75th percentile]. As age, BW and SBP at baseline might have an effect on the results, we additionally performed an analysis of covariance (ANCOVA) to assess the influence of these covariates and their interaction with treatment effects on the results. *p* values < 0.05 were considered significant.

#### **3. Results**

#### *3.1. Systolic Blood Pressure*

Baseline SBP as measured at 60 weeks of age was similar in all three groups (CTRL 231 ± 9 mmHg, CAP 228 ± 13 mmHg, CAP + NIF 245 ± 8 mmHg). During the observation period, SBP increased slightly in CTRL rats (239 ± 5 mmHg; *p* > 0.05). CAP and CAP + NIF induced a significant SBP reduction by about 20% after 22 weeks of therapy (*p* < 0.001; Figure 1). SBP measured at 82 weeks of age was not dependent on age, BW and SBP at baseline.

**Figure 1.** Systolic blood pressure of old SHR expressed as mean ± SEM at baseline (60 weeks of age; hatched columns) and at the final measurement after 22 weeks of experiment (82 weeks of age; filled columns). Significance marks: # significant vs. baseline measurement (*p* < 0.001); \* significant vs. CTRL (*p* < 0.001).
