Primary antibodies against cardiac troponin I (ab47003), Vimentin (ab8978), TGF-β1 (ab25121), collagen I (ab90395) were purchased from Abcam (England). Primary antibodies against DDR2 (N-20) (sc7555), von Willebrand Factor (vWF) (sc27649) were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Primary antibodies against p-Smad2 (No. 3108) and Smad4 (No. 9515) were purchased from Cell Signaling Technology (Boston, MA, USA). Primary antibody against CD44 (bs2507R) was purchased from Bioss Technology (Beijing, China). Donkey anti-Mouse IgG, Alexa Fluor 488 (A21202) and Donkey anti-Goat IgG, Alexa Fluor 594 (A11058) were purchased from Life Technologies (Waltham, MA, USA). Goat anti-Rabbit IgG (CW0103) and Goat anti-Mouse IgG (CW0102) were purchased from ComWin Biotch Co. (Beijing, China). Type II collagenase (17101-015) and angiotensin II (A9525) were purchased from Thermo Fisher Scientific (Waltham, MA, USA) and Sigma-Aldrich (St. Louis, MO, USA) respectively.

All procedures were performed in compliance with “The Guide for the Care of Use of Laboratory Animals” published by the US National Institute of Health (8th edition) and revised in 2011. The experimental protocol was approved by the Experimental Animal Management Committee of Shanxi Medical University, Taiyuan, China (Approval NO. 2015-0001).

One to three-day old Sprague-Dawley rats were obtained from the Animal Laboratory Center of Shanxi Medical University. Cardiac fibroblasts were isolated from rats using a modified version of the differential attachment method as described previously [13]. In brief, the intact hearts were rapidly removed into the cold phosphate buffered saline to wash and rinse out the blood for 3-4 times. After excluding the connective tissue and atrium, the ventricles were minced into 1 cubic mm tissue blocks and digested with mixed enzyme solution that contains 0.08% trypsin and 0.04% type II collagenase through a 6 min-period of 8 - 10 cycles at 37˚C. The minced tissue solution was filtered through a 70 µm-cell strainer and centrifuged at 800 g for 10 min at room temperature. The cells collected were plated in the wells with moderate DMEM medium supplemented with 10% FBS, and then incubated at 95% O₂ and 5% CO₂ under 37˚C. Two hours later, unattached cells were discarded and the attached cells were plated in fresh DMEM medium containing 10% FBS at 37˚C. When the degree of cell fusion reached to about 80%, the cells were digested with trypsin EDTA and passaged. Subsequent experiments were performed using cells at 2 - 4 passages. Before each assay, isolated cardiac fibroblasts were incubated and synchronized with serum-free DMEM medium for 24 h. In the following observations, the cells were exposed to a fresh medium solution (control), Ang II activation (10 nM to 1 µM) or Ang II (50 nM) plus telmisartan (100 nM). The selection of telmisartan dose was based on its inhibition on collagen secretion and production from Ang II stimulated cardiac fibroblasts [14].

Immunofluorescence staining with vimentin and DDR2 (N-20) for fibroblasts, vWF for endothelial cells and cardiac troponin I for cardiomyocytes were selected to identify whether isolated cells are cardiac fibroblasts, but not endothelial cells and cardiomyocytes. In brief, the cells were plated at a density of 1 × 10⁴ per well in 96-well plates, fixed with 4% paraformaldehyde, and then washed with 0.5% Triton X-100. Immunofluorescence staining was performed on incubated cells with vimentin (dilution at 1:50), DDR2 (N-20), vWF and cardiac troponin I (all dilution at 1:100), respectively, at 4˚C overnight. The secondary antibodies including Donkey anti-mouse IgG, Alexa Fluor 488 or Donkey anti-goat IgG, Alexa Fluor 594 were then applied for 90 min at room temperature correspondingly. The nuclei were stained with DAPI for 10 min. Finally, the morphology of the cells were observed and the photos were taken under the fluorescence microscopy using a digital camera (Olympus, Japan)

Cell proliferation assay was assessed by a Cell Counting Kit-8 (CCK-8, Dojindo, kumamoto, Japan) according to the manufacturer’s protocol. Cardiac fibroblasts were seeded into 96-well plates at a density of 7 × 10³ cell per well, and treated with Ang II at different concentrations for 24 h. CCK-8 reagents were then added into the culture medium at a ratio of 1:10 at 37˚C for 2 h to identify the cell proliferation based on manufacture’s instruction (Dojindo, Shanghai, China). The absorbance at 450 nm was measured using a microplate reader (Molecular Devices Corp, San Jose, CA, USA) to estimate cell proliferation.

The transwell migration assay is a commonly used test to study the migratory response of the cells to oxidative stress stimulation. In brief, the cell suspension before assay was prepared at the density about 1 × 10⁶/ml. After adding 500 μl serum free medium into 24-well plates, the transwell chambers (No. 3422, CORNING, NY, USA) were placed; 200 μl of cell suspension was pipetted into the upper chamber and cultured at 37˚C for 12 h without touching the membrane or introducing air bubbles. Following an incubation period (24 h), the migrated cells through membrane were fixed using 4% paraformaldehyde at room temperature for 30 min. After washing with PBS twice, the membranes were stained by 0.1% crystal violet solution for 20 min. To count the cell migration, the transwell chamber was viewed using an inverted microscope (Olympus, Japan), and the numbers of migrated cells were counted from eight high-powered fields (HPF, 400× magnification) to get an average sum of the cells.

The level of CD44 mRNA detected from the cells was measured by real-time PCR. Briefly, 0.5 μg of total RNA isolated from the cells using RNAiso plus (9108, TaKaRa, Shiga, Japan) was reverse-transcribed into cDNA by the Prime Script RT Master Mix (DRR036A, TaKaRa, Shiga, Japan). SYBR Premix Ex TaqTM II (rr820A, TaKaRa, Shiga, Japan) was selected to test CD44 mRNA expression. The thermal profile for SYBR Green PCR was 95˚C for 30 s, followed by 40 cycles of denaturation at 95˚C for 5 s and annealing at 60˚C for 20 s. The primer sequences were as follows:

CD44 (Gen-Bank accession number, NM_012924.2), forward: 5’-CTCAAGTGGGAATCAAGACAGTGG-3’ and reverse: 5’-TGCAGACGGCAAGAATCAGAG-3’, GAPDH (Gen-Bank ID NM_017008.3), forward: 5’-GGCACAGTCAAGGCTGAGAATG-3’ and reverse: 5’-ATGGTGGTGAAGACGCCAGTA-3’. The amplification was normalized to GAPDH and the expression of CD44 mRNA was quantified by the relative quantitative 2^−ΔΔCt method.

Isolated cells from different groups were washed with PBS and lysed with lysates buffer (P0013, Beyotime, Shanghai, China) containing protease inhibitors (R0100, Solarbio, Beijing, China) and phosphatase inhibitor (AR1183, Solarbio, Beijing, China) for 30 min on ice. The supernatant was collected after centrifugation at 14,000 rpm for 15 min at 4˚C and the protein concentration was determined using BCA Protein Assay kit (23225, Thermo Fisher Scientific, Waltham, MA, USA). The protein sample (50 μg) was loaded, separated using SDS-PAGE, transferred to PVDF membranes at 15 eV for 15 min, and blocked with 5% non-fat milk for 1 h at room temperature. The membranes were then incubated with primary antibodies (i.e. CD44, pSmad2, Smad4, TGF-β₁, Collagen I) overnight at 4˚C. After washing 3 times, the membranes were incubated with horseradish peroxidase conjugated secondary antibodies (i.e. goat anti-Rabbit IgG and goat anti-Mouse IgG). Finally, the bands were detected with Bio-Rad imaging system assisting with ECL reagent kit according to the manufacturer’s instructions. The gray-scale values of straps were analyzed via ImageJ software, and the relative expression of proteins was normalized on β-actin or GADPH.

CD44 on the cells was silenced using small interfering RNA (siRNA). The cells in the log phase of growth were separately transfected with adenovirus targeting rat CD44 (Ad-CD44 shRNA1, Ad-CD44 shRNA2 and Ad-CD44 shRNA3) according to the manufacturer’s recommendations, which were designed by HANBIO Company (Shanghai, China). A similar adenoviral vector encoding the GFP gene (Ad-GFP NC) was used as a control. After adenovirus were co-incubated with cells for 4 h, the fresh medium was changed and the cells were continually cultured for another 32 h. Target cells showing green fluorescent protein (GFP)-positive were observed under a fluorescence microscope. The virus that induces the most significant decrease in CD44 expression was selected for ongoing experiments. All procedures involving virus operation were performed in a biosafety cabinet.

SPSS 21.0 software served as the statistical analysis tool. Data from six individual experiments were averaged as mean ± standard deviation and analyzed by one-way ANOVA. Pearson correlation test was used as correlation analysis. A p value of less than 0.05 was considered to be statistically significant.

The first passage of neonatal rat cardiac fibroblasts grew rapidly on day 1. The cell passaging or splitting of fibroblasts on day 2 displayed their typical spread morphology to form a confluent monolayer culture. Inactive fibroblasts were smaller, spindle or polygonal-shaped, and irregularly branched cytoplasm with a large ovoid nucleus, clear boundaries and good refraction (Figure 1(A)). The purity of isolated cardiac fibroblasts during cell preparations was evaluated using immunofluorescence staining. As shown in Figure 1(B), the majority of the cells were stained positively with vimentin and discoid in domain receptor-2 (DDR2), the markers of fibroblasts. Absence of staining in these cells with cardiac troponin I or von Willebrand Factor indicated that the cells isolated are not cardiomyocytes and endothelial cells. To identify whether cardiac fibroblasts attach to the plates during the culture period, DAPI (4’,6-Diamidino-2-Phenylindole, Dihydrochloride), a cell-permeable fluorescent DNA-binding dye, was applied. Intensity of fluorescence staining showed the smaller, more intensely stained nuclei, suggesting that cardiac fibroblasts equally distribute well (Figure 1(B)).

After fibroblasts were originally plated at equal density on day 2, the dose-response effect of Ang II on the proliferative capacity of cardiac fibroblasts was assessed. Compared with non-treated cells (control), stimulation of cardiac fibroblasts with extracellular Ang II for 24 h significantly increased fibroblast proliferation (0.9 ± 0.04 in 50 nM vs 0.6 ± 0.07 in control, p < 0.05, Figure 2(A)). The maximal enhancement was seen at a dose range between 50 - 100 nM. Although a higher dose of 300 nM or 1 μM also induced fibroblast proliferation, the proliferative degree of fibroblasts started to decline (Figure 2(B)). The calculated values from the control, it was shown that Ang II at different concentrations increased fibroblast proliferation by 1.31, 1.45, 1.43, 1.19 and 1.16 times, respectively. Based on these results, 50 nM of Ang II was selected for subsequent observations.

Transwell migration assay was used to verify the migration of Ang II stimulated cardiac fibroblasts. The cells were seeded in the top compartment of the transwell chamber and the number of migrated cells at the bottom medium through a permeable polycarbonate film was identified. As shown in Figure 2(C), the population of purple-stained cells was significantly increased compared

with the control after treated with Ang II at 50 nM. Addition of Ang II AT1 receptor blocker, telmisartan at a dose of 100 nM before Ang II stimulation significantly reduced the number of migrated cells (Figure 2(D)), suggesting an involvement of the AT1 receptor in cell migration.

Stimulation of cardiac fibroblasts by Ang II was accompanied with change in the expression of TGF-β1 [9]. Along with enhanced proliferation and migration of cardiac fibroblasts, the protein level of TGF-β1 was increased by 1.57 time after 50 nM of Ang II was added (Figure 3(A)), which was significantly blocked by telmisartan, suggesting that the release of TGF-β1 from activated fibroblasts is mediated by the AT1 receptor. Consistent with modulation in the production of TGF-β1 from fibroblasts by Ang II, the protein levels of pSmad2, Smad4 and collagen I were simultaneously upregulated by Ang II. Compared with the control, the protein levels of pSmad2, Smad4 and collagen I were increased by 1.49, 2.21 and 1.97 times, respectively, which all were blocked by telmisartan treatment (Figures 3(B)-(D)).

To study whether stimulation of Ang II AT1 receptor is associated with upregulation in the expression of CD44 mRNA and protein, RT-PCR using primers designed to recognize the fibroblast CD44 gene was performed on day 2 to obtain control baseline values for the cell transcriptional activity. Contrary to the expected basal expression of gene transcripts in control, RT-PCR amplified significant detectable amounts of CD44 mRNA in fibroblasts after the cells were exposed to 50 nM of Ang II, when compared to controls (Figure 4(A)). The quantitation of CD44 mRNA expression from a control level revealed that Ang II increases CD44 mRNA by 80.9% ± 4%. Consistent with upregulation of CD44 mRNA, the expression in CD44 protein was also enhanced by 37.9% ± 2%, as shown by Western blot analysis (Figure 4(B)). Addition of the AT1 receptor blocker, telmisartan at a concentration of 100 nM significantly reduced CD44 transcriptional activity (47.0% ± 3%) and protein level (26.2% ± 1%) relative to

Ang II stimulation, respectively, (all p < 0.05, Figure 4). Furthermore, Pearson correlation analysis showed that there is a linear positive correlation between CD44 and TGFβ1, suggesting a role of CD44 in the promotion of TGFβ1 upregulation (Figure 4(C)).

CD44 was silenced by Ad-CD44 shRNA (siCD44) to further confirm whether CD44 is involved in the activation of cardiac fibroblasts. To validate the effect of siCD44 on the suppression of CD44 in cardiac fibroblasts, RT-PCR and Western blot were carried out. As expected, as negative control, a fluorescent specific shRNA negative (Ad-GFP NC) showed the green fluorescent protein (GFP)-positive cells (Figure 5(A)), which was further confirmed by CD44 mRNA level (Figure 5(B)) and protein expression (Figure 5(C)). Out of the three shRNA constructs, shRNA1 presented the highest efficiency in silencing CD44 expression on cardiac fibroblasts, with a reduction of approximately 29.3% ± 1%, whereas in the other two, Ad-GFP shRNA2 and Ad-GFP shRNA3 did not show significant effects on the number of green fluorescent protein (GFP)-positive cells, when they are compared with the non-silencing control (Ad-GFP NC) (Figure 5(B)). Thus,

Ad-shRNA1 was chosen for sequential establishment of CD44-knockdown on cardiac fibroblasts. Consistent with the results of CD44 silencing, CD44 protein level was also significantly downregulated with Ad-shRNA1 (Figure 5(C)).

Collagen I was selected as a parameter to demonstrate the effect of CD44 knockdown on activation of cardiac fibroblasts. As shown in Figure 5(D), quantification of total collagen I with a computer-assisted densitometric analysis showed a significant increase in collagen I expression after Ang II treatment at 50 nM when compared with control, which was not altered by Ad-GFP NC. However, no significant change in the expression of collagen I was seen when CD44 was silenced with Ad-GFP shRNA1 even after Ang II was re-added, further supporting the role of CD44 in Ang II stimulated cardiac fibroblasts. Based on data demonstrated in this study, we proposed the signaling pathways of Ang II AT1 receptor/CD44 upregulation in mediating the production of collagen (Figure 6).