Velvet antler tissue was prepared as previously described (Xiao et al., 2017). Fresh VA, produced in the Dongfeng Sika Deer Farm, Liaoyuan City, Jilin Province of China, was obtained from 4-year-old sika deer (Cervus nippon Temminck), which were at the early stage of their fast-growing antlers. The deer were bred in Jilin Province of China and identified by Professor Chunsheng Liu of the Department of Chinese Medicine Identification at the Institute of Traditional Chinese Medicine, Beijing University of Chinese Medicine. After administering general anesthesia, the antlers were removed by cutting the proximal region with a surgical hand saw and washed with 75% ethanol. Blood was then drawn from the VAs with a vacuum instrument (SHZ-IIIB, Tan Shi Vacuum Equipment Co., Ltd., Linhai, Zhejiang, China) for 6 h in 4°C, sliced, and freeze-dried by lyophilizer (FreeZone Plus 12 Liter Cascade Console Freeze Dry System, -84°C, LABCONCO, Kansas City, MO, United States). The temperature was -60°C and the baro-vacuum was 12 Pa until the VA quality change was less than 0.1 g. With superfine grinding technology, the freeze-dried VA was superfinely comminuted under the cryogenic environment and 54.08 g lyophilized VA powder was created. The dehydration percentage of VA was 67.02%. The powder was stockpiled in sealed tubes and numbered XX-DGF01 to XX-DGF04.

The method has been described briefly before (Xiao et al., 2017). Lyophilized VA powder was weighed and placed in a large beaker, mixed with ultrapure water at the proportion of 1:10, with vortex oscillation blending, then the lyophilized VA powder was mixed for 5 s and then paused for 15 s. Using homogenate and immersion methods, with 12 cycles in an ice-bath, the VA was ultrasonically extracted for 2 h. The supernate was obtained in a 50-ml centrifuge tube and the precipitate was removed by centrifugation (12,000 rpm, 30 min), the supernate was stored at -80°C for less than 12 h, and then freeze-dried by lyophilizer (FreeZone Plus 12 Liter Cascade Console Freeze Dry System, -84°C, LABCONCO, Kansas City, MO, United States) for 72 h. The precipitate was processed according to the above steps three times, and then all of the freeze-dried VA-pro were mixed, numbered from XX-VApro01 to XX-VApro03, and stored at -80°C. The protein content was measured using a bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific, Waltham, MA, United States). According to the specifications, the diluted bull serum albumin (BSA) standards and BCA working reagent (WR) were prepared first: 200 μl of WR and 25 μl of samples or 25 μl of diluted albumin (BSA) standards were added into the 96-well plate and mixed at 37°C for 30 min. The plate was removed and set at room temperature for 10 min, and the optical density (OD) was counted by microplate spectrophotometer (Spectra MR, Dynex, United States). The VA-pro mass concentration was 465.56 μg/mg.

We digested 120 μg of VA-pro with 4 μg of trypsin for 24 h at 37°C and we added 1 μg of trypsin for another 12 h at 37°C, then the salt was removed with a C18 column (Phenomenex StrataTM, Torrance, CA, United States) and the desalted peptide mixture was obtained. Nano LC-MS/MS analysis was performed as previously described (Xiao et al., 2017).

After HPLC-MS/MS measurement, the raw data were imported into the PD (Proteome Discoverer 1.3, Thermo Fisher Scientific, Waltham, MA, United States) to be filtered. The collected data files were processed and sent to a MASCOT server (Version 2.3.0, Matrix Science) for peptide identification using the UniProt database¹ with the Ruminantia taxonomy constraint (The retrieval parameters shown in Table 1).

The sample was annotated based on the GO database² and the KEGG database³ .

The total number of VA-pro identified by the MASCOT search was 386 using the taxonomy Ruminantia (Uniprot), so there were 145 uncharacterized proteins. The 386 pI ranged from 4.14 to 11.57, with a pI of 4.0 to 7.0 (68.4%) and 8.0 to 9.0 (12.8%) for most proteins (Figure 1A). The molecular weight (MW) ranged from 6 to 800 kDa, with most of the proteins from 10 to 80 kDa (74.4%) and 100 to 200 kDa (12.8%) (Figure 1B). For the VA-pro function analysis (GO annotation), part of cellular component results are shown in Figure 1C. As for the biological processes, 211 proteins were involved in the metabolic process, 102 proteins were involved in metabolic regulation, 151 proteins were involved in the biological process of response to stimulus, 110 proteins were involved in the localized biological processes, and 14 proteins were involved in the growth of related biological processes. In addition, there were 172 proteins involved in cell metabolism, 78 proteins involved in the regulation of cell metabolism, 90 proteins involved in cellular responses to stimuli, 59 proteins involved in cellular localization, and 10 proteins involved in the biological process of cell growth (Figure 1D). Further, 38 proteins were involved in the cell apoptosis process, 27 proteins were related to cell migration, 4 proteins were closely related to cell proliferation, and 2 proteins played a role in mitochondrial changes during apoptosis.

The CMECs were cultured from 2-day-old SD rat pup hearts using mature and accepted methods. The protocol was approved by the Animal Care and Welfare Committee of the Clinical Medicine Institute in China-Japan Friendship Hospital. Rat pups’ hearts were excised, minced, and digested in 2% Collagenase II solution (Gibco, Grand Island, NE, United States) for 7 min at 37°C five times. The digested product was collected in a centrifuge tube containing serum and was filtered through a 70-m nylon cell strainer, then the cells were collected and adhered differentially. CMECs were cultured in Endothelial Cell Medium (ECM, Sciencell, Carlsbad, CA, United States), consisting of 500 ml of basal medium, 25 ml of fetal bovine serum (FBS), 5 ml of endothelial cell growth supplement (ECGS), and 5 ml of penicillin/streptomycin (P/S) solution in standard incubator conditions. CMECs were determined by the presence of vWF VIII (Abcam, Cambridge, United Kingdom) and PECAM-1/CD31 (Abcam, Cambridge, United Kingdom).

After two passages, CMECs were used for the experiment, treated with ECM for 24 h, then the control cells were cultured in complete ECM while other cells were washed with PBS twice, then cultured in FBS-free medium with different VA-pro concentrations. CMECs remained in culture at 37°C with hypoxia conditions (5% CO₂-95%N₂) for 46 h. The control cells were cultured in a regular incubator at 37°C, 5% CO₂.

To find an effective dose range of VA-pro, we set up 6 dose groups (0.25, 0.5, 1, 2, 4, and 8 mg/ml, with no FBS in the medium) within the ischemia IH group (no FBS in the medium, hypoxia conditions) and the hypoxia group (H group, 10% FBS, hypoxia conditions). The control group of cells was cultured with complete medium at 37°C, 5% CO₂.

Cell viability was determined using the Cell Titer 96^®Aqueous One Solution Reagent (MTS assay, PROMEGA, United Kingdom). Cells were cultured in a 96-well plate. Before measuring, each well medium was replaced by 100 μl of new medium and we added 20 μl of reagent, then we cultured the cells for 3 h at 37°C, 5% CO₂. A blank well with the same volume of medium but without cells was as a control well. The optical density (OD) was measured by microplate spectrophotometer (Spectra MR, Dynex, United States).

Cell proliferation was determined using an EdU kit (RIB BIO, Guangzhou, China), with blue and red indicating the Hoechst 33342- and EdU-labeled cells, respectively. After adding the EdU (20 μM), cells were cultured for 4 h. Then, cells were fixed with 4% (w/v) formaldehyde in PBS for 10 min, and other testing steps were performed in accordance with the operation manual. The results were detected by a high-intensity scanning system and analyzed by MetaXpress software (Molecular Devices, Sunnyvale, CA, United States).

The Matrigel was thawed (BD, San Jose, CA, United States) at 4°C for 3 h. We added 100 μl of Matrigel to each well of a 48-well plate and then left the Matrigel to solidify at 37°C for 30 min. With 3 × 10⁴ CMECs seeded on top of the Matrigel layer in 250 μl of culture medium, the cells were subsequently incubated at 37°C, 5% CO2. The tube network was observed by bright field microscope (Leica Microsystems, Germany) during a 12-h period. The branch point cells were calculated in 4 random fields per well. These images were analyzed using NIS Elements AR Analysis 4.10.00 (Nikon, Tokyo, Japan).

We cultured 3 × 10⁵ CMECs in 12-well plates for 24 h. Subsequently, confluent cell monolayers were scratched with a P200 pipette tip to make three parallel “wounds” in each well, and we washed the cells with PBS two times. After cells were incubated with different treatments, the wounds were photographed and counted using a bright field microscope, and the pictures were analyzed with Prism 5.0 (GraphPad Inc., La Jolla, CA, United States).

Invasion chambers (Corning, NY, United States) were also used to detect CMEC migration. We added 150 μl of cell FBS-free suspension (1 × 105 cells/ml) to the inserts and medium with different drug concentrations to the lower chamber. After the treatment, non-invasive cells in the upper chamber were removed with cotton swabs and invasive cells were stained with DAPI. We counted six random fields for each chamber (magnification, ×200) using a microscope (Leica Microsystems, Germany).

To quantify cellular apoptosis in the early and late stages, cells were seeded in a 6-well plate, and after treatment, they were harvested and processed for the apoptosis assay using an Annexin V-FITC/PI kit (BD, San Jose, CA, United States) following the indicated protocol specified by the manufacturer. They were examined using an EPICS XL flow cytometer and EPICS XL ADC system (Beckman Coulter, Porterville, CA, United States). Data were analyzed using Cell Quest software (BD, San Jose, CA, United States).

The mitochondrial membrane potential (MMP) was measured using a JC-1 kit (5,5′,6,6′-tetrachloro-1,1′,3,30′-tetraethyl-imidacarbocyanine, BD, San Jose, CA, United States). We dissolved JC-1 in DMSO and diluted it with medium at 1:99 when it was used. We stained 1 × 10⁶ cells with 500 μl of working solution at 37°C for 15 min in the dark, washed with 1× assay buffer twice and resuspended in 1× assay buffer, followed by analysis using flow cytometry. A minimum of 10,000 events was acquired to assess JC-1 fluorescence in both FL-1 (monomers, green fluorescence) and FL-2 channels (aggregates, red fluorescence) using the EPICS XL flow cytometer.

To further explore the mechanisms involved in the effect of VA-pro on CMECs and the possible involvement of PI3K/Akt signaling, an inhibitor group was set and LY294002 (10 μM) was added to the medium. IH-injured cells were washed and lysed with RIPA lysis buffer (containing protease inhibitors and phosphatase inhibitors). With 20 μl/well loaded, the proteins were resolved on a 12% polyacrylamide gel and transferred to a nitrocellulose membrane. Membranes were blocked with 5% non-fat dry milk for 2 h, then incubated with primary antibodies against Akt (1:1000, CST), p-Akt (1:1000), Bcl-2, Bax, caspase-3, cleaved caspase-3 (all from CST, Boston, MA, United States), and GAPDH (1:1000, Abcam, Cambridge, United Kingdom) overnight at 4°C. Membranes were then washed and further incubated with goat anti-rabbit antibody (1:5000, LI-COR, Lincoln, NE, United States) at room temperature for 1 h. Bands were visualized and analyzed using Odyssey SA and Image Studio version 4.0 (LI-COR Biosciences, Lincoln, NE, United States).

For immunofluorescence, cells grown on glass cover slips were fixed with 4% paraformaldehyde at room temperature for 15 min. Cells were washed three times with PBS for 5 min and then they were treated with 0.3% triton-100, blocked with 3% BSA (PBS) for 1 h, and then were immunolabeled with primary antibody (anti-CD31 at 1:100, anti-vWF at 1:200) overnight at 4°C. They were then washed with PBS, and immunolabeled with Alexa 488-conjugated secondary antibody or Alexa 555-conjugated secondary antibody (1:200, all from CST, Boston, MA, United States) for 1 h at room temperature in the dark. Cells were washed in PBS, and the nuclei were stained with DAPI for 10 min. Glass cover slips were upended on the glass slide before storing at 4°C. The cells were imaged and counted using an inverted fluorescent microscope (Nikon, Tokyo, Japan) and Leica TCS SP5 confocal microscope (Leica, Germany). Digital images were prepared using NIS Elements AR Analysis 4.10.00 and Adobe Photoshop CS6 software (Adobe Systems, San Jose, CA, United States).

Quantitative data were presented as means ± SD and were analyzed by GraphPad Prism 5 (GraphPad Software, Inc., United States). Statistical differences among groups were assessed by ANOVA followed by Tukey’s or Dunnett’s posttest. P < 0.05 was considered to be significant. All of the experiments were set in three parallel groups in support of a three-time repeat.

After two passages, CMECs were ovoid and could form tubes on the Matrigel (Figures 2A,B). Cells were identified by their surface markers vWF and CD31 (Figure 2C). The morphology, tubes, and the positive surface marker manifested a successful culture of CMECs.

Cardiac microvascular endothelial cell viability in the IH group was significantly reduced (P < 0.001). VA-pro treatment resulted in an improvement in CMEC viability, and the best dose was 1 mg/ml (P < 0.001) because the group of cells almost had the same viability as the control group. Other dose groups were no different when compared with the H group, except for 4 and 8 mg/ml (Figure 3A). As shown in Figures 3B,C, IH suppressed copies of DNA to hold back CMEC proliferation. VA-pro treatment resulted in better cell proliferation, especially at 0.5 and 1 mg/ml (P < 0.001). These resulted suggested that VA-pro improved the viability of CMECs subjected to IH.

Normal CMECs were negative for both AV and PI, early apoptotic CMECs were AV positive and PI negative, and late apoptotic and necrotic CMECs were positive for both AV and PI. Severe apoptosis was induced after CMECs were cultured in IH conditions for 46 h. Flow cytometry results are shown in Figure 4. When compared to the control group, the IH group of normal CMECs decreased significantly (P < 0.001), while in the VA-pro groups with a concentration of 0.25, 0.5, and 1 mg/ml, the number of apoptotic cells decreased (P < 0.001). These results suggested that VA-pro reduced the apoptosis of CMECs.

The horizontal axis of the flow cytometric dot plot is the blue fluorescence signal (JC-1 monomers) and the vertical axis is the red fluorescent signal (JC-1 aggregates). Figure 5 shows that there is a lower red fluorescent intensity and a higher blue fluorescent intensity in the IH group of CMECs (P < 0.001); this means that MMP decreased markedly when CMECs were subjected to IH, but MMP was increased when the cells were treated with VA-pro, especially at 0.5 and 1 mg/ml (P < 0.05). It indicated that VA-pro retained the stability of the MMP of CMECs that were cultured in the IH conditions.

For the scratch assay, as shown in Figures 6A,B, the control group migration rate was set at 100%, and quantitative analysis indicated that the migration rate of the IH group was decreased (P < 0.001). VA-pro treatment (0.25, 0.5, 1, and 2 mg/ml dose) reversed the migration rate as the wound areas of these groups were decreased compared to the IH group, while the 1 mg/ml and 2 mg/ml VA-pro groups had the best migratory ability (P < 0.001). This indicated that VA-pro promoted the migration of CMECs subjected to IH. Meanwhile, a better cell morphology was observed in the 1 mg/ml-VA-pro group under a microscope. For transwell assay, only 0.5 and 1 mg/ml dose groups showed good results (P < 0.05) (Figures 6C,D).

In vitro, endotheliocytes can spontaneously form a tubular capillary-like network in Matrigel culture. As shown in Figure 7A and quantified in Figure 7B, CMECs’ tubule formation ability was decreased when CMECs were subjected to IH (P < 0.001). VA-pro treatment (0.5, 1, and 2 mg/ml dose) potentially reversed the CMECs’ tubule formation ability that had been reduced by IH, especially at 1 mg/ml (P < 0.001), so VA-pro maintained the tube formation ability of CMECs.

Considering the results of Figures 3–7, we only retained the 1 mg/ml VA-pro group. As shown in Figure 8, a lower expression of phosphorylated Akt (Figures 8A,B) and Bcl-2 (Figures 8A,D), and a higher expression of Bax (Figures 8A,C) and cleaved caspase-3 (Figures 8A,E) were observed in the IH group. VA-pro could reverse these changes. Further, the inhibitor of PI3K/Akt (LY294002) treatment suppressed the effects of VA-pro on Akt, Bax, Bcl-2, and cleaved caspase-3 expression, indicating that VA-pro protected the CMECs challenged with IH by activating the PI3K/Akt signaling pathway.