Male Sprague-Dawley (SD) rats weighing 180–200 g were obtained from the Peking University Health Science Center Animal Center (Beijing, certificate no. SCXK 2006-0008) and handled according to the guidelines of the Peking University Health Science Center Animal Research Committee. All experimental procedures were approved by Peking University Biomedical Ethics Committee Experimental Animal Ethics Branch, complying with the “Guidelines for the Care and Use of Laboratory Animals”, published by the National Institutes of Health (Zheng et al., 2019).

T₈₉ (270 mg per capsule, lot number: 20160601S) was obtained from Tasly Pharmaceutical Co., Ltd. (Tianjin, China), which was prepared from water-ethanol extract of SM, PN, and borneol under the guideline of Good Manufacturing Practice and Good Laboratory Practice verified by the Chinese, and United States Government agencies. The doses of T₈₉ for rats were 111.6 and 167.4 mg/kg, equivalent to low and high dose for human in FDA phases II and III clinical trials. ISO was purchased from Tokyo Chemical Industry Co., Ltd. (product number: I0260).

The SD rats were randomly divided into five groups: (1) Sham group (Sham); (2) Treatment with 167.4 mg/kg T₈₉ in Sham group (Sham + T₈₉ 167.4); (3) ISO-induced model (ISO); (4) Treatment with 111.6 mg/kg T₈₉ in ISO-induced model (ISO + T₈₉ 111.6); and (5) Treatment with 167.4 mg/kg T₈₉ in ISO-induced model (ISO + T₈₉ 167.4). The rats in the Sham and ISO-induced groups received subcutaneous injections of normal saline and ISO saline solution (10 mg/kg), respectively, at 24 h intervals for the inchoate three consecutive days and then at 48 h intervals for the later 4–15 days. Rats treated with T₈₉ in the ISO-induced groups were given corresponding drugs by intragastric administration for 15 consecutive days starting from 5 min before ISO injection. The animals in the Sham + T₈₉ 167.4 group was given T₈₉ at 167.4 mg/kg by intragastric administration for 15 consecutive days, while those in the Sham group received normal saline instead the same way.

During the 15-day observation, survival of each group was monitored daily for 15 days. At the end of the experiment, survival curves were plotted and rats were anesthetized with 2% pentobarbital (60 mg/kg) by peritoneal injection.

Two hours after ISO injection on Day 15 of the experiment, myocardial blood flow (MBF) in each group was measured using Laser-Doppler Perfusion Imager (PeriScan PIM3 System; PERIMED, Stockholm, Sweden) equipped with a computer. Briefly, heart was exposed after left thoracotomy, and a computer-controlled optical scanner directed a low-powered He–Ne laser beam over the exposed heart. The scanner head was positioned in parallel to the surface of the heart at a distance of 18 cm, and the beam illuminated the tissue to a depth of 0.5 mm. A color-coded image to denote specific relative perfusion level was displayed on a video monitor, and all images were evaluated with the software LDPIwin 3.1 (PeriScan PIM3 System; PERIMED, Stockholm, Sweden). The MBF magnitude is represented by different colors, with blue to red denoting low to high (Wei et al., 2013).

Heart function was tested by a biofunction experiment system BL-420F (Chengdu Taimen Technology Ltd., Chengdu, China), which was connected to a cannulation inserted into the left ventricle through the right carotid artery. Heart rate (HR), left ventricular systolic pressure (LVSP), left ventricular diastolic pressure (LVDP), left ventricular end diastolic pressure (LVEDP), left ventricular maximum upstroke velocity (+dp/dt max), and left ventricular maximum descent velocity (-dp/dt max) were evaluated 2 h after ISO injection on the Day 15 of the experiment (Zheng et al., 2019).

On Day 15 of the experiment, 2 h after ISO injection, hearts were removed and washed with normal saline. Both heart weight and body weight were weighed, and heart weight to body weight (HW/BW) was calculated to evaluate the cardiac hypertrophy response to the procedure (Huang R. et al., 2019).

At the end of the experiment, rat hearts were removed and fixed in 4% paraformaldehyde. Serial paraffin sections (5 μm thickness) were prepared and stained with hematoxylin eosin (HE) as routine (Yan et al., 2018).

The immunofluorescence staining of wheat germ agglutinin (WGA, Invitrogen) and Connexin 43 (CX 43) was performed. Images of immunofluorescence were observed at ×40 magnification of objective with a Laser Scanning Confocal Microscope (TCS SP5, Leica, Mannheim, Germany). Cardiomyocyte cross-section area was determined on sections stained with WGA in randomly selected five fields, calculating the average of cross-section areas of 10 cells in each, using Image-Pro Plus 6.0 (Media Cybernetic, Bethesda, MD, United States) (Huang R. et al., 2019).

An approximately 1 mm³ fresh myocardial tissue block was taken from the left ventricle. Tissues were fixed with 3% glutaraldehyde and postfixed with 1% osmium tetraoxide. The specimens were processed as routine for ultrathin sections. Sections were stained with uranium acetate and lead citrate, and ultrastructural changes were then evaluated by using a transmission electron microscope (JEM-1230; JEOL, Tokyo, Japan) (Wei et al., 2013).

Enzyme-linked immunosorbent assay (ELISA) was undertaken using a specific kit indicated by a microplate reader (Multiskan MK3, Thermo, New Bedford, MA, United States), respectively, to determine the content of creatine kinase MB (CK MB), and Troponin I (cTn I) in plasma (Huanya Biomedicine Technology Co., Ltd., Beijing, China); pyruvic acid, 3-hydroxybutyric acid (3-HYA), atrial natriuretic peptide (ANP), thiobarbituric acid reacting substances (TBARS), ATP, and AMP in the left ventricle (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Likewise, the activities of mitochondrial ETC complex I, IV, and V were also detected by ELISA kit according to the manufacturer’s instructions (Abcam, Cambridge, MA, United States) (Huang D. D. et al., 2019).

At the end of 15-day observation, left ventricular tissue proteins from three groups (Sham, ISO, ISO + T₈₉ 167.4) were extracted and quantified using the BCA Protein Assay Kit (Thermo Fisher Scientific, United States). The proteins were then labeled with tandem mass tags (TMT). TMT6/10 (Pierce, Rockford, IL, United States) with different reporter ions (126–131 Da) were applied as isobaric tags for relative quantification. Then the labeled peptide aliquots were combined for fractionation and further LC-MS analysis. The LC-MS/MS analysis was carried out in CapitalBio Technology by using Q Exactive mass spectrometer (Thermo Fisher Scientific, United States). Twenty most intense precursor ions from a survey scan were selected for MS/MS detected at a mass resolution of 35,000 at m/z of 400 in Orbitrap analyzer. All the tandem mass spectra were produced by higher-energy collision dissociation (HCD) method. The MS/MS data was collected and searched against the Oryctolagus cuniculus database using the SEQUEST algorithms. A protein with fold change ≥1.5, and the presence of least two unique peptides with a P-value < 0.05, was considered to be a differentially expressed protein (DEP). Finally, DEPs were analyzed by reference to three key databases: Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Reactome and the Clusters of Orthologous Groups (COG) (Park et al., 2019). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (Perez-Riverol et al., 2019) partner repository with the dataset identifier PXD024641.

Western blotting assay was performed to verify the differentially expressed proteins obtained from quantitative proteomics analysis. Myocardial tissues were taken from left ventricle, and total proteins were extracted using a protein extraction kit (Applygen Technologies, Beijing, China) and mixed with 5× electrophoresis sample buffer. After electrophoresis, the separated proteins were transferred to polyvinylidene difluoride membrane. After being blocked with 5% non-fat dry milk, the membrane with target proteins was incubated overnight at 4°C with antibodies against Eno-1, Mcee, Bdh1, Ces1c, Cbr4, ND2, Cox6a, Cox17, ATP 5j, Atl3, heat shock protein (HSP)70, HSP40, extracellular signal-regulated kinase (ERK) 1/2, P-ERK1/2, and CX 43 (Abcam, Cambridge, MA, United States). Blotted antibodies were visualized by HRP-conjugated second antibody (Cell Signaling Technology) and ECL detection system (Applygen Technologies, Beijing, China). Densitometric analyses of blots were performed using the Quantity One image analyzer software (Bio-Rad, Richmond CA, United States). The result of each band was expressed as relative optical density compared with internal controls of β-actin (Zheng et al., 2019).

All parameters were expressed as means ± SE. Statistical analysis was performed using one-way ANOVA, followed by Bonferroni test for multiple comparisons. A probability of less than 0.05 was considered statistically significant.

As shown in Figure 1A, no rats in the Sham (n = 28) and Sham + T₈₉ 167.4 group (n = 28) died during the whole experiment. The 15-day survival rate of rats in ISO group (n = 58) was 48.28%, significantly lower than that in the Sham group (^∗P < 0.05 vs. Sham). However, compared with the ISO group, the survival rate of T₈₉ treatment groups were significantly higher, which were 62.22% (n = 45 in the ISO + T₈₉ 111.6 group) and 68.29% (n = 41 in the ISO + T₈₉ 167.4 group), respectively (^#P < 0.05 vs. ISO).

In view of improvement of rat survival, we wonder T₈₉ could preserve the cardiac perfusion. Therefore, the effects of T₈₉ on MBF were assessed. The representative images are shown in Figure 1B, which were manifested as distinct color with the red color representing the highest MBF and the blue color as the lowest. Impressively, MBF decreased after ISO in comparison with the Sham group (^∗P < 0.05 vs. Sham), whereas treatment with T₈₉ at dose of 167.4 mg/kg significantly attenuated ISO-induced decline in MBF (^#P < 0.05 vs. ISO). The above results were confirmed by quantitative evaluation, as shown in Figure 1C.

The improvement of heart function contributes to elevation of rat survival rate, so heart function was detected in different groups. As demonstrated in Figure 1D, ISO administration caused a noticeable decline in LVSP (d2) and +dp/dt max (d3), and an inclement in LVDP (d4), LVEDP (d5), and -dp/dt max (d6) (^∗P < 0.05 vs. Sham), indicating an impairment on heart function. However, this impairment was significantly alleviated by treatment with T₈₉, especially at dose of 167.4 mg/kg (^#P < 0.05 vs. ISO).

Figure 2A displays the myocardial morphology stained by HE in Sham (A1 and A6), Sham + T₈₉ 167.4 (A2 and A7), ISO (A3 and A8), ISO + T₈₉ 111.6 (A4 and A9), and ISO + T₈₉ 167.4 (A5 and A10) groups. The upper panels (A1 to A5) are representative micrographs captured under ×10 magnification of objective, and the lower panels (A6–A10) under ×20 magnification of objective. Compared with the Sham group, ISO elicited a distinct morphological injury, such as myocardial fiber disarrangement and disruption, leukocyte infiltration, and connective tissue hyperplasia. Obviously, T₈₉ administration significantly preserved myocardium structure after ISO.

The myocardial ultrastructure in varies groups is presented in Figure 2B. In comparison with the Sham group (B1), ISO (B3) evoked significant alteration of myocardial ultrastructure, including myocardial fibers rupture, tissue edema, and mitochondrial swelling. Apparently, the ISO-induced alterations of myocardial ultrastructure were ameliorated in groups receiving T₈₉ (B4 and B5).

To investigate the effect of T₈₉ on myocardial injury, the level of CK MB and cTn I in plasma, and content of ANP in left ventricle were estimated by ELISA. Noticeably, the levels of CK MB (Figure 2C) and cTn I (Figure 2D) in plasma were significantly increased by ISO in comparison with the Sham group (^∗P < 0.05 vs. Sham), while treatment with T₈₉ remarkably inhibited ISO-induced elevation of CK MB and cTn I (^#P < 0.05 vs. ISO). As shown in Figure 2E, ISO evoked obvious increase of ANP compared with the Sham group (^∗P < 0.05 vs. Sham). While T₈₉ 167.4 mg/kg treatment markedly reduced the elevation of ANP content caused by ISO (^#P < 0.05 vs. ISO).

TMT-quantitative proteomics were used to investigate the protective mechanisms of T₈₉. A total of 4,092 proteins and 30,319 peptides were successfully identified by LC-MS/MS, and 1201 DEPs were detected among the three groups. A total of 516 DEPs were identified between the ISO group and Sham group, 293 of which were upregulated, while 223 were downregulated. The Volcano Plot of ISO vs. Sham is shown in Figure 3A. The top 30 GO terms between the ISO and Sham group that showed remarkable enrichment are shown in Figure 3B. Biological process (BP) analysis showed that the majority of the DEPs identified between ISO and Sham were classified as “negative regulation of enzyme activity” and “biological metabolic process.” With regard to cellular component (CC) terms, most of the DEPs identified between the ISO and Sham groups were associated with the extracellular and membrane-bounded vesicles. With regard to molecular function (MF), the identified DEPs were predominantly involved in binding processes, and in particular, ionic binding and protein binding. Next, the biological pathway enrichment of DEPs identified between ISO and Sham groups were statistically analyzed. As shown in Figure 3C, either KEGG pathway or Reactome annotation displayed that the main signaling pathways undergoing modulation were those related to metabolism. Furthermore, KEGG pathway detail analysis was performed and showed in Figure 3D, demonstrating 46, 23, and 16 DEPs being related to carbohydrate metabolism, lipid metabolism and energy metabolism, respectively.

Statistical analysis of pathway enrichment between ISO + T₈₉ 167.4 group and ISO group disclosed that the greatest changes occurred in fatty acid metabolism and ATP formation pathway (Figure 4A). The heat maps of DEP variations among Sham, ISO, and ISO + T₈₉ 167.4 groups are displayed in Figure 4B, including Eno1, Mcee, Bdh1, Apoc2, Decr1, Cbr4, Acaa2, ND2, Cox6a, Cox17, ATP5mc1, ATP5pf, etc.

Carbohydrate metabolism alteration was found as the main metabolic pathway changed after ISO challenge. As shown in Table 1, in comparison with the Sham group, ISO reduced 31 proteins and increased 15 proteins that are associated with carbohydrate metabolism. The 31 reduced proteins were mainly involved in citrate cycle, gluconeogenesis, and other pathways that inhibit glucose synthesis. Whereas, the 15 increased proteins were mainly enriched in glycolysis, indicating enhancement of glucose oxygenolysis. These alterations of carbohydrate metabolism implied that more carbohydrates were mobilized to provide energy through glycolysis after ISO challenge. Of the above changes, T₈₉ prominently inhibited the expression of Eno1, which was confirmed by Western blot (Figure 5A).

Carbohydrate metabolism alteration is closely linked to lipid metabolism. As shown in Table 2, ISO challenge elicited upregulation of seven lipid metabolism proteins and downregulation of 16 proteins. Among them, the enhancement of Mcee, Bdh1, and Ces1c were apparently reversed by T₈₉, which was confirmed by Western blotting analysis (Figures 5A,B). In contrast, the 16 reduced proteins by ISO were mainly enriched in lipid digestion, mobilization and transport, and fatty acid β-oxidation. The remarkable decline of these proteins suggested an inclement of fatty acid metabolism after ISO. The above inclement was partly alleviated by T₈₉, manifesting as the significant enhancement of Apoc2, Decr1, Acaa2, and Cbr4 (Figure 4B).

Consistent with enhancement of glycolysis and inclement of fatty acid oxidation after ISO, ISO evoked an obvious increase of pyruvic acid (Figure 5C), whereas a remarkable decline of 3-HYA (Figure 5D) in the left ventricle (^∗P < 0.05 vs. Sham). The above alterations induced by ISO were significantly alleviated by T₈₉ (^#P < 0.05 vs. ISO).

The alterations in glycolipid metabolism inevitably lead to energy metabolism disorders, which mainly manifests as oxidative phosphorylation dysfunction. Not surprisingly, ISO reduced 16 energy metabolism proteins, of which 13 proteins (ND1, ND2, Ndufa5, Ndufa10, Ndufa11, Ndufab1, Cox5a, Cox6b1, Cox6a2, Cox17, ATP5fle, ATP5g, and ATP5j) were subunits of mitochondrial ETC complexes, engaged in oxidative phosphorylation. Excitingly, five subunits reduced by ISO were significantly reversed by T₈₉, they were ND2 (subunit of complex I), Cox6a, and Cox17 (subunits of complex IV), ATP 5g, and ATP 5j (subunits of complex V), which was confirmed by Western blotting analysis (Figures 5E,F). Accordingly, the activities of complex I, IV, and V were remarkably inhibited after ISO (^∗P < 0.05 vs. Sham) and were obviously restored by T₈₉ (^#P < 0.05 vs. ISO) (Figure 5G). These data highlight the involvement of mitochondrial function in the beneficial role of T₈₉.

To further evaluate the effect of mitochondrial function on oxidative stress and energy metabolite, the contents of TBARS, AMP, and ATP in the left ventricle were determined by ELISA. As shown in Figures 5H,I, ISO elicited an obvious elevation of TBARS and a significant decline of ATP/AMP compared with the Sham group (^∗P < 0.05 vs. Sham). On the contrary, T₈₉ at 167.4 mg/kg significantly reversed the alteration of TBARS and ATP/AMP caused by ISO (^#P < 0.05 vs. ISO).

In addition to metabolism-related proteins, proteomics analysis showed that T₈₉ could also regulate other proteins, such as Hspa1a (HSP70), and Dnajc8 (HSP40). Western blotting examination (Figures 6A–H) demonstrated that ISO-induced elevation of HSP70, HSP40 and p-ERK 1/2 and reduction of CX43, indicating that these proteins are involved in cardiac hypertrophy induced by ISO. Nevertheless, all the alteration caused by ISO was significantly reversed by T₈₉ treatment (^#P < 0.05 vs. ISO).

Wheat germ agglutinin is used to demarcate cell boundaries for determination of cardiomyocyte size. CX 43 is immunolabeled in the interstitial disk as markers of the gap junction of cardiomyocytes. Figure 7A shows the images of WGA and CX 43 by immunofluorescent staining in each group, where WGA is red and CX 43 is green. Obviously, in the Sham (A1) and Sham + T₈₉ 167.4 (A2) groups, the myocardial boundaries stained by WGA were complete and distinct, and CX 43 expression was abundant. Compared with the Sham group, the cardiomyocyte boundary was notably enlarged and CX 43 expression was significantly decreased in ISO group (A3). However, T₈₉ treatment (A4 and A5) inhibited the enlargement of cardiomyocyte boundary and the breakdown of CX 43 caused by ISO.

The quantitative analysis of cardiomyocyte cross-sectional area is presented in Figure 7B, revealing that ISO elicited a remarkable augment on cardiomyocyte cross-sectional area, which was alleviated by T₈₉ (^∗P < 0.05 vs. Sham, ^#P < 0.05 vs. ISO). To evaluate the hypertrophic response to ISO, the ratio of HW/BW was measured. As noticed in Figure 7C, ISO educed obvious increase of HW/BW compared with the Sham group (^∗P < 0.05 vs. Sham), whereas T₈₉, especially at dose of 167.4 mg/kg, significantly inhibited ISO-induced elevation of HW/BW (^#P < 0.05 vs. ISO).