All experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of the First Affiliated Hospital of Dalian Medical University (AEE20027). The wild-type (WT) male C57BL/6J mice were fed under specific pathogen-free conditions for 8–10 weeks. Mice were anesthesia with 1.5% isoflurane, then the osmotic pumps (Alzet, 1,002; DURECT, Cupertino, CA, United States) were subcutaneously implanted to deliver angiotensin (Ang) II (1,000 ng kg⁻¹ min⁻¹) for 14 days to construct a hypertensive cardiac remodeling model. Then anti-VCAM-1 neutralizing antibody (0.1 or 0.2 mg, BE0027, Bioxcell) or control IgG was intraperitoneally (i.p.) injected every 2 days as previously described (Yin et al., 2022).

The mouse systolic blood pressure (SBP) of each group was measured and recorded with a tail-cuff instrument (BP-98A, Softron, Japan). All mice were placed in a restraint device on a warming platform to keep calm for at least half an hour. Then the SBP was measured with a tail cuff after mouse heart rate was stabilized as previously described (Wang et al., 2016). The mouse cardiac function was measured by echocardiography after Ang II infusion for 14 days as previously described (Wang et al., 2018). Briefly, all mice were anesthesia with 1.5% isoflurane and keep heart rate steady, then placed on a warming platform. The M-model echocardiographic analysis was performed by a 30 MHz probe (Vevo 1,100 system; VisualSonics, Toronto, Ontario, Canada). The parameters of ejection fraction, fractional shortening, ventricular wall and internal dimension at end-diastole and end-systole were captured and calculated.

All mice were anesthesia with an overdose of isoflurane (>5%), and hearts were carefully collected. The hearts were fix by 4% paraformaldehyde and dehydrated, and then embedded in paraffin or OCT. Cardiac paraffin sections (4 μm) were performed hematoxylin-eosin (HE) staining, Masson’s trichrome staining and immunohistochemical staining with antibodies against VCAM-1 (1:200, Proteintech), α-smooth muscle actin (α-SMA, 1:200, Arigo), CD68 (1:200, Abcam) and nitrotryrosine (1:200, Bioss) as previously described (Wang et al., 2018). Eight-micrometer frozen sections were subjected to wheat germ agglutinin (WGA) staining, dihydroethidium (DHE) staining and immunofluorescence staining with antibodies against collagen III (1:200, Abcam), VLA-4 (1:200, Bioss) and γ-H2AX (a sensitive marker of DNA damage, 1:200, Abcam) as previously described (Wang et al., 2018; Yin et al., 2022).

Total RNA was extracted from heart tissues by Trizol regent (Sangon Biotech, B511311) according to the manufacturer’s instructions. Total RNA (1–2 μg) was used to synthesize the single stranded cDNA by using the PrimeScript RT Kit (Yeasen, 11141ES60). Quantitative polymerase chain reaction (qPCR) was performed with an Applied Biosystems 7,500 system after the cDNA samples were mixed with SYBR reagent (Accurate Biotechnology Co., Ltd., Hunan, AG11701). The specific primer sequences of VCAM-1, ANF, BNP, MYH7, collagen I, collagen III, α-SMA, IL-1β, IL-6, TNF-α, NOX1, NOX2, NOX4, GAPDH and β-actin are listed in Table 1.

Protein was isolated from cardiac tissue or cells by an extraction kit (Keygenbio, KGP250). Protein concentration was calculated by using a BCA Protein Assay Kit (Life-iLab, AP12L025). Samples (30–50 μg) were separated on SDS-polyacrylamide gels and transferred onto PVDF membranes. The membranes were incubated with appropriate antibodies against VCAM-1 (1:200, Proteintech), calcineurin A (1:800, CST), phosphor (p)-STAT3 (1:500, CST), STAT3 (1:500, Arigo), TGF-β1 (1:800, CST), p-Smad2/3 (1:500, CST), Smad2/3 (1:1,000, CST), p-IKKα (1:500, CST), IKKα (1:1,000, CST), p-p65 (1:500, CST), p65 (1:1,000, CST), NOX1 (1:800, CST), NOX4 (1:800, CST) and GAPDH (1:2000, HUABIO). Then, membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (Sino Biological Inc.). Blots were visualized using an ECL chemiluminescence system. GADPH served as the internal reference.

Human serum was collected from the First Affiliated Hospital of Dalian Medical University and Beijing Chao-Yang Hospital of Capital Medical University between August 2021 and January 2022. The criteria of HF diagnosis and treatment were selected according to the 2021 ESC Guidelines (McDonagh et al., 2021), and 30 HF patients and 30 health control were included in the current study. All patients and controls underwent routine physical examinations, echocardiography, urine and blood routine tests. The functional examinations of the lever, lung, kidney and lung were conducted. HF patients with a reduced LV ejection fraction (LVEF <40%), increased BNP concentration (BNP >35 pg/ml) as well as clinical symptoms and signs of HF within 6 months are included. The patients with unstable angina or myocardial infarction during the last 6 months, autoimmune or infectious diseases, cancer, lung disease, and renal failure were excluded. The control subjects with normal cardiac function and without any cardiovascular risk factors, cardiac diseases or any family history of coronary artery disease. as previously described (Wang et al., 2018). The experimental process was approved by the Ethics Committees of First Affiliated Hospital of Dalian Medical University (No. LCKY 2016-31) and Beijing Chao-Yang Hospital of Capital Medical University (2022-human-244). VCAM-1 concentration was measured by ELISA according to the manufacturer’s instructions (Elabscience).

Mouse bone marrow macrophages (BMMs) was isolated and cultured in medium containing 10% fetal bovine serum (Life-iLab, AC03L055) as previously described (Yin et al., 2022). Briefly, the tibia and femur from WT mice were isolated, the whole BM cells were flushed out and cultured overnight. The suspended blood cells were removed and the adherent BM cells were stimulated by recombinant murine colony stimulating factor (M-CSF, PeproTeCH) to induce macrophage maturation. The VLA-4⁺ macrophages were confirmed by immunostaining with anti-VLA-4 antibody. Human umbilical vein endothelial cells (HUVECs) were cultured in EC medium and pretreated with anti-VCAM-1 (5 μg/ml) or IgG control for 4 h and then stimulated with Ang II (100 nM) or saline for an additional 24 h as describe previously (Yin et al., 2022). For the adhesion assay, BMMs were fluorescently labeled with PKH26 (Red, Sigma‒Aldrich) and added to the HUVECs that had been treated as described above. For the migration experiment, BMMs were seeded in the upper chamber of a Transwell insert, which was inserted into plates containing HUVECs treated as described above. DAPI staining was performed to identify migrated BMMs. Images were obtained by microscopy (Olympus, IX73, 100 x), and six randomly selected fields were analyzed.

Neonatal rat Primary cardiomyocytes (NRCMs) and cardiac fibroblasts (CFs) were obtained enzymatically from 1-day-old Sprague-Dawley rat hearts as previously described (Wang et al., 2018). Briefly, hearts were digested into single-cell suspensions and cultured in DMEM/F-12 medium for 90 min. The adherent fibroblasts were cultured in DMEM medium (VivaCell, Shanghai, China, C3113-0500), and the suspended cardiomyocytes was harvested and cultured in new DMEM/F-12 medium, which containing 5-Brdu (100 μm) to eliminate residual CFs. A coculture system was constructed with a transwell plate as previously described (Wang et al., 2018). BMMs and HUVECs were co-cultured and treated with anti-VCAM-1 (5 μg/ml) and Ang II (100 nM) as described in Macrophage Adhesion and Migration. The supernatants were collected and added to CMs or CFs for 24 h. Cardiomyocyte size was determined using immunostaining with anti-α-actinin antibody. The protein levels of CaNA, STAT3, TGF-β1 and p-Smad2/3 in lysates from CMs or CFs were evaluated by immunoblotting analysis. The mRNA levels of collagen I and collagen III in CFs were assessed by qPCR analysis.

All data are expressed as the mean ± standard deviation (M ± SD) and statistically analyzed using SPSS 19.0. Data distribution and heteroscedasticity were analyzed by the Shapiro–Wilk test. Independent t test, one-way ANOVA, or chi-square test were utilized to compare differences in data. P < 0.05 was considered statistically significant.

To verify VCAM-1 participation in Ang II-induced hypertensive cardiac remodeling and clinical HF, we first measured the serum VCAM-1 level in HF patients. The baseline characteristics of humans are shown in Table 2. Serum VCAM-1 and BNP levels were significantly upregulated in HF patients compared with normal controls (Figures 1A,B). Moreover, we measured VCAM-1 level in mouse heart after Ang II treatment. qPCR indicated that VCAM-1 mRNA level was markedly increased in Ang II-infused mouse hearts (Figure 1C). The increase in the expression of VCAM-1 in the hearts of Ang II-infused mice was verified by both immunoblotting and immunohistochemical analysis (Figure 1D,E). Therefore, these results show that VCAM-1 upregulation may be associated with the development of Ang II-induced cardiac remodeling.

To determine the role of VCAM-1 in cardiac remodeling, WT mice were treated with a VCAM-1 neutralizing antibody (0.1 or 0.2 mg/mouse) or IgG every 2 days and infused with Ang II (1,000 ng/kg/min) for 14 days (Figure 2A). We found that compared with saline, Ang II observably elevated the mouse systolic blood pressure (SBP), whereas anti-VCAM-1 significantly inhibited this increase in a dose-dependent manner (Figure 2B). The echocardiography data showed that Ang II-induced cardiac dysfunction, as reflected by increased EF% and FS%, was dose-dependently alleviated by anti-VCAM-1 (Figure 2C; Table 3). Next, we analyzed the inhibited effect of VCAM-1 inhibition on cardiac hypertrophy. H&E and WGA staining indicated that Ang II promoted cardiac hypertrophy, as indicated by increases in heart size, the heart weight to body weight (HW/BW) ratio, the cross-sectional area of myocytes, and the mRNA levels of ANF, BNP and MYH7, while these effects were dose-dependently inhibited by anti-VCAM-1 (Figures 2D–F). Accordingly, we examined the hypertrophy-related signaling molecules in mouse heart. Immunoblotting showed that compared with IgG, anti-VCAM-1 treatment dose-dependently blocked Ang II-induced increase in calcineurin A (CaNA) and phosphorylated (p)-STAT3 protein levels (Figure 2G). Together, these data show that VCAM-1 overexpression accelerates pathological cardiac hypertrophy.

Cardiac fibrosis often leads to a hallmark of a majority of cardiac diseases. Therefore, we first evaluated the degree of cardiac collagen deposition. Compared with that in the saline group, the fibrotic area in Ang II-infused heart was significantly increase, whereas anti-VCAM-1 antibody dose-dependently alleviated this increase (Figure 3A). Moreover, staining with an anti-collagen III or anti-α-SMA antibody revealed that administration of the anti-VCAM-1 markedly inhibited the Ang II-induced upregulation of collagen III and α-SMA (a marker of myofibroblast activation) in a dose-dependent manner (Figures 3B,C). Similarly, the mRNA expression of collagen I, collagen III and α-SMA were dose-dependently downregulated in anti-VCAM-1 antibody-treated heart compared with IgG-treated heart (Figure 3D). Accordingly, Ang II-induced activation of profibrotic TGF-β1 and p-Smad2/3 in IgG-treated mice was dose-dependently attenuated by anti-VCAM-1 (Figure 3E). Overall, our results indicate that VCAM-1 overexpression promotes cardiac fibrosis.

Inflammatory cell infiltration is associated with pathological cardiac remodeling. We next evaluated the role of VCAM-1 in macrophage accumulation and inflammatory reactions. H&E staining revealed that compared with saline control, Ang II infusion promoted cardiac inflammatory cell infiltration, and this effect was dose-dependently inhibited by anti-VCAM-1 treatment (Figure 4A). Then, we identified the infiltrated inflammatory cells using immunohistochemical and immunofluorescence staining with an anti-CD68 and anti-VLA-4 antibody. The results revealed that compared with saline treatment, Ang II infusion observably increased cardiac infiltration of CD68⁺ and VLA-4⁺ macrophages, and this effect was dose-dependently attenuated in anti-VCAM-1-treated group (Figures 4B,C). Accordingly, the mRNA levels of inflammatory cytokines IL-1β, IL-6 and TNF-α in anti-VCAM-1-treated animals were significantly lower than in IgG-treated animals after Ang II infusion (Figure 4D). In addition, we assessed the activation of proinflammatory pathways and found that the Ang II infusion-induced increases in the protein levels of VLA-4, p-IKKα and p-p65 were greatly attenuated in anti-VCAM-1-treated mice (Figure 4E). These findings indicate that VCAM-1 promotes cardiac inflammatory by increasing the infiltration of VLA-4⁺ macrophages into the heart.

Oxidative stress plays a significant role in cardiac remodeling. Therefore, DHE staining was performed to measure ROS levels in heart and found that Ang II-induced augment in ROS level, which was evidenced by an increase in the DHE fluorescence intensity, was markedly dose-dependently suppressed in anti-VCAM-1-treated animals compared with IgG-treated animals (Figure 5A). Moreover, immunostaining for γ-H2AX revealed that Ang II increased γ-H2AX level in nuclei in IgG-treated animals, and this effect was dose-dependently abolished in anti-VCAM-1-treated animals (Figure 5B). Furthermore, we performed immunohistochemical staining to measure the level of nitrotyrosine (an important indicator of the oxidative stress level). As expected, the nitrotyrosine-positive area as was dose-dependently decreased in anti-VCAM-1-treated animals compared with IgG-treated animals after Ang II treatment (Figure 5C). Consistently, the Ang II-caused increase of NADPH oxidase isoforms (NOX1, NOX2 or NOX4), the major source of ROS, was also dose-dependently abrogated by anti-VCAM-1 treatment (Figures 5D,E). These results suggest that increased VCAM-1 expression aggravates oxidative, stress leading to DNA damage.

VCAM-1 has significant effects in mediating leukocytes infiltration into heart tissue (Takamatsu, 2018). Firstly, we performed ELISA to evaluate the concentration of the IL-1β, IL-6 and TNF-α in cultured VLA4⁺ macrophages after saline or Ang II stimulation, and the results showed that Ang II markedly upregulated the levels of these cytokines (Figure 6A). Then, we performed a transwell assay to test whether blocking VCAM-1 can affect BMMs adhesion and migration to HUVECs. The results showed that Ang II markedly upregulated the number of PKH-26-labeled adhered macrophages and migrated macrophages in the IgG-treated group, but this increase was dose-dependently suppressed by anti-VCAM-1 treatment (Figures 6B,C). Furthermore, we tested the effect of BMMs on neonatal rat cardiomyocytes (CMs) or cardiac fibroblasts (CFs). Our results indicated that pretreatment of BMMs with the anti-VCAM-1 obviously inhibited Ang II-caused cardiomyocyte hypertrophy, as indicated by the decrease in CM size, the mRNA levels of ANF and BNP, and the protein levels of CaNA and p-STAT3 compared with IgG-treated group (Figures 6D–F). Similarly, pretreatment of BMMs with the anti-VCAM-1 also markedly reduced Ang II-stimulated differentiation of CFs, as showed by the reduction in the mRNA levels of collagen I and collagen III, as well as the protein levels of TGF-β1 and p-Smad2/3 compared with IgG-treated group (Figures 6G,H).