Group A Streptococcus (GAS, American Type Culture Collection) was grown in brain-heart culture medium (024053, Guangdong Huankai Microbial Technology Co., Ltd., Guangdong, China) at a steady temperature of 37 °C. After 24 hours, the GAS bacteria were rinsed with normal saline (NS) and then inactivated by immersion in 4% paraformaldehyde (P1110, Beijing Solarbio Science & Technology Co., Ltd., Beijing, China). After inactivation, the GAS bacteria were rinsed and placed in NS at a concentration of 4.0 $\times$ 10¹¹ CFU/mL [36]. The GAS bacteria were emulsified at the time of administration via a sonicator (VCX130, Sonics & Materials, Inc, Newtown, CT, USA).

Twenty-four healthy female Lewis rats, aged 10 weeks, were obtained from Beijing Viton Lihua Laboratory Animal Technology Co., Ltd. (Beijing, China). The rats weighed between 160 and 180 g. The rats were kept in a laboratory at Guangxi Medical University’s Animal Experiment Center, which is a specific pathogen-free (SPF) room. The temperature was maintained at 23 $\pm$ 2 °C, with a 12–hour cycle alternating between light and darkness. Prior to the start of the experiment, the rats were allowed to acclimate to the novel environment for one week. The rats were granted unrestricted mobility within their enclosures and were supplied with adequate drinking water and standardized rat chow. The research followed ethical standards for the care and use of laboratory animals. This study was approved by the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University (Approval Number: 202209145).

The experimental rats were pretreated with a recombinant adeno-associated virus (AAV) vector (AAV-siACVR2A; Hanheng Biotechnology (Shanghai) Co., Ltd., Shanghai, China) carrying the ACVR2A-siRNA sequence. To analyze the possible effects of the AAV vector on the rats, a negative control group (AAV-control group) was also established in which the rats were pretreated with AAV without the target sequence. The siRNA sequence of the ACVR2A target gene was 5^′-GGAAGTTGTTGTGCATAAA-3^′. The ACVR2A target gene shRNA sequences were as follows: top strand, 5^′-AATTCGGAAGTTGTTGTGCATAAACTCGAGTTTATG CACAACAACTTCCTTTTTTG-3^′; and bottom strand, 5^′-GATCCAAAAGGAAGTTGTTGTGCATAAACTCGAGT TTATGCACAACAACTTCCG-3^′.

The rats were arbitrarily grouped into four groups (6 rats/group): the control group, RHD group, RHD+AAV-control group (hereinafter referred to as the AAV-control group) and RHD+AAV-siACVR2A group (hereinafter referred to as the AAV-siACVR2A group). The RHD experimental model was constructed over 8 weeks via an approach described in prior investigations [37, 38]. First, a combination of GAS (4.0 $\times$ 10¹¹ CFU/mL) and complete Freund’s adjuvant (F5881, Sigma‒Aldrich, Darmstadt, Germany) was made by mixing them at a 1:1 ratio. Then, 100 µL of this mixture was injected into the hind paw pad of each rat under gas anesthesia at a 2% isoflurane (R510-22-10, RWD Life Science Co., Ltd., Shenzhen, China) concentration. After waiting for one week, 500 µL of the above mixture was administered into the abdomen of the rats, and this mixture was maintained for 4 weeks once a week. During the remaining weeks, 4.0 $\times$ 10¹¹ CFU/mL GAS was injected subcutaneously into the abdominal cavity of the rats at a dose of 500 µL once a week. Before the experiment, rats in the AAV-control and AAV-siACVR2A groups were injected individually with a single dose of 2.5 $\times$ 10¹¹ viral particles diluted with 200 µL of NS via the tail vein, and after 3 weeks, these rats underwent the same injection procedure for modeling. The control rats were subjected to the same protocol as the model rats were, but the injection medium used was NS. The euthanasia endpoint was defined as a reduction of more than 15% of the animal’s weight, together with a decrease in feeding and drinking. In this study, no rat reached this endpoint. At the end of all dosing periods, the rats were euthanized via the intraperitoneal administration of pentobarbital sodium (P3761, Sigma‒Aldrich, Darmstadt, Germany) at a concentration of 150 mg/kg to collect the heart valves.

The mitral valve specimens from dead rat hearts were collected in a timely manner, rapidly transferred to liquid nitrogen, and then stored at –80 °C for additional experimental analysis.

After incubation in 4% paraformaldehyde for 12 hours, the tissue blocks were dehydrated through a range of ethanol concentrations. Afterward, the tissue blocks were promptly transferred into molten paraffin to maintain the temperature and subsequently embedded once the paraffin fully enveloped the tissue blocks. Each block was sequentially sliced into 5 µm thick sections for hematoxylin and eosin (H&E) staining and Sirius red staining. The H&E staining process involved immersing the samples in a hematoxylin solution at ambient temperature for 4–10 minutes, followed by 2–3 minutes of immersion in an eosin staining solution containing alcohol. H&E staining kits (G1005) were purchased from Wuhan Servicebio Technology Co., Ltd (Wuhan, China). A BX43 light microscope (BX43, Olympus Corporation, Tokyo, Japan) was used to record the results. The Sirius red staining process involved the use of Sirius red staining solution (MM1036, Shanghai Maokang Biotechnology Co., Ltd., Shanghai, China) at ambient temperature for one hour. Polarized light pictures were then captured with a BX43 confocal microscope (BX43, Olympus Corporation, Tokyo, Japan).

Immunohistochemical analysis was conducted via the use of purified polyclonal antibodies against rat CD31, $\alpha$-SMA, collagen type I alpha 1 (COL1A1) and vimentin. Paraffin sections (5 µm thick) were treated with a 3% aqueous solution of hydrogen peroxide at ambient temperature and maintained in darkness for 25 minutes. The sections were subsequently blocked with 3% bovine serum albumin (GC305006, Servicebio) under ambient conditions for 30 minutes. Subsequently, the samples were incubated with primary antibodies for 12 hours. The samples were then incubated with a horseradish peroxidase (HRP)-conjugated anti-rabbit secondary antibody (1:200, G1213, Servicebio) for 50 minutes at room temperature. The results were recorded with a BX43 microscope.

HUVECs (Shanghai Zhong Qiao Xin Zhou Biotechnology Co., Ltd., Shanghai, China) were grown in endothelial cell medium (ECM, 1001, ScienCell Research Laboratories, Carlsbad, CA, USA) at 37 °C and 5% CO₂. HUVECs were stimulated with 10 ng/mL recombinant human TGF-$\beta$1 (HZ-1011, Proteintech Group, Inc., Wuhan, China) or 10 ng/mL recombinant human activin A (APA001Hu01, Cloud-Clone Corp, Wuhan, China) for 48 hours. TGF-$\beta$1 at a concentration of 10 ng/mL successfully induces EndMT in HUVECs, as demonstrated in previous studies [39, 40, 41]. Therefore, TGF-$\beta$1 was used as a positive control. To investigate the role of activin A in endothelial cell mesenchymal transition, we first transfected endothelial cells with ACVR2A-siRNA and subsequently treated endothelial cells with activin A. All cell lines were validated by short tandem repeat profiling and tested negative for mycoplasma.

Hanheng Biotechnology (Shanghai) Co., Ltd. (Shanghai, China). provided the siRNAs. HUVECs were transfected with siRNAs via Lipofectamine 3000 (L3000015, Thermo Fisher Scientific, Inc., Waltham, MA, USA). Two microliters of siRNA and 5 µL of Lipofectamine 3000 transfection reagent were added to 250 µL of Opti-MEM (31985070, Thermo Fisher Scientific) solution, mixed separately, and left to stand to form the siRNA-Lipofectamine 3000 complex. The complex was used to treat HUVECs. The ACVR2A-siRNA used were as follows: forward, 5^′-GGAAGUUGUUGUGCAUAAATT-3^′ and reverse, 5^′-UUUAUGCACAACAACUUCCTT-3^′. The negative control (NC)-siRNA used were as follows: forward, 5^′-UUCUCCGAACGUGUCACGUTT-3^′ and reverse, 5^′-ACGUGACACGUUCGGAGAATT-3^′.

TRIzol® reagent (15596026CN, Thermo Fisher Scientific, Inc., Waltham, MA, USA) was used to extract total RNA from heart valves or cultured HUVECs. The RNA was converted into cDNA via a reagent kit (YFXM0009, YIFEIXUE Biotech. Co., Ltd., Nanjing, China). RT-qPCR was conducted with 2$\times$ SYBR Green Fast qPCR Master Mix (YFXM0001, YIFEIXUE Biotech. Co., Ltd., Nanjing, China) on a real-time PCR system (Bio-Rad). The results were normalized to the relative expression of each target gene via the 2^{-Δ⁢Δ⁢Ct} method, and either glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or $\beta$-actin was used as the reference gene. Table 1 summarized the primer sequences used in this study.

PMSF (P0100, Solarbio Science & Technology Co., Ltd., Beijing, China), protein phosphatase inhibitor (P1260, Solarbio), and RIPA buffer (R0010, Solarbio) were made into a cell lysate with a ratio of 1:1:100. Valve tissues or cells were lysed for a period of 1 hour. Protein concentration was determined by BCA assay and individual samples contained 30 mg of protein. Then electrophoresis on 7.5–12.5% sodium dodecyl sulfate (SDS)/polyacrylamide gel electrophoresis (PAGE) gels, followed by transfer to polyvinylidene fluoride (PVDF) membranes (ISEQ00005, EMD Millipore, Darmstadt, Germany). The membranes were blocked with Protein-Free Rapid Sealing Solution (PS108P, Shanghai Epizyme Biomedical Technology Co., Ltd., Shanghai, China) for 30 min at room temperature, and followed by the following specific primary antibodies incubated overnight at 4 °C: anti-GAPDH (1:10,000, 10494-1-AP, Proteintech), anti-Activin A (1:500, ab89307, Abcam, Cambridge, UK), anti-ACVR2A (1:1000, ab134082, Abcam), anti-p-Smad2 (1:500, AF3449, Affinity Biosciences, Jiangsu, China), anti-p-Smad3 (1:500, AF3362, Affinity), anti-Smad2 (1:2000, 12570-1-AP, Proteintech), anti-Smad3 (1:500, AF6362, Affinity, or 1:500, PAC123Hu01, Cloud-Clone Corp), anti-LEF1 (1:1000, ab137872, Abcam), anti-Snail1 (1:500, 13099-1-AP, Proteintech), anti-TWIST (1:1000, 25465-1-AP, Proteintech), anti-ZEB1 (1:500, 21544-1-AP, Proteintech), anti-$\alpha$-SMA (1:1000, ab137872, Abcam), anti-vimentin (1:2000, 10366-1-AP, Proteintech), and anti-VE-cadherin (1:1000, ab231227, Abcam, or 1:1000, ab33168, Abcam). The membranes were washed and then incubated with an HRP-coupled secondary antibody (1:10,000, ab97051, Abcam) for 1 h. An enhanced chemiluminescence (ECL) chemiluminescence kit (P2200, Suzhou New Saimei Biotechnology Co., Ltd., Suzhou, China) was used to visualize the final bands. The grayscale values of the bands corresponding to the final target protein were quantified via ImageJ software 1.53t (National Institutes of Health, Bethesda, MD, USA).

HUVECs were added to a 6-well plate. Once the HUVECs had adhered and proliferated, a pipette tip was used to gently make a wound in the central area of the cell layer. PBS was added for rinsing, and then, serum-free medium was added for culture. Images of the scratches were acquired with a phase contrast microscope at identical positions for 0 hours and 24 hours. The cell migratory capacity was assessed on the basis of the extent of healing observed in the scratched area.

HUVECs were transfected with ACVR2A-siRNA and treated with activin A. After they reached a sufficient treatment time, they were digested with trypsin. Then, 200 µL of HUVECs were combined with serum-free medium and transferred to the upper chamber of a 24-well dish with an 8 µm well. Additionally, 600 µL of HUVECs mixed with 10% FBS medium was transferred to the bottom chamber. After a 12-hour incubation period, HUVECs were stabilized with 4% paraformaldehyde for 15 minutes and then stained with crystal violet solution (G1064, Solarbio) for 20 minutes. Images of HUVECs located on the underside of the luminal membrane were subsequently captured and quantitatively analyzed via an inverted microscope.

GraphPad Prism 9.5 (GraphPad Software, San Diego, CA, USA) was used for statistical analysis of the data. The data are shown as the means $\pm$ standard errors of the means. For comparisons of normally distributed data between two groups, an unpaired Student’s t-test was used, whereas for pairwise comparisons among three or more groups, one-way ANOVA with Tukey’s multiple comparisons test was used. In cases where the data were not normally distributed, a nonparametric test was employed. The data were considered statistically significant if the p-value was less than 0.05.

H&E staining revealed increased inflammatory infiltration in the mitral valvular tissues of rats in the RHD group. However, this inflammatory infiltration was reduced after the inhibition of ACVR2A (Fig. 1A). Sirius red staining was used to identify the collagen type, with type I collagen (COL1) appearing as tightly packed yellow and red fibers and type III collagen (COL3) appearing as loosely packed green fibers under a polarized light microscope. In nonfibrotic valves, COL1 is the primary collagen type, while the proportion of COL3 steadily increases as fibrosis progresses. As fibrosis progresses, the proportion of COL3 gradually increases, and an increase in the COL3/COL1 (COL3/1) ratio can be used to detect the onset of valvular fibrosis [42]. Our results revealed that the COL3/1 ratio was elevated in the RHD group. However, after inhibition of ACVR2A, the COL3/1 ratio decreased (Fig. 1B). The above results indicated that inflammatory infiltration and fibrotic changes occurred in the mitral valvular tissues of the rats in the RHD group, whereas the inflammatory infiltration and fibrosis of the valvular tissues were reduced after the inhibition of ACVR2A.

Fig. 1.

AAV-siACVR2A attenuates the inflammatory response and fibrosis in RHD-affected valves. (A) H&E staining showing the severity of inflammatory cell infiltration in the mitral valvular tissues of the rats in each group. Magnification, $\times$400. Scale bar = 50 µm. (B) Sirius red staining showing the degree of fibrosis in the mitral valvular tissues of the rats in each group (COL1 fibers are tightly arranged (yellow and red), and COL3 fibers are loosely arranged (green)) and the COL3/1 ratio. Magnification, $\times$400. Scale bar = 50 µm. ^#p 0.05 compared with the control group. *p 0.05 compared with the AAV-control group. n = 6. AAV, adeno-associated virus; RHD, rheumatic heart disease; H&E, hematoxylin and eosin; COL1, type I collagen; COL3, type III collagen.

The immunohistochemistry results revealed that the expression of $\alpha$-SMA, COL1A1, and vimentin was elevated in the valvular tissues of the RHD group. However, in the AAV-siACVR2A group, the above indices were decreased in the valvular tissues. In addition, CD31 was slightly reduced in the RHD group, but this slight reduction was reversed after the inhibition of ACVR2A (Fig. 2A). RT-qPCR analysis revealed that the mRNA levels of $\alpha$-SMA, COL1A1, and COL3A1 were elevated in the valvular tissues of the RHD group. After pretreatment with AAV-siACVR2A, the levels of these indicators were reduced. In addition, the mRNA level of VE-cadherin, which is a characteristic indicator of endothelial cells, was decreased in the RHD group but recovered after the inhibition of ACVR2A (Fig. 2B). Western blotting analysis revealed that the protein expression of $\alpha$-SMA and vimentin was increased in valvular tissues in the RHD group, whereas their expression was decreased in the AAV-siACVR2A group. In addition, the expression of VE-cadherin was decreased in the RHD group but recovered after the inhibition of ACVR2A (Fig. 2C). These results suggested that EndMT occurred in the valvular tissues of RHD rats and that inhibition of ACVR2A suppressed EndMT.

Fig. 2.

AAV-siACVR2A inhibited activin-related pathway-mediated endothelial‒mesenchymal transition (EndMT) in RHD-affected valves. (A) Immunohistochemical staining for CD31, $\alpha$-SMA, COL1A1 and vimentin. Magnification, $\times$400. Scale bar = 50 µm. (B) RT-qPCR was used to measure the mRNA expression levels of $\alpha$-SMA, COL1A1, COL3A1 and vascular endothelial (VE)-cadherin in rat valves. (C) Western blot analysis of the protein expression of VE-cadherin, vimentin and $\alpha$-SMA in rat valves. (D) RT-qPCR analysis of the mRNA expression levels of factors related to the activin pathway in rat valves. (E) Western blot analysis of the protein expression of factors related to the activin pathway in rat valves. ^#p 0.05 compared with the control group. *p 0.05 compared with the AAV-control group. n = 6.

RT-qPCR analysis revealed that the mRNA levels of ACVR2A, LEF-1, Snail1, TWIST, ZEB1, and ZEB2 were significantly elevated in the RHD group, but these metrics decreased after the inhibition of ACVR2A. The mRNA expression of Smad2 and Smad3 did not significantly differ between the groups (Fig. 2D). Western blotting analysis revealed that the protein expression of activin A, ACVR2A, p-Smad2, p-Smad3, Snail1, TWIST and ZEB1 was elevated in the RHD group, whereas these indices were reduced in the AAV-siACVR2A group. The protein expression of Smad2 and Smad3 did not differ significantly among the groups (Fig. 2E). The above results indicate that during EndMT in RHD valvular tissues, activin is activated, and the expression of ACVR2A is subsequently elevated, which leads to the phosphorylation of Smad2/3 and then the activation of the intranuclear transcription factors involved in EndMT. In contrast, after the inhibition of ACVR2A, activin did not effectively phosphorylate Smad2/3, and related intranuclear transcription factors were not effectively activated, thus inhibiting EndMT in the valvular tissues of RHD rats.

RT-qPCR analysis revealed that the mRNA expression of ACVR2A, LEF-1, Snail1, TWIST, ZEB1, ZEB2, $\alpha$-SMA, and vimentin was greater in the TGF-$\beta$1 group and activin A group than in the control group, whereas the expression of VE-cadherin was lower in both groups. Smad2 and Smad3 mRNA expression did not differ significantly (Fig. 3A). Western blotting analysis revealed elevated protein expression of activin A, ACVR2A, p-Smad2, p-Smad3, LEF-1, Snail1, TWIST, ZEB1, $\alpha$-SMA, and vimentin in the TGF-$\beta$1 group and activin A group, whereas the expression of VE-cadherin was reduced in both groups. There was no significant difference in the protein expression of Smad2 and Smad3 (Fig. 3B). These results suggested that activation of activin A stimulated the mesenchymal transformation of HUVECs and that activin pathway-related factors were also activated during this process. The mesenchymal transformation of endothelial cells is closely related to the activation of activin and its signaling factors.

Fig. 3.

Activin A promoted EndMT in human umbilical vein endothelial cells (HUVECs). (A) RT-qPCR was performed to measure the mRNA expression levels of activin pathway-related factors and EndMT-related factors in HUVECs during EndMT. (B) Western blot analysis of the protein expression of activin pathway-related factors and EndMT-related factors in HUVECs during EndMT. *p 0.05 compared with the control group. n = 3. TGF, transforming growth factor.

RT-qPCR analysis revealed that the mRNA levels of ACVR2A, LEF-1, Snail1, TWIST, ZEB1, ZEB2, $\alpha$-SMA, and vimentin were decreased in the ACVR2A-siRNA+activin A group and that the mRNA level of VE-cadherin was elevated in the ACVR2A-siRNA+activin A group compared with those in the NC-siRNA+activin A group. There was no significant difference in the mRNA expression of Smad2 and Smad3 (Fig. 4A). Western blotting analysis revealed that the protein expression of activin A, ACVR2A, p-Smad2, p-Smad3, Snail1, TWIST, ZEB1, $\alpha$-SMA, and vimentin was reduced in the ACVR2A-siRNA+activin A group and that the protein expression of VE-cadherin tended to increase. There was no significant difference in the protein expression of Smad2 and Smad3 (Fig. 4B). In addition, the results of the scratch and Transwell migration assays revealed that migration was slower in the ACVR2A-siRNA+activin A group than in the NC-siRNA+activin A group (Fig. 4C,D). The above results indicated that after the inhibition of ACVR2A, the activation of activin and its associated pathway factors was suppressed, and the mesenchymal transition of HUVECs was also inhibited. In addition, the motility of HUVECs during mesenchymal transformation was suppressed.

Fig. 4.

ACVR2A-siRNA inhibited EndMT in HUVECs. (A) RT-qPCR analysis of the mRNA expression levels of activin pathway-related factors and EndMT-related factors in HUVECs after ACVR2A-siRNA treatment. (B) Western blot analysis of the protein expression of activin pathway-related factors and EndMT-related factors in HUVECs after ACVR2A-siRNA treatment. (C) Cell scratch assay using HUVECs. Magnification, $\times$40. Scale bar = 100 µm. (D) Transwell migration assay using HUVECs. Magnification, $\times$200. Scale bar = 50 µm. *p 0.05 compared with the NC-siRNA+activin A group. n = 3.