Antibody responses in COVID-19 patients

Introduction

Myocardial fibrosis is a substantial contributor to aberrant cardiac remodeling and the root of many pathological alterations connected to pressure overload-induced chronic heart disease^[1]. Heart failure develops due to an abundance of extracellular matrix (ECM) buildup because ventricular compliance is reduced^[2]. The principal effector cells causing fibrosis are cardiac fibroblasts (CFs)^[3]. When CFs are stimulated, they transdifferentiate into myofibroblasts and are better able to secrete the extracellular matrix between myocytes^[4]. To stimulate and facilitate myofibroblast transdifferentiation, various cytokines, chemokines, and growth factors are required for the differentiation of CFs. This causes an excessive buildup of ECM and an imbalance in the collagen ratio, which eventually leads to cardiac fibrosis. These results in excessive accumulation of ECM and imbalance of collagen ratio, ultimately leading to myocardial fibrosis. Unfortunately, there is no effective treatment to halt the progression of cardiac fibrosis.

Over the past few decades, mounting data has established that TGF-β1/SMADs signaling pathway is a classic route leading to fibrosis^[5]. In cardiac pressure overload, TGF-β signaling is activated^[5]. Specifically, active TGF-β1 attaches to its receptor, which acts on the C-terminal serine residue of R-SMADs, stimulating the phosphorylation of SMAD2 and SMAD3, then builds a transcription complex with SMAD4 that transfers into the nucleus for target gene transcription. TGF-β1 is the most critical cytokine that leads to CF activation, triggering TGF-β1 signaling in fibroblasts through SMAD3, resulting in α-smooth muscle actin (α-SMA) transcription and collagen production^[6]. Thus, it is possible to decrease proliferation and differentiation in CFs by decreasing TGF-β1 signaling pathway, attenuating the development of cardiac fibrosis. However, it remains uncertain how to suppress the TGF-β1 signaling pathway.

A transmembrane glycoprotein named BAMBI has 260 amino acids in total, including the intracellular C-terminus, the transmembrane segment, and the N-terminus of the extracellular ligand-binding domain. It is interesting to note that BAMBI resembles the extracellular domain of type Ⅰ receptors like activin and bone morphogenetic protein (BMP), and that even a section of it shares 50% of its structure with that of the TGF-β receptor^[7]. However, BAMBI is intracellularly short and lacks kinase activity. BAMBI is a TGF-β pathway pseudo-receptor as a result. Due to the degree of similarity between the extracellular domains of BAMBI protein and those of TGF-β type Ⅰ receptors, TGF-β type Ⅱ receptors can bind to TGF-β type Ⅰ receptors during the binding process^[8]. Intriguingly, there is a quiescent response to TGF-β signaling because the intracellular region lacks the serine/threonine kinase domain and is unable to phosphorylate the R-SMADs protein in the cytoplasm. Recent research has demonstrated that aberrant BAMBI expression is crucial to the pathophysiology of inflammation, fibrosis, and the growth of cancer. Inhibiting BAMBI expression, for instance, promotes the TGF-β/SMADs signaling pathway and prevents colon cancer spread, according to studies on the disease^[9]. Thus, BAMBI may lessen cardiac fibrosis by preventing the TGF-β signaling pathway and decreasing collagen formation.

MicroRNAs, which are tiny noncoding RNAs with a length of 21-24 nucleotides, limit protein-coding gene production by targeting messenger RNAs (mRNAs)^[10]. Through imperfect complementary base pairing, these molecules link to the 3' untranslated regions (UTR) of the target mRNA, which can cause mRNA degradation and gene silence^[11]. The activation of numerous signaling pathways and the course of disease are both directly correlated with post-transcriptional control mediated by microRNA. For instance, ventricular hypertrophy and heart failure have been associated with miR-133 overexpression^[12]. The most significant microRNA in the miR-17-92 gene cluster (miR-17/18a/19a/19b/20a/92a), miR-19a, has been causally linked to several illnesses, including cancer and fibrosis^[13]. The miRNA-19 promotes lung cancer by decreasing E-cadherin and initiating EMT in H1299 cells^[14]. Furthermore, miRNA-Target analysis revealed that the binding location for miR-19a-3p at the 3' UTR of BAMBI mRNA is highly conserved across species, indicating that miR-19 might be crucial in the development of fibrosis.

In the present research, we aimed to identify the role of miR-19a-3p in TGF-β1-induced cardiac fibroblast activation, which may provide a novel marker for myocardial fibrosis clinical treatment.

Material and methods

Human tissue sampling

Samples of human myocardium from organ donors or cardiac valve replacement surgeries were utilized in this investigation. For this study, written informed agreement was obtained from patients or family members of donors. This study was approved by the Institutional Review Board of the First Affiliated Hospital of Nanjing Medical University and verified by the ethical guidelines of the 1975 Declaration of Helsinki. All the detailed information is contained in Table 1.

Table  1.  General information of patients

Age	Gender	Etiology
Donation hearts
57	Male	Accidental death
47	Female	Accidental death
50	Man	Accidental death
17	Man	Accidental death
48	Female	Accidental death
Patients with hypertrophic cardiomyopathy (HCM)
57	Female	HCM
47	Male	HCM
58	Female	HCM
43	Male	HCM
28	Male	HCM
Demographics and diagnosis of control subjects of heart organ donors and patients with hypertrophy.

 | Show Table

DownLoad: CSV

Animals

Sprague-Dawley rats (1-3 days old) regardless of gender and male 8-week-old C57BL/6J mice were obtained from the Animal Core Facility of Nanjing Medical University (Nanjing, China). The Animal Care and Use Committee of Nanjing Medical University gave its approval to all animal research, which were conducted in accordance with the National Research Council's Guide for the Care and Use of Laboratory Animals (Permit Number: IACUC 1811030).

Transverse aortic constriction (TAC)

Male C57BL/6J mice (8 weeks, 22–25 g) were put under anesthesia with 1.5% isoflurane inhalation. The TAC operation was carried out exactly as reported before^[15]. In a nutshell, the aortic arch was revealed, and a 0.5-mm wire was implanted alongside the transverse aorta. The wire was ligated with the artery by the 5-0 silk and withdrawn immediately, leaving the aortic arch confined to the wire's diameter. A similar approach was used for the sham operation; however, the suture was not tied around the aorta.

Masson's trichrome staining

Heart samples from each group were fixed in 4% polyformaldehyde solution. They were imbedded in paraffin (Cat. #A6330, Millipore, Billerica, MA, USA) after 24 hours and sliced at 5 μm. To detect cardiac collagen deposition, the tissues were counter stained with hematoxylin, ponceaux and toluidine blue (Cat. #1340, Solarbio, Beijing, China), and viewed using brilliant field microscopy (BX51, OLYMPUS, Tokyo, Japan).

Cell culture

Primary NRCFs isolation and treatment

As previously disclosed, newborn rat ventricular cardiac fibroblasts (NRCFs) were obtained from Sprague-Dawley rats that were 1 to 3 days old^[16]. NRCFs were generated in Dulbecco's Modified Eagle Medium (DMEM, Cat. #12800-017, Life technologies corporation, Gaithersburg, MD, USA) with 10% Fetal Bovine Serum (FBS, Cat. #04-001-1A, Biological Industries, Kibbutz Beit Haemek, Israel) after being plated at a density of 10⁶ cells per milliliter. At 80% to 90% cell density, NRCFs were passaged. The second passage of NRCFs was used for formal experiments^[17–18]. The immunofluorescence of the first passage cells was added to supplement.

We utilized miR-19a-3p mimic and inhibitor created and produced by RiboBio (Guangzhou, China) to research the function of miR-19a-3p in cardiac fibroblast activation. NRCFs were placed on Opti-MEM, then transfected with 50 nmol/L mimic or 100 nmol/L inhibitor using Lipofectamine 2000 (Cat. #11668-019, Invitrogen, Carlsbad, CA, USA) once they had reached 70% to 80% confluence. The mixture of transfection was replenished after 6 hours with DMEM supplemented with 10% FBS. After 48 hours of incubation, NRCFs were stimulated with 10 ng/mL TGF-β1(Cat. #100-21C-10UG, Peprotech, Rocky Hill, NJ, USA) in DMEM containing 1% FBS.

HT1080 cell and HEK 293T cell culture

The Chinese Academy of Sciences' Cell Bank was where the HEK 293T and HT1080 cell lines were bought. The authenticity of the cell lines was verified. In a cell incubator with 5% CO₂, Cells were raised in DMEM supplied with 10% FBS. Cell lines in the study have been tested and found free of Mycoplasma.

After HT1080 cells grew to 80% of confluence, plasmids expressing Flag-BAMBI and GFP were transfected into cells with or without miR-19a-3p mimic implementing Lipofectamine 2000. The culture medium with mixture was changed to DMEM supplemented with 10% FBS after 6 hours. 48 hours later, 10 ng/mL TGF-β1 was used to treat HT1080.

Western blotting

RIPA buffer (Cat. #P0013B, Beyotime, Beijing, China) was used to harvest the cells, and a BCA protein assay kit (Cat. #23225, Thermo Fisher Scientific, Waltham, MA, USA) was applied to determine the protein content. Proteins were denatured and then put through western blot. Cell Signaling Technology (CST, Danvers, MA, USA) provided the primary antibodies for p-SMAD2 (S465/467)/SMAD3(E423/425) (Cat. #8828S, 1∶1000), SMAD2/3 (Cat. #8685S, 1∶1000); Abcam (Cambridge, MA, USA) supplied the antibody for BAMBI (Cat. #ab203070, 1∶1000); Sigma-Aldrich (St. Louis, MO, USA) offered the antibody for α-SMA (Cat. #A5228, 1∶1000); and Beyotime (Beijing, China) provided the antibodies for GAPDH (Cat. #AF0006, 1∶1000) and Tubulin (Cat. #AF1216, 1∶1000). Proteintech (Chicago, IL, USA) supplied collagen type Ⅰ alpha 2 polyclonal antibody (Cat. # 14695-1-AP, 1∶1000).

RNA extraction and qRT-PCR

The RNA isolator Total RNA Extraction Reagent (Cat. #R401-01-AA, Vazyme, Nanjing, China) was a tool to extract total RNA from cells. According to the directions provided by Takara's PrimeScriptTM RT reagent Kit with gDNA Eraser (Perfect Real Time) kit (Cat. #RR047A, TaKaRa, Tykyo, Japan), RNA was reverse transcribed into cDNA. As per the Vazyme miRNA 1st Strand cDNA Synthesis Kit's instructions (via stem-loop) (Cat. #MR101-02), microRNAs were reverse transcribed. On the StepOnePlusTM Real-Time PCR System with Tower (Applied Biosystems, USA), specified product amplification along with detection were carried out. To detect the amount of miR-19a-3p, Bulge-Loop hsa-miR-19a-3p Primer Set (Cat. # MQPS0000773-1-200, RiboBio, Guangzhou, China) was utilized. Invitrogen created the primers for real-time PCR, which were utilized to access mRNA expression. For normalizing mRNA template and microRNA template, Hprt and U6, respectively, were utilized as controls. Cycle times (Ct) for each target PCR were compared in order to determine the relative gene expression, as previously mentioned. The primer sequences are in Table 2.

Table  2.  List of primers

Gene	Forward	Reverse
Rat Hprt	GTTGGATACAGGCCAGACTTTGTT	GATTCAACTTGCGCTCATCTTAGGC
Rat Col1a2	TGTCGATGGCTGCTCCAAAA	CCGATGTCCAGAGGTGCAAT
Rat Col3a1	TGGGAAAGGTGAAATGGGTCC	ATTCCTCCCACTCCAGACTTG
Mus Hprt	GTTGGATACAGGCCAGACTTTGTT	GATTCAACTTGCGCTCATCTTAGGC
Mus Col1a2	AATGGTGGCAGCCAGTTTGA	TCCAGGTACGCAATGCTGTT
Mus Col3a1	TGACTGTCCCACGTAAGCAC	GAGGGCCATAGCTGAACTGA
Mus Tgfb1	ACTGGAGTTGTACGGCAGTG	GGGGCTGATCCCGTTGATTT
Mus Acta2	CTTCGTGACTACTGCCGAGC	AGGTGGTTTCGTGGATGCC
Hsa HPRT	CCTGGCGTCGTGATTAGTGA	CGAGCAAGACGTTCAGTCCT
Hsa COL1A2	GCCTAGCAACATGCCAATC	GCAAAGTTCCCACCGAGA
Hsa COL3A1	TGAAGGGCAGGGAACAAC	GAGGGCGAGTAGGAGCAGT
Abbreviations: Hprt, hypoxanthine phosphoribosyltransferase; Col1a2, collagen type Ⅰ alpha 2 chain; Col3a1, collagen type Ⅲ alpha 1 chain; Tgfb1, transforming growth factor beta 1; Acta2, actin alpha 2, smooth muscle.

 | Show Table

DownLoad: CSV

Dual-Luciferase reporter assay

Lipofectamine 2000 was used to co-transfect 50 nmol/L miR-19a-3p mimic with 10 ng of TK plasmid and 50 ng of WT BAMBI mRNA 3'UTR reporter gene or mutant BAMBI mRNA 3'UTR reporter gene, respectively. Following a 24-hour growth period, cells were washed twice with PBS. According to the manufacturer's protocol, Dual-Luciferase® Reporter (DLRTM) Assay System (Cat. #E1910, Promega Corporation, Madison, WI, USA) was used to detect relative luciferase activity.

Immunofluorescence staining

At room temperature for 30 minutes, NRCFs or sections were exposed to 4% buffered paraformaldehyde. The membrane was then blocked with 0.3% TritonX-100 at ambient temperature for 30 minutes, followed by 30 min of treatment with 10% BSA. Prior to being exposed to Alexa FluorTM 594 donkey anti-mouse antibody (Cat. #R37115, Invitrogen, Carlsbad, CA, USA, 1:500) or Alexa FluorTM 488 donkey anti-rabbit antibody (Cat. #R37116, Invitrogen, Carlsbad, CA, USA, 1:500) for one hour at 37 °C, cells were first treated with the primary antibodies at 4 °C overnight. A fluorescent microscope was used to record these results.

Proliferation assay

Using 5-ethynyl-2-deoxyuridine (EdU) incorporation (Cat. #C0078S, Beyotime Biotechnology, Beijing, China), the proliferation of NRCFs was assessed in accordance with the manufacturer’s instructions. By comparing the quantity of EdU-positive cells to the number of Hoechst-stained cells, the proliferation rate of NRCFs was calculated.

Wound scratch assay

In a six-well plate, NRCFs were seeded before being transfected and given treatment. Scratching was done on cell monolayers using 200 μL plastic pipettes. Applying an inverted light microscope, NRCFs which dramatically migrated into the cell-free area 24 hours after injury were tracked. The residual wound rate, calculated as (residual scratch width /original scratch width), was used to measure cell migration.

Statistical analysis

Prism 9 was used for statistics and graphics. The data were presented as means ± standard deviation. All comparisons between 2 groups were performed with two-tailed unpaired Student's t-test in the condition of homogeneity of variance (p >0.1). If not, we used unpaired t-test with Welch's correction. One-way ANOVA analysis, with Tukey's post hoc test, was used for comparing among ≥3 groups. P<0.05 was deemed statistically significant.

Results

The expression of miR-19a-3p is increased in cardiac fibrosis.

Given that cardiac pressure overload might result in left ventricular fibrosis and, ultimately, heart failure, we investigated the profibrotic signal of murine left ventricular exposed to TAC for 4 weeks. Notably, TAC-induced cardiac hypertrophy was observed in WT mice (Fig. 1A). Histologic analysis reflected that heart weight-to-body weight ratio (HW/BW) was increased by TAC (Fig. 1B). Indeed, it was shown in Masson’s trichrome staining that TAC caused both myocardial interstitial and perivascular collagen deposition (Fig. 1C). Consistently, mRNA expression of collagen and fibroblast activation marker Acta2 (Fig. 1D-F), as well as Tgfb1 mRNA expression (Fig. 1G), were elevated in the left ventricular of TAC mice for 4 weeks compared to the sham group. Indeed, after four weeks of TAC surgery, the protein expression of ΤGF-β1 obviously increased (Fig. 1H-I). Besides, the protein expression of BAMBI was slightly upregulated in TAC mice (Fig. 1J-K). Furthermore, fluorescence co-localization showed that BAMBI expression was upregulated in mouse fibroblasts after TAC (Fig. 1L). Surprisingly, miR-19a-3p expression increased dramatically in pressure overload-induced fibrosis (Fig. 1M). Moreover, RT-PCR of heart tissues showed that the expression of miR-19a-3p were slightly upregulated in human hearts with cardiac hypertrophy (Fig. 1N). Collectively, these results suggest that miR-19a-3p is upregulated in cardiac fibrosis.

Figure  1.  MiR-19a-3p is increased in the myocardium of cardiac fibrosis and in TGF-β1-induced NRCFs.

A–J: Eight-week-old male mice were subjected to transverse aortic constriction (TAC) or sham surgery for 4 weeks. Representative gross morphology of hearts (A). Scale bar: 1 mm. n=6. Heart weight-to-body weight ratio (HW/BW) (B). n=6. Representative graphs of Masson's trichrome staining (C). Scale bar: 100 μm. n=6. Real-time PCR was performed to analyze the expression of fibrotic markers, including Col1a2 (D), Col3a1 (E) and Acta2 (F), and Tgfb1 (G) using Hprt as the internal reference. n=6. Western blotting was used to assess the expression of TGF-β1 (H). Grayscale values for TGF-β1 protein were statistically analyzed (I). n=6. Western blotting was used to assess the expression of BAMBI (J). K: Grayscale values for BAMBI were statistically analyzed. n=6. L: Colocalization of vimentin and BAMBI. Scale bar: 40 μm. M: The RNA levels of miR-19a-3p were identified and standardize them to U6. n=6. N: Levels of miR-19a-3p was measured by Real-time PCR in 5 patient hearts with cardiac hypertrophy and 5 normal hearts from donations. n=5. O: 10 ng/mL TGF-β1 was administered to cultured NRCFs over a period of time. Through the application of qRT-PCR, the RNA levels of miR-19a-3p were identified and standardize them to U6. n=3. All data were presented as the mean ± standard deviation. Statistical analyses were performed by student's t-test. ^*P <0.05, ^**P <0.01. Abbreviations: TGF-β1, transforming growth factor β1; BAMBI, bone morphogenetic protein and activin membrane-bound inhibitor; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; DAPI, diamidino-phenyl-indole.

DownLoad: Full-Size Img PowerPoint

TGF-β1 is a known contributor to myocardial fibrosis. To investigate the involvement of miR-19a-3p in cardiac fibroblast activation, NRCFs were exposed to 10 ng/mL TGF-β1 at various times. According to RT-PCR analysis, miR-19a-3p was considerably higher in TGF-β1-treated NRCFs than in control groups (Fig. 1O). These results collectively suggested that TGF-β1 induces the expression of miR-19a-3p in NRCFs.

MiR-19a-3p promotes TGF-β1-induced NRCF differentiation.

Due to the fact that synthetically active fibroblasts express α-SMA^[19], we proposed that overexpressing miR-19a-3p could facilitate TGF-β1-induced NRCF activation. Following TGF-β1 administration, the expression of α-SMA in NRCFs was evaluated though western blot. In fact, miR-19a-3p mimic significantly increased the level of the protein α-SMA in response to TGF-β1 stimulation (Fig. 2A-B), but miR-19a-3p inhibitor transfection reversed the effects of miR-19a-3p (Fig. 2C-D). Consistently, there was an increase of α-SMA positive NRCFs with transfection of miR-19a-3p mimic as evidenced by immunofluorescence (Fig. 2E-F). The immunofluorescence results of the first passage cells were consistent to the second (Supplementary Fig. 1A-B). Nevertheless, inhibition of miR-19a-3p reduced this effect. Therefore, these data support the idea that miR-19a-3p crucially promotes TGF-β1-induced NRCF differentiation.

Figure  2.  MiR-19a-3p is involved in TGF-β1-induced NRCF differentiation.

A–F: For 48 hours, miR-19a-3p mimic or inhibitor were transfected into NRCFs. Transfected cells were starved before being incubated for a further 24 hours with or without 10 ng/mL TGF-β1. A: Western blotting was used to assess the expression of α-SMA following transfection with miR-19a-3p mimic. B: Grayscale values for α-SMA protein were statistically analyzed. n=3. C: Western blotting was used to detect the quantity that α-SMA was expressed after miR-19a-3p was inhibited. D: Grayscale values for α-SMA protein were statistically analyzed. n=3. E: Immunofluorescent expression of α-SMA. 50 microns wide was the scale bar. F: The extent of NRCF activation by a histogram of the proportion of cells that were positive for α-SMA. n=3. All data were presented as the mean ± standard deviation. Statistical analyses were performed by one-way ANOVA followed by Tukey's tests for multiple comparisons. ^*P <0.05, ^**P <0.01, ^***P <0.001. Abbreviations: TGF-β1, transforming growth factor β1; α-SMA, α-smooth muscle actin; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; DAPI, diamidino-phenyl-indole.

DownLoad: Full-Size Img PowerPoint

Inhibition of miR-19a-3p restrains TGF-β1-induced NRCF proliferation, migration, and collagen Ⅰ/ Ⅲ expression.

An assay for the incorporation of EdU was used to gauge the impact of miR-19a-3p on NRCF proliferation. The proliferation of NRCFs was significantly increased after treatment with 10 ng/mL TGF-β1. When compared to the negative control, miR-19a-3p overexpression accelerated TGF-β1-induced NRCF proliferation (Fig. 3A-B). However, miR-19a-3p inhibition showed the opposite trend. Using a wound-healing test, the impact of miR-19a-3p on NRCF migration was examined. As illustrated in Figure 3A, up-regulation of miR-19a-3p resulted in increased cell motility after TGF-β1 therapy. In contrast, miR-19a-3p suppression limited cell migration toward the wound (Fig. 3C–D). These findings indicate that lessened miR-19a-3p inhibits the proliferation and migration of NRCFs in vitro.

Figure  3.  Down-regulation of miR-19a-3p restricts TGF-β1-induced NRCF proliferation, migration, and collagen deposition.

A–H: MiR-19a-3p mimic or inhibitor have been introduced for 48 hours into cultured NRCFs. For an additional 24 hours, cells were stimulated with or without 10 ng/mL TGF-β1. A: The proliferation of NRCFs was investigated employing an EdU incorporation assay. The scale bar had a width of 50 microns. B: The Histogram displayed the individually proliferation rates relative to the control group. n=3. C: Measurement of NRCF migration utilizing a wound closure assay. The scale bar was 100 microns in width. D: The degree of NRCF motility was displayed by a histogram of migration rate. n=3. E–H: Real time-PCR was performed to analyze the expression of fibrotic markers including Col1a2 (E and G) and Col3a1 (F and H) using Hprt as the internal reference. n=3. I: Western blotting was used to assess the expression of collagen Ⅰ following transfection with miR-19a-3p mimic. J: Grayscale values for collagen Ⅰ protein were statistically analyzed. n=3. K: Western blotting was used to assess the expression of collagen Ⅰ following transfection with miR-19a-3p inhibitor. L: Grayscale values for collagen Ⅰ protein were statistically analyzed. n=3. All data were provided as the mean ± standard deviation. Statistical analyses were performed by one-way ANOVA followed by Tukey's tests for multiple comparisons. ^*P <0.05, ^**P <0.01, ^***P <0.001. Abbreviations: TGF-β1, transforming growth factor β1; EdU, 5-Ethynyl-2-deoxyuridine.

DownLoad: Full-Size Img PowerPoint

Based on the findings, we reasoned that miR-19a-3p might be involved in TGF-β1-induced collagen expression. With miR-19a-3p mimic transfection, the relative gene expression levels of the collagen synthesis genes Col1a2 and Col3a1 were significantly increased in TGF-β1-stimulated NRCFs in vitro (Fig. 3E-F). Consistently, miR-19a-3p inhibition reduced Col1a2 and Col3a1 mRNA levels in NRCFs (Fig. 3G-H). Steadily, under TGF-β stimulation, transfection with miR-19a-3p mimic led to a further increase in the expression of collagen Ⅰ, while the inhibitor induced the opposite effect (Fig. 3I-L). Together, these findings imply that miR-19a-3p may play a role in the way that TGF-β1 induces the expression of the collagen gene.

MiR-19a-3p augments TGF-β1/SMAD2/3 signaling by targeting BAMBI.

Since miR-19a-3p was involved in TGF-β1-induced cardiac fibroblast activation, we implemented bioinformatic target-scan analysis (TargetScanHuman 8.0: https://www.targetscan.org/vert_80/) to determine the precise target locations of miR-19a-3p. A highly conserved miR-19a-3p binding site across species was discovered at locations in the BAMBI mRNA 3'UTR (Fig. 4A and Supplementary Fig. 2). Furthermore, dual luciferase reporter assays were utilized to determine if the anticipated binding sites in BAMBI 3'UTR are necessary for miR-19a-3p regulatory function. Strikingly, luciferase activity was decreased when miR-19a-3p mimic was overexpressed in HEK293T cells that had been transfected with the WT BAMBI 3'UTR, which contains miR-19a-3p binding sites (Fig. 4B). Collectively, miR-19a-3p binds to the 3'UTR of BAMBI, as it was removed in HEK 293T cells transfected with mutant BAMBI 3'UTR, in which each predicted miR-19a-3p binding site was altered.

Figure  4.  Fig. 4MiR-19a-3p promotes TGF-β1 signaling by targeting BAMBI.

A: TargetScanHuman8.0 predicted miR-19a-3p binding sites in the BAMBI 3'UTR. B: The pGL3-luciferase reporter vectors carrying either WT or mutant BAMBI 3'UTR were transfected into HEK 293T cells. Further treatment for the cells either included miR-19a-3p mimic or a negative control. 24 hours later, the activity of luciferase was discovered. n=4. C–E: MiR-19a-3p mimic or inhibitor were introduced into NRCFs. miR-19a-3p mimic or its negative control (NC) were introduced into NRCFs. The RNA expression of miR-19-3p was assessed by RT-PCR, using U6 as a reference (C). n=3. Western blotting was performed to measure the BAMBI protein level (D). E: Quantitative analysis of the protein expression levels in the panel D using GAPDH as an internal reference. n=3. F–K: NRCFs were grown in vitro and either miR-19a-3p mimic or its inhibitor were transfected into them. TGF-β1 was administered after 48 hours. The protein levels of p-SMAD2/3, SMAD2/3 and BAMBI were detected by Western blotting (F and I). Phosphorylation of SMAD2/3 and BAMBI expression level were quantitative analyzed (G, H, J and K). n=3. Data are presented as the mean ± standard deviation. Statistical analyses were performed by student's t-test for comparison between the two groups and one-way ANOVA followed by Tukey's tests for multiple comparisons. ^*P <0.05, ^**P <0.01, ^***P <0.001. ns, no significance. Abbreviations: TGF-β1, transforming growth factor β1; BAMBI, bone morphogenetic protein and activin membrane-bound inhibitor.

DownLoad: Full-Size Img PowerPoint

To further explore the role of miR-19-3p in regulating BAMBI, we evaluated BAMBI protein in NRCFs after being transfected with miR-19a-3p mimic or inhibitor. After transfection with miR-19a-3p mimic, miRNA overexpression was validated using qPCR (Fig. 4C). Predictably, BAMBI was dramatically reduced when miR-19a-3p mimic was overexpressed, while increased after inhibition of miR-19a-3p (Fig. 4D-E). Thus, these findings support that miR-19a-3p directly binds to BAMBI 3’UTR to limit its expression.

Both malignant fibrosis and normal tissue healing increase TGF-β1 expression^[20]. TGF-β type Ⅰ and type Ⅱ receptors bind to TGF-β1 once it has been activated. SMAD2/3 are phosphorylated by the TGF-β type Ⅰ receptor, which causes them to attach to SMAD4 and translocate into the nuclei to trigger gene transcription^[21]. To investigate whether miR-19a-3p could regulate the expression of BAMBI and TGF-β1 signaling in cardiac fibroblasts, NRCFs were transfected with miR-19a-3p mimic or negative control respectively. After 48 hours, TGF-β1 was administered, and the phosphorylation level of SMAD2/3 were detected by protein extraction (Fig. 4F). Western blot showed that TGF-β1 stimulation activated SMAD2/3 phosphorylation compared to the negative control. However, overexpression of miR-19a-3p significantly increased phosphorylation of SMAD2/3 (Fig. 4F-G). Specially, miR-19a-3p mimic alleviated the expression of BAMBI induced by TGF-β1 (Fig. 4H). These results suggested that miR-19a-3p promotes the activation of TGF-β1/SMAD2/3 signaling pathway in cardiac fibroblasts.

To further understand the molecular mechanisms underlying the association between miR-19a-3p and TGF-β1 signaling pathway, NRCFs were transfected with miR-19a-3p inhibitor or negative control. We also found that the inhibition of miR-19a-3p decreased SMAD2/3 phosphorylation after the stimulation of TGF-β1 (Fig. 4I-J). In addition, the expression of BAMBI was upregulated by miR-19a-3p inhibitor (Fig. 4K). Together, these results strongly supported that miR-19a-3p activated TGF-β1/SMAD2/3 signaling via suppressing BAMBI expression.

Overexpression of BAMBI ameliorates miR-19a-3p-induced fibroblast collagen expression through TGF-β1 signaling.

To confirm the role of BAMBI in TGF-β1/SMADs cascade, we transfected miR-19a-3p mimic into HT1080 cells, overexpressed BAMBI during that period, and then investigated the phosphorylation of SMAD2/3. As seen in Fig. 5A-B, over-expression of BAMBI dramatically decreased phosphorylation of SMAD2/3 in fibroblasts brought on by miR-19a-3p mimic. These findings provided additional evidence that miR-19a-3p stimulates TGF-β1/SMAD2/3 signaling by inhibiting BAMBI.

Figure  5.  Elevated BAMBI ameliorates miR-19a-3p-induced fibroblast collagen expression through TGF-β1 signaling.

A–E: In addition to miR-19a-3p mimic or NC, pcDNA3.1-BAMBI-3$\times$Flag or pcDNA3.1-GFP were transfected into HT1080 cells. TGF-β1 was given 48 hours later. Western blotting was used to identify the protein levels of p-SMAD2/3, SMAD2/3 and Collagen Ⅰ (A). SMAD2/3 was used as an internal reference in order to quantify the level of p-SMAD2/3 protein expression (B). Grayscale values for collagen Ⅰ protein were statistically analyzed used Tubulin as an internal reference (C). Quantitative RT-PCR was applied to gauge the amounts of fibrotic biomarkers (D and E). n=3. All data were showed as the mean ± standard deviation. Statistical analyses were performed by one-way ANOVA followed by Tukey's tests for multiple comparisons. ^*P <0.05, ^**P <0.01, ^***P <0.001. Abbreviations: TGF-β1, transforming growth factor β1; BAMBI, bone morphogenetic protein and activin membrane-bound inhibitor.

DownLoad: Full-Size Img PowerPoint

The main characteristic of fibroblast activation is increased extracellular matrix, such as collagen expression. To further validate whether BAMBI participates in the miR-19a-3p-mediated collagen gene expression, we examined the expression of collagen. MiR-19a-3p mimic or its negative control, and pcDNA3.1-3$\times$flag-BAMBI were co-transfected into HT1080 cells with or without TGF-β1 treatment. After transfected with miR-19a-3p mimic, the protein expression of collagen Ⅰ was decreased by overexpression of BAMBI in fibroblasts upon TGF-β1 stimulation (Fig. 5A and Fig. 5C), so did the mRNA level of COL1A2 and COL3A1 (Fig. 5D-E). These results demonstrated that BAMBI ameliorates miR-19a-3p-mediated collagen gene expression.

Discussion

In this study, it was demonstrated that miR-19a-3p expression is increased in patients with cardiac hypertrophy and mice following TAC. We further confirmed that TGF-β1 upregulated the expression of miR-19a-3p which inhibited the expression of BAMBI via post-transcriptional regulation in NRCFs, ultimately promoting the activation of TGF-β1/SMAD2/3 and enhancing cardiac fibroblast activation (Fig. 6).

Figure  6.  miR-19a-3p augments TGF-β1-induced cardiac fibroblasts activation via targeting BAMBI.

BAMBI acts as a pseudo receptor of TGF-β1 to block the phosphorylation of SMAD2/3. TGF-β1 increases miR-19a-3p expression, which suppresses BAMBI and encourages the phosphorylation of SMAD2/3 and cardiac fibroblast activation under pressure overload.

DownLoad: Full-Size Img PowerPoint

Left ventricular pathological fibrosis was probably brought on by pressure overload^[22]. The most significant aspect in the pathological phase of myocardial fibrosis is the activation of cardiac fibroblasts. Increased interstitial collagen deposition and myocardial systolic and diastolic dysfunction are brought on by the stress-induced conversion of cardiac fibroblasts into myofibroblasts^[23]. It is well known TGF-β1 is an important component that contributes to increased extracellular matrix. The expression of cardiac TGF-β is elevated when the heart is under pressure overload, as shown by Yufeng Yao et al^[24]. Our findings, which were in line with this research, demonstrated elevated TGF-β1 expression following TAC. TGF-β1/SMAD3 signaling is shown by Rodriguez P. et al. to be important for fibroblast differentiation^[25]. The α-SMA, a key source of ECM proteins in cardiac remodeling, is expressed by myofibroblasts, which differ from fibroblasts physically and functionally. They are a manifestation of the cardiac fibrotic response. Similarly, TGF-β1 signaling has been shown by Pchejetski D. et al. to encourage the formation of extracellular matrix, including collagen, further increasing the onset of ventricular fibrosis^[26]. Our experiments were conducted in vitro using NRCFs. Upon TGF-β1 stimulation, we observed the activation of p-SMAD2/3 and an increase in the expression of collagen.

MicroRNAs have been implicated in fibrotic processes in recent years. For instance, inhibition of miR-221-3p/ miR-222-3p accelerates TGF-β-induced myocardial fibrosis^[27], and several studies have shown that microRNAs modulate the TGF-β signaling pathway in fibrotic processes^[28]. In our study, TGF-β1 activated the expression of miR-19a-3p both in vivo and in vitro. However, Zou M et al. revealed that miR-19a-3p/19b-3p expresses with low levels in the plasma of heart failure patients, and miR-19a-3p inhibits autophagy via TGF-β signaling in a human cardiac fibroblast cell line^[29]. Since miR-19-3p in plasma may be impacted by multiple organs such as the liver and kidney, and miR-19a-3p expression in cardiac tissues has not been observed, interestingly, we detected and noticed miR-19a-3p increasing in both patients with cardiac hypertrophy and mice following. And in vitro, neonatal rat cardiac fibroblasts were employed in our research, TGF-β1 induced the expression of miR-19a-3p. Additionally, Zhang Y et al. demonstrated that TGF-β increases the expression of miR-19, and miR-19 promotes the development of renal fibrosis by inhibiting PTEN^[30]. Given that the TGF-β1/SMADs signaling pathway is involved in cardiac fibrosis^[5], we also observed cardiac fibroblast activation with transfection of miR-19a-3p mimic. The study provides the first evidence that miR-19a-3p might facilitate differentiation, proliferation, and collagen gene expression of cardiac fibroblasts generated by the TGF-β1/SMAD2/3 signaling pathway.

Importantly, through attaching directly to the 3' untranslated regions of their target mRNAs, microRNAs normally inhibit gene expression at the post-transcriptional level^[31]. In our study, microRNA TargetScanHuman8.0 was used to investigate the presence of the miR-19a-3p binding site in the 3' UTR of BAMBI. For this purpose, we further demonstrate the post-transcriptional regulation of miR-19a-3p on BAMBI via luciferase reporter gene assay. Whereas BAMBI is a transmembrane protein that inhibits TGF-β1 signaling by lacking an intracellular kinase domain but having an extracellular domain identical to the TGF-β1 receptor^[32]. Endogenous BAMBI may bind to TβR-II, resulting in a decrease in the TβR-I/TβR-II complex and the quantity of phosphorylated TβR-I in this complex, according to Onichtchouk D et al^[33]. These processes may explain the capability of BAMBI to block TGF-β signaling, although they do not rule out interaction with type Ⅱ receptors^[33]. BAMBI has been shown to suppress β-catenin and TGF-β1 signaling pathways in HCC cells^[34]. Villar AV et al observed that TGF-β1 stimulation of primary mouse fibroblasts promoted an increase in BAMBI expression via positive feedback^[35]. Furthermore, BAMBI deletion worsened the TGF-β1-induced profibrotic response in primary cardiac fibroblasts, but BAMBI overexpression alleviated this response in NIH-3T3 fibroblasts^[35]. Interestingly, transfection of miR-19a-3p mimic in NRCFs inhibited BAMBI expression and promoted the expression of p-SMAD2/3, resulting in further increases in Col1a2 and Col3a1 mRNA levels. Consistently, this effect was reversed upon inhibition of miR-19a-3p. Therefore, we demonstrate that miR-19a-3p participates in the regulation of TGF-β1 signaling pathway by inhibiting the expression of BAMBI.

In summary, the expression of miR-19a-3p in myocardial tissue has been demonstrated to be strongly associated to myocardial fibrosis. Moreover, our study provides evidence that TGF-β1 induces high expression of miR-19a-3p in NRCFs. Importantly, our findings indicate that miR-19a-3p augments TGF-β1-induced differentiation and proliferation of cardiac fibroblasts by targeting BAMBI, which has not been previously demonstrated. And miR-19a-3p may be a feasible new indicator for clinical treatment of myocardial fibrosis.