miR-199a-5p inhibits aortic valve calcification by targeting ATF6 and GRP78 in valve interstitial cells

2.1 Samples

During heart valve replacement surgery, the calcific aortic valve leaflets were obtained from 21 patients diagnosed with aortic valve stenosis. Patients under 65 years of age were excluded. The valve leaflets obtained from 14 heart transplantation recipients suffering from dilated cardiomyopathy served as the control group (Table A1). All human valve samples were instantly resected in surgeries and stored in a −80°C freezer.

1. Ethics approval and consent to participate: The present study was approved by the Ethic Committee of Changhai Hospital, Naval Medical University (No. CH20180630). Patients whose valve leaflets were collected provided written informed consent.

2. Patient consent for the use of valve leaflets: All patients provided written informed consent for the scientific use of valve leaflets.

2.2 Isolation and culture of primary VICs

The aortic valve leaflets in the control group collected from the heart transplantation recipients were soaked into phosphate-buffered solution (PBS). The valvular endothelial cells (VECs) on the sample surface were eliminated by 0.2% collagenase II on a shaker for 15 min at 37°C. The tissue was cut into small pieces, and then, digestion with the aforementioned collagenase II was performed for 2 h. Primary VICs were cultured in a mixed medium including Dulbecco’s modified Eagle’s medium, 1% penicillin and streptomycin, and 10% fetal bovine serum. The third to fifth passages of VICs were used for further experiments.

2.3 Differentiation of VICs into osteoblasts

VICs were induced to differentiate into osteoblasts by osteogenesis medium, which is a complete medium containing sodium dihydrogen phosphate dihydrate (2 mmol/L), ascorbic acid (50 μg/mL), and insulin (10⁻⁷ mol/L). The osteogenesis medium was replaced every 2 days.

2.4 Transfection

The downregulation and upregulation of miR-199a-5p in VICs were produced by transfecting with 100 pmol miR-199a-5p agomiR or antagomiR (GenePharma, China) individually by lipofectamine 2000 (Thermo, USA) in accordance with protocol. To knockdown GRP78 or ATF6 in VICs, cells were transfected with GRP78 or ATF6 siRNA (Obio Technology, China). 8 h ahead of transfection, the medium was changed with osteogenic differentiation medium.

2.5 Real-time quantitative polymerase chain reaction (RT-qPCR)

The miRNAs, mRNAs, and total RNA were extracted by TRIzol (Invitrogen, USA) under the guidance of protocol. The PrimeScript RT reagent Kit (TaKaRa, Japan) was applied to the synthesis of cDNA. miRNA First-Strand Synthesis Kit (TaKaRa, Japan) was used for reverse-transcribed microRNAs. TB Green RT-PCR Kit and Light Cycler 480 System were employed for quantitative real-time PCR assay. The amplification protocol was as follows: pre-denaturation at 95°C for 5 min, denaturation at 95°C for 30 s, annealing at 62°C for 30 s, and extension at 72°C for 30 s. The cycle value was set as 40. At last, the expressions of mRNA and microRNA were normalized to the reference gene GAPDH and U6, respectively. Results were analyzed using the 2^−△△Ct method. All primers are summarized in Table 1.

2.6 Alizarin red staining

Alizarin red staining was conducted to the identification of calcium deposition after the osteogenic differentiation procedure in a period of 7 days. In brief, after being washed twice with PBS, VICs were fixed by 75% ethyl alcohol for 15 min, and then, VICs were washed twice again. Finally, the Alizarin red solution (Servicebio, China) stained VICs for 10 min. For the removal of non-specific staining, cells were further washed with 95% ethanol. The positive stain photos were evaluated and taken by the digital microscope. The quantification of the Alizarin red stain was carried out by measuring the absorbance of the supernatant.

2.7 Dual-luciferase reporter assay

The wild type (WT) of GRP78 and ATF6 3′-UTR which contained the predicted binding site of miR-199a-5p was inserted into the psiCHECK-2 vector (Sangon Biotech, China) individually. Cotransfection of HEK293T cells with luciferase reporter plasmid and miR-199a-5p agomiR or agomiR-NC was performed by Lipofectamine 2000 (Thermo, USA). The firefly and renilla luciferase activities were determined by a Dual-luciferase reporter assay system (Promega, USA).

2.8 Western blotting (WB)

The cellular protein was extracted with SDS buffer. Cell lysates were separated by 10% SDS-PAGE and transferred onto polyvinylidene difluoride membranes. After the blocking procedure, the membrane was stained with several primary antibodies followed by corresponding secondary antibodies (Jackson Immuno Research; 1:5,000 dilution). Protein bands on membranes were detected by an ECL kit (Thermo). The relevant primary antibodies applied above were GAPDH (Protein tech; 60004-1-Ig; 1:5,000 dilution), Runx2 (CST; # 12556; 1:500 dilution), GRP78 (Proteintech; 11587-1-AP; 1:800 dilution), ATF6 (Abcam; ab203119; 1:800 dilution), ATF4 (Santa Cruz; sc-390063; 1:50 dilution), and XBP1 (Abcam; ab37152; 1:800 dilution).

2.9 Immunohistochemistry

These aortic valve specimens were made into paraffin slides. After that, deparaffination and hydration were performed before the blocking procedure. Subsequently, these slides were incubated with primary antibodies separately, including ATF4 (Affinity; DF6008; 1:100 dilution), CHOP (Affinity; DF6025; 1:100 dilution), GRP78 (Proteintech; 11587-1-AP; 1:200 dilution), Runx2 (Abcam; ab23981; 1:200 dilution), ATF6 (Abcam; ab203119; 1:200 dilution), and XBP1 (Abcam; ab37152; 1:200 dilution). Nuclei were counterstained with hematoxylin. In the next stage, the positive staining was detected by an UltraSensitive SP IHC Kit and a DAB Kit (Maixin Biotech; KIT-9710; China).

2.10 Immunofluorescence

The fixations of VICs were carried out with 4% paraformaldehyde. Triton X-100 (0.5%) was used for permeabilization. Primary antibodies from Vimentin (Santa Cruz; 1:1,000 dilution), α-SMA (Affinity; 1:1,000 dilution), to CD31 (Abcam; 1:50 dilution) were incubated overnight at 4°C on a shaker. Next, relevant secondary antibodies labeled with Cy3 (Servicebio; 1:500 dilution) were incubated for 1 h at RT. The nuclei were stained with 4′,6-diamidino-2-phenylindole (Servicebio; 1:500 dilution). Fluorescent images were captured using an inverse fluorescent microscopy.

2.11 Statistical analysis

The statistical analyses were carried out by IBM SPSS v. 21.0 and GraphPad Prism software. Mean values and standard deviation were described for continuous variables. Shapiro–Wilk test was used for normality analysis. Student’s t-test and one-way analysis of variance were used for analysis between two groups and multiple comparisons separately. p values <0.05 were considered statistically significant for all tests.

3.1 Downregulation of miR-199a-5p in calcified aortic valve and VICs’ osteogenic induction model

To investigate whether miR-199a-5p was associated with the pathological process of CAVD, we first analyzed the expression level of miR-199a-5p in calcified aortic valves. Our results demonstrated that the miR-199a-5p expression in calcified aortic valves was significantly lower than that of normal valves (Figure 1a). The normal aortic valve is composed of VICs colonizing in the interstitial substance, VECs cover the surface of the valve leaflets and the extracellular matrix. The results of immunofluorescence staining were a combination of vimentin-positive staining and CD31-negative staining, demonstrating that these cells isolated and cultured from the normal aortic valves were pure VICs, rather than compounds containing VECs (Figure 1b). Then, the VICs were stimulated by the osteogenic induction medium containing inorganic phosphate.

Figure 1

Downregulation of miR-199a-5p in calcified aortic valve and valve interstitial cells (VICs) osteogenic induction model. (a) RT-qPCR analysis of miR-199a-5p in the calcific aortic valve. The miR-199a-5p level dramatically decreased among the CAVD samples. (b) The expression of Vimentin and CD31 in VICs. Scale bar, 100 μm. (c) RT-qPCR results of miR-199a-5p in VICs after osteogenic induction. The miR-199a-5p level dramatically decreased on day 1. *p < 0.05, **p < 0.01.

We then used the VICs’ osteogenic induction model to identify the effect of miR-199a-5p in the osteogenic differentiation of VICs. miR-199a-5p expression was dramatically reduced to a low level from day 1 to day 7 (Figure 1c). These results, taken together, enable us to support our hypothesis that miR-199a-5p deficiency plays a prominent role in CAVD progression.

3.2 ERS was upregulated in VICs’ osteogenic induction model and calcified aortic valve

The RT-qPCR results indicated that, on the third day after osteogenic induction, the mRNA levels of ERS-related proteins from GRP78, ATF6 to ATF4 were remarkably increased (Figure 2a–c), suggesting that the activated ERS may be involved in the process of osteogenic differentiation of VICs. The Runx2 and XBP1 were also increased on the third day after osteogenic induction (Figure 2d and e). An identical tendency was observed by WB (Figure 2f). The protein expression levels of GRP78, ATF6, ATF4, XBP1, and Runx2 were upregulated on the third day.

Figure 2

ERS was upregulated in VICs’ osteogenic induction model and calcified aortic valve. (a–e) RT-qPCR analysis of GRP78, ATF6,ATF4, XBP1, and Runx2 of VICs after osteogenic induction. The mRNA level significantly increased on the third day. (f) WB of GRP78, ATF6, ATF4, XBP1, and Runx2 of VICs. The protein level significantly increased on the third day. Representative immunohistochemical staining of ERS-related proteins (g) GRP78, (h) ATF6, (i) ATF4, and (j) XBP1 in the aortic valve. (k) Representative immunohistochemical staining of CHOP. The expression of CHOP was hardly detected in the two groups. (l) Representative immunohistochemical staining of Runx2. The Runx2 level was upregulated in the CAVD sample. Scale bars, 50 μm. *p < 0.05, **p < 0.01, ***p < 0.001.

The previous studies have confirmed that overactive ERS participated in aortic valve calcification. Immunohistochemistry staining showed that compared to those in the control group, the expression of GRP78, ATF6, ATF4, and XBP1 in the CAVD group was significantly upregulated (Figure 2g–j). However, CHOP expression was hardly detected in both the CAVD and control groups (Figure 2k). Additionally, the expression level of Runx2, which was a pivotal osteogenic-related protein, was also significantly upregulated (Figure 2l). These results demonstrated that the activated ERS may be related to the pathological process of CAVD.

3.3 miR-199a-5p suppressed osteoblastic differentiation of VICs

For the verification of the inhibition effect of miR-199a-5p in the osteogenic differentiation of VICs, the miR-199a-5p in VICs was overexpressed and knocked down by transfecting agomiR-199a-5p and antagomiR-199a-5p, separately. The RT-qPCR analyses revealed that the mRNA levels of GRP78 and ATF6 did not change significantly (Figure 3a and b). Moreover, the mRNA levels of Runx2 and ATF4 were decreased with agomiR-199a-5p transfection. Conversely, silencing miR-199a-5p promoted the expression level of Runx2 (Figure 3c and d). Besides, the WB analysis showed that the protein expressions of GRP78, ATF6, ATF4, and Runx2 were inhibited by agomiR-199a-5p transfection. In contrast, they were increased by antagomiR-199a-5p transfection (Figure 3e). Similarly, the mineralization deposition was suppressed by transfecting agomiR-199a-5p while accelerated by transfecting antagomiR-199a-5p as evidenced by Alizarin red staining (Figure 3f). Taken together, miR-199a-5p attenuated osteogenic differentiation of VICs in vitro.

Figure 3

miR-199a-5p attenuates osteoblastic differentiation of VICs in vitro. (a–d) RT-qPCR analysis of GRP78, ATF6, ATF4, and Runx2 of VICs after transfection of agomiR-199a-5p and antagomiR-199a-5p. (e) The protein levels of GRP78, ATF6, ATF4, and Runx2. (f) Representative Alizarin red staining of calcium deposition. Scale bars, 2 mm. *p < 0.05.

3.4 miR-199a-5p directly targeted at ATF6 and GRP78

The potential molecular mechanism of miR-199a-5p in regulating the osteogenic differentiation of VICs was further analyzed. ATF6 and GRP78 were selected for further analysis (Figure 4a). At first, the luciferase reporter plasmid vector, which contained ATF6, GRP78 3′-UTR, and corresponding mutation vector, was constructed. Second, miR-199a-5p overexpression evidently reduced the luciferase activity of WT-ATF6 3′-UTR and WT-GRP78 3′-UTR (Figure 4b and c). However, the vector that contained relevant mutant sequence eliminated this inhibitory effect, indicating that miR-199a-5p specifically targeted ATF6 and GRP78.

Figure 4

miR-199a-5p directly target ATF6 and GRP78. (a) Schematic illustration of the putative binding site (highlighted in red) of miR-199a-5p. The luciferase activity of WT-GRP78-3′-UTR (b) and WT- ATF6-3′-UTR (c) was remarkably suppressed by the transfection of agomiR-199a-5p. (d) Downregulation of GRP78 significantly suppressed the calcium deposition of VICs. (e) Downregulation of ATF6 dramatically alleviated the calcium deposition of VICs. WT, wild type. *p < 0.05, **p < 0.01.

At last, to investigate whether ATF6 and GRP78 were essential for miR-199a-5p to suppress the osteogenic differentiation of VICs, the rescue experiments were performed. The results of rescue experiments presented that the calcium deposition was obviously weakened after si-GRP78 transfection. In contrast, the mineralization deposition was aggravated by antagonmiR-199a-5p transfection. Co-transfection with si-GRP78 eliminated the aggravating mineralization deposition induced by antagomiR-199a-5p (Figure 4d). Similarly, co-transfection with si-ATF6 also alleviated the effect of the aggravating mineralization deposition induced by antagomiR-199a-5p (Figure 4e). To sum up, we confirmed that GRP78 and ATF6 were the functional targets of miR-199a-5p in VICs’ osteogenic differentiation.