Nucleolin promotes Ang II‐induced phenotypic transformation of vascular smooth muscle cells by regulating EGF and PDGF‐BB

Abstract RNA‐binding properties of nucleolin play a fundamental role in regulating cell growth and proliferation. We have previously shown that nucleolin plays an important regulatory role in the phenotypic transformation of vascular smooth muscle cells (VSMCs) induced by angiotensin II (Ang II). In the present study, we aimed to investigate the molecular mechanism of nucleolin‐mediated phenotypic transformation of VSMCs induced by Ang II. Epidermal growth factor (EGF) and platelet‐derived growth factor (PDGF) inhibitors were used to observe the effect of Ang II on phenotypic transformation of VSMCs. The regulatory role of nucleolin in the phenotypic transformation of VSMCs was identified by nucleolin gene mutation, gene overexpression and RNA interference technology. Moreover, we elucidated the molecular mechanism underlying the regulatory effect of nucleolin on phenotypic transformation of VSMCs. EGF and PDGF‐BB played an important role in the phenotypic transformation of VSMCs induced by Ang II. Nucleolin exerted a positive regulatory effect on the expression and secretion of EGF and PDGF‐BB. In addition, nucleolin could bind to the 5′ untranslated region (UTR) of EGF and PDGF‐BB mRNA, and such binding up‐regulated the stability and expression of EGF and PDGF‐BB mRNA, promoting Ang II‐induced phenotypic transformation of VSMCs.

division. Although the expression of nucleolin in tumours and other rapidly dividing cells is quite high, its expression remains very low in non-dividing cells. Therefore, nucleolin, as an effective marker, is often used to determine the extent of cell proliferation. 9 Phenotypic transformation of vascular smooth muscle cells (VSMCs) is a critical initial step in the proliferation and migration of VSMCs. Our previous studies have indicated that the expression of nucleolin at the mRNA and protein levels is gradually increased after VSMCs are stimulated with angiotensin II (Ang II) at different concentrations and durations, and Ang II can induce nucleolin translocation from nucleus to cytoplasm. Such findings further confirm that nucleolin plays a positive regulatory role in Ang II-mediated phenotypic transformation of VSMCs. 10 However, how the cellular nucleolin promotes phenotypic transformation of VSMCs, as well as its specific molecular mechanism, remains largely undetermined.
A large number of studies have revealed that the RNA-binding properties of nucleolin play a fundamental role in a variety of biological functions, and the specific nucleic acid-binding element is '(T/G) CCC G (A/G)'. 6,[11][12][13][14][15] Therefore, we first analysed the mRNA sequences of several VSMC phenotype transformation genes by bioinformatics analysis and found that the mRNA sequences of 12 VSMC phenotype-related genes, such as epidermal growth factor (EGF) and platelet-derived growth factor (PDGF), contained different amounts of nucleolin-specific binding element (Table 1). Based on our previous findings, we hypothesized that Ang II-induced up-regulation of nucleolin, and then it bound to the mRNA of EGF and PDGF through the RNA-binding properties of nucleolin, and regulated the stability of EGF and PDGF mRNA, thus positively regulating the phenotypic transformation. In the present study, we used Ang II to induce phenotypic transformation of VSMCs, and the regulatory role of nucleolin in the phenotypic transformation of VSMCs was identified by molecular biology methods, such as nucleolin gene mutation, gene overexpression, RNA interference technology and protein-mRNA interaction analysis (post-transcriptional level). Moreover, we assessed the protein-RNA interaction to elucidate the molecular mechanism underlying the regulatory effect of nucleolin on phenotypic transformation of VSMCs. Collectively, our study provided valuable insights into the regulatory mechanism of VSMC phenotypic transformation and offered a new strategy for controlling and reversing the phenotypic transformation of VSMCs.

| Bioinformatics analysis of nucleolinbinding elements
The mRNA sequences of 12 VSMC phenotype-associated genes, such as epiregulin, were searched against PubMed (ncbi.nlm.nih.gov/ pubme d/) and UCSC Genome Browser gene databases (genome. ucsc.edu/), and the mRNA sequences of these genes were used to screen nucleolin-binding elements, such as '(T/G) CCCG (A/G)' and analyse whether these binding elements were located in the 5′ UTR, 3′ UTR or coding region.
Subsequently, the cell lysates were transferred into the 1.5-mL centrifuge tubes, denatured at 100°C for 10 minutes and then centrifuged at 12 000 g for 10 minutes at 4°C. The supernatant was transferred to a 0.5-mL centrifuge tube and stored at −70°C. The protein concentration was determined using a Bicinchoninic Acid (BCA) Protein Assay kit (Shanghai Biyuntian Biotechnology, Ltd.) according to the manufacturer's instructions.

| Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)
Total RNA was isolated using the Rneasy kit (Qiagen) according to the manufacturer's instructions, and 2 μg purified RNA was reversely transcribed into cDNA using oligo(dT) primers. The expressions of target genes were examined on a 7500 Fast Real-Time PCR system (Applied Biosystems) using a QuantiTect SYBR Green PCR Kit (Qiagen). Briefly, after an initial denaturation step at 95°C for 10 seconds, amplifications were carried out with 40 cycles at a melting temperature of 95°C for 5 seconds and an annealing temperature of 60°C for 30 seconds. Primer sequences used in the present study were listed in Table 1. The specificity of amplification was verified by melting curve analysis and validated by electrophoresis on agarose gels. The relative expressions of target genes were calculated by 2 −ΔΔCt method, and β-actin was selected as the housekeeping gene. Each experiment was conducted in triplicate.

| Amplification and extraction of recombinant plasmids
The recombinant plasmids of pcDNA3.

| Transient transfection
Transfection of cells was carried out following the manufacturer's instructions (MegaTran 1.0; OriGene). Briefly, 5 × 10 5 cells were grown in 5 mL appropriate complete growth medium at 37°C in a CO 2 incubator until the cells reached 70%-80% confluence (24 hours).
After washed with serum-free and antibiotic-free medium, the cells were transfected with pcDNA3.1-Nuc/psiRNA-Nuc (experimental), pcDNA3.1/psiRNA (vector control) or Nuc1-309 by mixing with 6 μL MegaTran 1.0 containing 2 μg DNA, and the mixture was placed at room temperature for about 10 minutes. Subsequently, the mixture was added to a 6-well plate, followed by gentle agitation and incubation at 37°C for 24 hours in a CO 2 incubator.

| Western blotting analysis
Following various treatments, VSMCs cells were lysed with radioimmunoprecipitation assay (RIPA) buffer (Shanghai Biyuntian Biotechnology, Ltd.). The protein concentration was determined using BCA assay. Whole-cell lysates were subjected to SDS-PAGE and then transferred onto polyvinylidene fluoride (PVDF) membranes. The blots were incubated with respective primary antibodies against nucleolin (rabbit polyclonal antibody, Sigma), α-SM-actin (mouse monoclonal antibody, Boster Biotech), SM22a (rabbit polyclonal antibody, Abcam), calponin (mouse monoclonal antibody, Abcam), OPN, EGF, PDGF-BB (rabbit polyclonal antibody, Abcam) and β-actin (mouse monoclonal antibody, Abcam) at 25°C for 2 hours. Subsequently, the blots were incubated with peroxidase-conjugated secondary antibodies at 25°C for 1 hours. Immunoreactive bands were visualized utilizing enhanced chemiluminescence detection kit (Beyotime Institute of Biotechnology) according to the manufacturer's instructions, and the densitometry analysis was performed by scion image software.

| Enzyme-linked immunosorbent assay (ELISA)
Levels of EGF and PDGF-BB in the culture medium were determined by commercially available rat ELISA kits for EGF and PDGF-BB (Abcam) according to the manufacturer's instructions.
Considering that EGF and PDGF-BB are cell secreting proteins, the cells were cultured in a 96-well plate for 48 hours. When the cells reached 80%-90% confluence, the serum-free medium was replaced before adding Ang-II. Cell culture supernatant was col-

| Co-immunoprecipitation of nucleolin protein and EGF, PDGF-BB mRNA
After treatment, VSMCs were homogenized in 1 mL RIPA buffer containing protease inhibitors (10 µL/0.1 g tissue weight; Sigma-Aldrich; Merck KGaA). Soluble proteins were collected after centrifugation at 12 000 g for 15 minutes at 4°C. After quantitative analysis by BCA method, protein supernatant was divided into three equal fractions (500 μL of each) as follows: one was used as 'Input' sample; one was used as control IgG for immunoprecipitation; and the third one was immunized with antibody against nucleolin. An aliquot (500 µL) of cell lysate was pre-cleared by incubation with 200 μL of protein A/G beads on ice for 60 minutes, followed by centrifugation. Subsequently, 10 μg of anti-nucleolin antibody was added to the pre-cleared cell lysate, and the mixture was incubated at 4°C for 1 hours, followed by the addition of 200 μL protein A/G beads. The lysate was incubated at 4°C with agitation, and the immune complexes were separated by centrifugation at 10 000 g for 30 seconds at 4°C. RNA was extracted from the immunoprecipitate, and cDNA was prepared by reverse transcription with random primers and subjected to PCR.

| Measurement of mRNA stability
Vascular smooth muscle cells were treated with either 10 −5 mmol/L Ang II for 48 hours or nucleolin siRNA plasmid, pcDNA3.1-Nuc plasmid or Nuc1-309 plasmid for the indicated periods, and the cells were then incubated with either 0.5% ethanol or 5 μg/mL actinomycin D in 0.5% ethanol. Aliquots were removed from the cultures at 30-minutes intervals over a 3-hours time course.
Actinomycin D (5 μg/mL) induced no RNA fragmentation during this period. At the indicated time-points (0, 0.5, 1, 2 and 3 hours), 2 × 10 5 cells were harvested, and total RNA was isolated using the Rneasy kit. PCR amplification of the pooled cDNA was carried out.

| Synthesis of biotinylated RNA probes
The biotinylated EGF and PDGFB RNA probes were synthesized by Beijing Dingguo Changsheng Biotechnology Co., Ltd. (Table 1). The biotin-labelled RNA probe was denatured and annealed into a double strand (program: 94°C for 5 minutes, gradually returned to room temperature), and the annealing effect was examined by 12% PAGE gel, after which biotin-labelled RNA was diluted, packed and stored at −20°C prior to further analysis.
as follows. Briefly, primers were designed and synthesized according to the gene sequence of interest and the vector MCS site.
The EGF gene was cloned from the cDNA of rat, and the PDGFB was synthesized by whole gene. PCR was carried out using primers harbouring XhoI + HindIII and KpnI + HindIII restriction sites, fragments of about 395 and 388 bp in length were amplified, and the recovered PCR products and pGL3-basic vector were simultaneously subjected to a double enzyme (XhoI/HindIII) digestion. In addition, the enzyme-digested product gel was recovered and then subjected to a ligation reaction (attached with T4 ligase overnight), followed by PCR amplification. The fragment was cloned into the pGL3-luc vector. The bacterial transformation was carried out, and the monocolonies were subjected to enzyme digestion and sequencing to confirm the successful construction of the recombinant plasmids.

| Luciferase reporter gene assay
Cells cultured in 96-well plates were cotransfected with the firefly-Luciferase pGL3-promotor vector (pGL3p; Promega) or pGL3p harbouring the 5′ UTR of EGF and PDGF-BB, as well as the Renilla-

| Statistical analysis
Data were expressed as mean ± SEM based on at least three independent experiments. Statistical analysis was performed by oneway ANOVA (LSD test) for multiple testing. P < .05 was considered as statistically significant.

| Bioinformatics analysis
The mRNA sequences of phenotypic transformation-associated genes in VSMCs were screened by bioinformatics analysis, and multiple phenotypic transformation-associated genes containing nucleolin-binding elements '(T/G) CCCG (A/G)' were identified.
Bioinformatics analysis found that the mRNA sequences of 12 genes contained nucleolin-binding elements, including EGF and PDGF-BB. 10 The analysis results suggested that nucleolin bound to the mRNA of these genes, regulated the expressions of these  ELISA results showed that the secretion of EGF and PDGF-BB was also increased with the increase of AngII concentration and exposure duration ( Figure 1C).

| The role of Nuc1-309 (lacking the carboxy terminus of nucleolin) in the expression and secretion of EGF and PDGF-BB
The above-mentioned results indicated that nucleolin had a regulatory effect on the expressions and secretion of EGF and PDGF-BB, while such regulatory mechanism remained unknown. Since nucleolin is an RNAbinding protein, we speculated that it might bind to EGF and PDGF-BB mRNA molecules through its RBD to regulate their mRNA stability and expression. Therefore, we further used a nucleolin mutant lacking a carboxy terminus (Nuc1-309) (ie the amino acid containing the RBD was could not significantly increase the expression of nucleolin (Figure 4).

| The effect of nucleolin overexpression and low expression on the stability of EGF and PDGF-BB mRNA
The results showed that the expressions of EGF and PDGF-BB at the mRNA level were gradually decreased after the treatment of transcription inhibitor actinomycin D (5 μg/mL) in normal cells. Compared

| Binding of nucleolin protein to EGF and PDGF-BB mRNA
To confirm the presence of interaction between nucleolin and EGF or PDGF-BB mRNA, we used protein immunoprecipitation and RT-qPCR to verify their interaction. Results showed that only a small amount of EGF and PDGF-BB mRNA was precipitated in normal cell lysate.
However, the amount of EGF and PDGF-BB mRNA precipitated by nucleolin antibody was increased compared with the control IgG group, suggesting that nucleolin could bind to EGF and PDGF-BB mRNA under the normal condition. The amount of EGF and PDGF-BB mRNA in cell lysate was significantly increased after Ang II stimulation. After coimmunoprecipitation with nucleolin antibody, the EGF and PDGF-BB mRNA precipitated by nucleolin antibody were significantly increased compared with the control IgG group, suggesting that the binding ability of nucleolin to EGF and PDGF-BB mRNA molecules was increased after VSMCs were treated with Ang II. We found that the binding of nucleolin to EGF and PDGF-BB mRNA molecules was specific using β-actin as a control. Western blotting analysis was used to detect the content of nucleolin in the cell extracts and sediments, which confirmed the effectiveness of the nucleolin antibody ( Figure 6A,B).

| Binding of nucleolin to the 5' UTR of EGF and PDGF-BB mRNA
Preliminary results confirmed that nucleolin could bind to the mRNA of EGF and PDGF-BB. However, it remained unclear how nucleolin bound to EGF and PDGF-BB mRNA and then regulated their stability.
Based on bioinformatics analysis, we speculated that nucleolin bound to the 5′ UTR of EGF and PDGF-BB mRNA to increase their stability.
We extracted the cytoplasmic proteins of normal VSMCs, nucleolin over-expressing and mutant (Nuc 1-309) over-expressing VSMCs and used RNA-EMSA to detect the binding of nucleolin to 5′ UTR of EGF and PDGF-BB mRNA. The results showed that the probe retention signal (probe binding band) was observed in normal VSMCs, and the probe binding band was significantly enhanced after overexpression of nucleolin. The binding band of nucleolin mutant Nuc1-309 was not significantly enhanced. When the non-biotinylated EGF and the PDGF-BB 5′ UTR RNA probe (100-fold concentration) competed, the binding band disappeared. If the nucleolin antibody was added to the reaction system, the signal of the original binding band was significantly weakened, and a supermigration band appeared above the binding band ( Figure 7A,B). This result suggested that nucleolin could bind to the 5′ UTR of EGF and PDGF-BB mRNA. In order to observe the effect of Ang II treatment on the binding activity of nucleolin to EGF and PDGF-BB mRNA, the nuclear proteins of nucleolin overexpression and control plasmids were extracted, respectively.
The binding of nucleolin to EGF and PDGF-BB mRNA was detected by EMSA. The results showed that after Ang II treatment, the binding activity of nucleolin to EGF and PDGF-BB mRNA in control plasmid and nucleolin over-expressing groups was significantly higher compared with that without Ang II treatment, and the binding ability of nucleolin over-expressing group was stronger compared with the control plasmid ( Figure 7C,D).

| D ISCUSS I ON
Our previous studies have shown that nucleolin expression is up- The effect of nucleolin overexpression and low expression on the stability of EGF and PDGF-BB mRNA. A, Effect of nucleolin overexpression on the stability of EGF and PDGF. B, Effect of nucleolin mutant (Nuc1-309) on the stability of EGF and PDGF. VSMCs were transiently transfected with nucleolin overexpression plasmid and its mutant expression plasmid for 48 h. After adding the transcription inhibitor actinomycin D (actinomycin D, 5 μg/mL), the cells were collected at different time-points (0, 0.5, 1, 2 and 3 h), and the mRNA levels of EGF and PDGF were detected by RT-qPCR. Control: normal VSMCs; pcDNA3.1: control plasmid group; pcDNA3.1-Nuc: nucleolin overexpression plasmid group; Nuc1-309: nucleolin mutant plasmid lacking the carboxy terminus of nucleolin (ie deleting the amino acid containing the RNA-binding domain); β-actin was used an internal control (Data were expressed as X ± S, n = 5; * P < .05, # P < .01 vs the pcDNA3.1 and Nuc1-309 group.) (C) Effect of Ang II on the stability of EGF and PDGF. After treatment of VSMCs with 10 −5 mmol/L Ang II for 48 h, cells were harvested at different time-points (0, 0.5, 1, 2 and 3 h) after adding the transcription inhibitor actinomycin D (5 μg/mL), and the mRNA levels of EGF and PDGF were detected by RT-qPCR. Control: normal VSMC group; Ang II: Ang II treatment group; β-actin was internal control (Data were expressed as X ± S, n = 5; * P < .05, # P < .01 vs the control group). D, Effect of low expression of nucleolin on the stability of EGF and PDGF induced by Ang II. After VSMCs were transfected with nucleolin interference plasmid for 24 h, VSMCs were treated with 10 −5 mmol/L Ang II for 48 h, cells were harvested at different time-points (0, 0.5, 1, 2 and 3 h) after adding the transcription inhibitor actinomycin D (5 μg/mL), and the mRNA levels of EGF and PDGF were detected by RT-qPCR. Ang II: Ang II treatment group; PsiRNA: control plasmid; PsiRNA-Nuc: nucleolin RNA interference plasmid. β-actin was used as an internal control (Data were expressed as X ± S, n = 5; * P < .05, # P < .01 vs the PsiRNA and Ang II-treated group)

F I G U R E 6
The binding of nucleolin protein with EGF and PDGF-BB mRNA in VSMCs. A, Binding of nucleolin protein to PDGF-BB mRNA; B, binding of nucleolin protein to EGF mRNA Immunoprecipitation and RT-PCR were used to analyse the combination of nucleolin and EGF, PDGF-BB mRNA. Normal VSMCs and Ang II-treated VSMCs were collected to prepare the cell extracts, and cell extracts were divided into three equal groups as follows: Input group, negative control IgG group and nucleolin antibody group. The nucleolin antibody was used for co-immunoprecipitation, and the total RNA in half of the precipitate was extracted. The EGF, PDGF-BB and β-actin specific primers were used for RT-qPCR detection, and the other half of the precipitate was detected by Western blotting analysis for the amount of nucleolin. inhibit phenotypic transformation, proliferation and migration of VSMCs, thereby reducing the formation of angiogenic intima. 24,25 Similarly, EGF is also one of the most important cytokines known to promote proliferation and migration of VSMCs. As an important cell growth factor, EGF binds to its specific receptor (EGFR) and activates a series of cell signalling pathways, playing an important role in phenotypic transformation, proliferation and differentiation of VSMCs. 26 Studies have shown that EGF and EGFR-mediated signalling pathways play a very important role in remodelling of blood vessel walls and neointimal formation after vascular injury.
Moreover, the inhibition of atherosclerosis by suppressing EGF signalling has been validated in different animal models. 27,28 In

| CON CLUS IONS
In conclusion, our current study demonstrated that nucleolin increased

E TH I C A L A PPROVA L S TATE M E NT
The procedures in the study were scrutinized and approved by Medical Ethics Committee of Xiangya Hospital, Central South University.