Proliferation of Vascular Smooth Muscle Cells under ox-LDL Is Regulated by Alismatis rhizoma Decoction via InhibitingERK1/2 and miR-17∼92a Cluster Activation

Context: Alismatis rhizome decoction (AD) exhibits antiatherosclerotic activities. The activity of AD against vascular smooth muscle cell (VSMC) proliferation remains unclear. Objective. The mechanisms and effects of AD on oxidized low-density lipoprotein (ox-LDL)-induced VSMC proliferation were explored. Materials and methods. The male SD rats were fed with AD (2.56 g/mL) or 0.9% NaCl by oral gavage 4 mL twice daily for 7 d. Then, AD-containing serum (ADcs) was collected. MTS assay was applied to measure the VSMC viability. The proliferation of VSMCs was detected by 5-bromodeoxyuridine (BrdU) immunocytochemistry. The microRNA (miRNA) profiling was performed, and the target genes of miRNAs were searched from the TargetScan 7.2 database. The expressions of matrix metalloproteinases-2/9 (MMP-2/9), cyclin D1/E, cyclin-dependent kinase inhibitor 1B (p27), extracellular regulated protein kinases 1/2 (ERK1/2), and ERK1/2 phosphorylation were examined by western blotting or quantitative reverse transcription PCR. Results. The ox-LDL-induced miR-17-92a expression promoted VSMC proliferation. AD and the ERK1/2 inhibitor U0126 (10 μmol/L) inhibited VSMC proliferation and reduced the overexpression of miR-17∼92a. AD was found to inhibit phosphorylation of ERK1/2 and reduced the expression of MMP-2/9 in VSMCs. The expression of cyclin D1/E was suppressed, and p27 was elevated following treatment with AD as well as ERK1/2 inhibitor. According to the TargetScan 7.2 database, the target genes of miR-17∼92a act on tissue inhibitors of metalloproteinases (TIMPs)-MMPs, p27/21 cyclins, and peroxisome-proliferator-activated receptor α (PPARα) ATP-binding cassette transporter (ABC) A1/G1, which are involved in the process of atherosclerosis. Conclusions. AD inhibits ox-LDL-induced VSMC proliferation via inhibiting ERK1/2 and miR-17∼92a activation. The results provide the multitarget mechanisms for application of AD in the treatment of atherosclerosis. It would be helpful to the treatment of cardiovascular and cerebral diseases.


Introduction
As the main constitutive stromal cells of the vascular wall, vascular smooth muscle cells (VSMCs) are the only cell type in the medial layer of the arterial wall. It is recognized that VSMCs proliferate at a very low frequency in the normal vessel wall [1]. However, vascular injury has been found to induce aberrant VSMC proliferation and migration [2]. Increasing evidence shows that abnormal VSMC proliferation plays a central role in vascular pathogenesis [3,4].
Cumulative evidence indicates that cell-cycle progression is mediated by cell-cycle regulatory proteins, such as cyclin D1, cyclin E, and cyclin-dependent kinase inhibitor 1B (p27), which are associated with G0/G1 phase cell-cycle arrest [5]. In vascular diseases, matrix metalloproteinases (MMPs) have been identified as a key enzyme in cell wall remodeling, plaque formation, and restenosis [6]. MMPs (MMP-2 and MMP-9) are reported to promote the proliferation of VSMCs [7] and repressed by the tissue inhibitors of metalloproteinases (TIMPs). In addition, MMP-2 and MMP-9 expression is stimulated by ox-LDL in VSMCs, which is mediated by the activation of the ERK signaling [8,9].
Alismatis rhizoma decoction (AD), a classic traditional Chinese medicinal formula used for the treatment of cardiovascular and cerebral diseases, consists of the combination of Alisma orientalis and Atractylodes macrocephala [15]. AD exhibited multiple pharmacological actions, including hepatoprotective, anti-inflammatory, anticancer, and antioxidant activities [9,[16][17][18][19]. Our previous study showed that AD reduced lipid deposition in macrophagesderived foam cells [20] and a triterpenoid derived from Alisma orientale (Alisol A 24-acetate) suppressed the migration and phenotypic transformation of ox-LDL-induced VSMCs [8]. However, little is known about the effect of AD on VSMC proliferation. e present study investigates the impact of AD on ox-LDL-induced VSMC proliferation and explores the underlying mechanisms.

Ethical Statement and Animals.
is study was approved by the Ethics Review Committee of the Affiliated Rehabilitation Hospital of Fujian University of Traditional Chinese Medicine (FJKFYY-201400127). All animal experiments were performed in accordance with the Management of Animal Care and Use Programs in Research, Education, and Testing [21]. Male 6-week-old Sprague-Dawley (SD) rats, each weighing 200 to 220 g, were housed in a temperaturecontrolled (21°C) environment under a 12-h light/dark cycle and given free access to water and food.

Preparation of Alismatis rhizoma Decoction and ADcs.
AD and ADcs was prepared as described previously [20]. Briefly, crude materials of Alismatis rhizoma decoction were supplied by the Fujian University of Traditional Chinese Medicine Rehabilitation Hospital Pharmacy and were carefully identified. Alisma orientalis and Atractylodes macrocephala (5 : 2) were soaked in water for 30 min and mixed in proportion and decocted twice by refluxing with water (1: 6 and then 1: 4, w/v) for 1 h. e filtrates were combined and condensed. ey were then stored at 4°C until use. e SD rats were assigned to 2 groups (15 rats for each group). e AD group was given AD by oral gavage 4 mL twice daily for 7 d (approximately 2.56 g/mL Alisma orientalis and Atractylodes macrocephala; Affiliated Rehabilitation Hospital of Fujian University of Traditional Chinese Medicine, Fuzhou, China), and the control group was given 0.9% NaCl by oral gavage 4 mL twice daily for 7 d. en, blood samples were collected from the rat abdominal aorta following induction of anesthesia with diazepam/ketamine (1 : 1) at a dose of 0.1 mL/100 g, and serum was separated by centrifugation at 3000 rpm at 4°C for 10 min and inactivated in a 56°C water bath for 30 min. en, the bacteria were filtered through a microporous membrane, and the serum was stored at −20°C for the subsequent experiments.

Isolation and In Vitro Culture of Rat VSMCs and
Grouping. Rats were sacrificed, and the thoracic aorta was isolated under aseptic conditions as described previously [8].
e intima and adventitia of the thoracic aorta were removed, and the specimens of thoracic aorta were cut into pieces measuring 1 × 1 mm, digested with type II collagenase at 37°C for 14 h, and then incubated in the DMEM/F-12 medium (Hyclone, Logan, UT, USA) supplemented with 20% fetal bovine serum (FBS; Hyclone). e culture medium was replaced once every 2-3 days with the DMEM medium supplemented with 10% FBS at 37°C in humidified atmosphere containing 5% CO 2 . VSMCs from 4 to 8 passages were used for experiments. Cells grown to 80-95% confluence were made quiescent by serum starvation for 24 h, and then VSMCs were grouped. Cells in the ox-LDL group were cultured in the DMEM/F12 medium supplemented with 20% normal rat serum containing 50 mg/L ox-LDL (Peking Union-Biology Co, Ltd., Beijing, China) for 12, 24, and 48 h, cells in the AD group were cultured in the DMEM/ F12 medium supplemented with 20% ADcs and 50 mg/L ox-LDL for 12, 24, and 48 h, and cells in the U0126 group were cultured in the DMEM/F12 medium supplemented with 20% rat serum containing 10 μmol/L U0126 (Sigma-Aldrich, Inc., St. Louis, MO, USA) and 50 mg/L ox-LDL for 12, 24, and 48 h, while cells in the control group were cultured in the DMEM/F12 medium supplemented with 20% normal rat serum for 12, 24, and 48 h.

MTS Assay.
To measure the cell viability, VSMCs were seeded onto 96-well plates (Corning, Inc.; Corning, NY, USA) at a density of 1 × 10 4 cells per well and cultured for 24 h until 80% confluence. Following incubation in serumfree medium during the initial 24 h, VSMCs were exposed to ADcs at concentrations of 0%, 10%, 20%, 30%, 40%, and 50% for 12 and 24 h, respectively. en, the viability of VSMCs was measured using the MTS assay (Promega, Madison, WI, USA) following the manufacturer's instructions.

Incorporated BrdU of Immunohistochemistry Assay for VSMC Proliferation.
To measure the VSMC proliferation, VSMCs were seeded onto 6-well plates at a density of 5 × 10 5 cells per well and incubated in the DMEM medium supplemented with 10% FBS for 24 h, followed by serum starvation for 24 h to make a synchronization of cell growth in the G0 phase of the cell cycle. ey were then assigned to 4 groups: the control group, ox-LDL group, ox-LDL + 20% ADcs, and ox-LDL + U0126 group. Following treatment with ox-LDL, 20% ADcs, and U0126 for 12 and 24 h, BrdU (30 μmol/L; APExBIO, USA) was added to the cell culture medium 12 h before collecting the cells. Immunocytochemical staining for BrdU expression was performed on VSMCs by the SP method and following the manufacturer's instructions. Briefly, VSMCs were fixed in 4% paraformaldehyde, rinsed with PBS, 2 mol/L hydrochloric acid for 30 min at 37°C, blocked with goat serum, incubated with mouse anti-rat BrdU monoclonal antibody (1 : 1000; Proteintech, CA, USA), rinsed with PBS, incubated with biotin-conjugated goat anti-mouse IgG antibody (1 : 1000; BOSTER Biological Technology, Wuhan, China), followed by incubation in streptavidin, avidin, and peroxidase, visualized with diaminobenzidine (DAB; Maxim-Bio, Fuzhou, China) and DAPI dye solution (100 ng/mL; BOSTER Biological Technology, Wuhan, China), drained the dye solution, and observed under microscopy. In the negative control group, the primary antibody was replaced with PBS. Five fields of vision were randomly selected from each slip under a microscope at 400× magnification and then photographed. e images captured were analyzed using the image processing software Image-Pro Plus v6.0 (Media Cybernetics, Bethesda, MD, USA). Positive BrdU expression was defined as the presence of pale-yellow or deep-tan staining, while nonspecific background staining was excluded. Positive DAPI cells were defined as the presence of blue. All experiments were repeated in triplicate. e ratio of BrdU-positive cells and total cells was estimated.

miRNA Microarray Assay.
e cells were treated as described above. Total RNA was extracted from VSMCs. e miRNA microarray profiling was performed using the Lianchuan Biological Small RNA Sequencing Analysis (Lianchuan Bio, China) according to the manufacturer's protocol. Small RNA sequencing library preparation was performed using the TruSeq Small RNA Sample Prep Kits (Illumina, San Diego, CA USA) kit. After the library preparation work was completed, the constructed library was sequenced using Illumina Hiseq 2000/2500, and the sequencing read length was single-ended 1 × 50 bp.

Western Blotting Analysis. Protein was extracted from
VSMCs, and the protein concentration was quantified with the BCA protein assay kit (TransGen Biotech Co., Ltd., Beijing China). Total protein was separated by 10% SDS-PAGE, and the blots were electrotransferred to a PVDF membrane with a 0.45 μm pore size (Millipore, Billerica, MA, USA). e membranes were blocked with 5% (w/v) fatfree milk in 0.1% (v/v) TBST at room temperature for 2 h. (1 : 1000; Cell Signaling Technology, Inc., Danvers, MA, USA) at 4°C overnight, while anti-β-actin mouse monoclonal antibody (BOSTER Biological Technology Co, Ltd., Wuhan, China) served as a loading control. e membranes were then washed three times with TBST, of 10 min each time, and incubated with anti-mouse or anti-rabbit IgG HRP-conjugated secondary antibodies (1 : 3000; BOSTER Biological Technology Co, Ltd., Wuhan, China) at room temperature for 1 h. After the membranes were rinsed three times with the wash buffer, the peroxidase activity was detected by an enhanced chemiluminescence (ECL) kit with the ChemiDoc XRS + system (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and then the quantification of the band intensity was analyzed using the Image Lab Software, version 3.0 (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

qRT-PCR Assay.
Total RNA was extracted from VSMCs with the Trizol reagent (TaKaRa Biotechnology, Dalian, China) following the manufacturer's instructions, and the RNA concentration was quantified. e total RNA was reversely transcribed into cDNA. e expression of MMP-2 and MMP-9 was quantified using the qRT-PCR assay according to the previous study [19]. PCR was performed according to the manufacturer's instructions. According to GenBank, the RT-PCR primers were designed as follows: MMP-9 forward: 5′-CAAGGACGGTCGGTATTGGAAG-3′ and reverse: 5′-AAACGAGTAACGCTCTGGGGAT-3′, with 348 bp in length; MMP-2 forward: 5′-TCCCGTTATGA-GACCCTGAGC-3′ and reverse: 5′-AGGACGCA-GAGAACCCTGAGAG-3′, with 205 bp in length; and GAPDH forward: 5′-ACGGCAAGTTCAACGGCACAG-3′ and reverse: 5′-GAAGACGCCAGTAGACT CCACGAC-3′, with 149 bp in length. e PCR products were checked by electrophoresis on 1.5% agarose gel, and the data were analyzed using the Image Lab Software, version 3.0. e relative mRNA expression level was estimated by the ratio of the gray scale value of target gene to GAPDH.
2.9. Statistical Analysis. All the experimental data were expressed as mean ± SD. All statistical analyses were performed using the statistical software SPSS, version 18.0 (SPSS, Inc., Chicago, IL, USA), of three independent experiments and assessed by one-way analysis of variance (ANOVA) followed by Fisher's protected least-significant difference test to compare two treatments, and P value <0.05 was considered statistically significant.

Morphology of Rat VSMCs.
e rat VSMCs appeared as a single shape or an irregular triangle, with abundant high-intensity cytoplasm, and the nucleus appeared round and was located in the center. After growing to confluence, VSMCs displayed a typical peak and valley growth morphology (Supplementary Materials available here). As shown in Supplementary Materials, cells were positive for α-smooth muscle actin (α-SMA, 1 : 1000; BOSTER Biological Technology, Wuhan, China) Evidence-Based Complementary and Alternative Medicine staining and myofilaments in the cytoplasm ran along the long axis of the cell. erefore, cells were identified as VSMCs.

Cytotoxicity of ADcs to VSMCs.
ere was no significant difference in body weight, mortality, and disability between the AD group and the control group during the preparation of ADcs. Quiescent VSMCs were exposed to ADcs at concentrations of 10%, 20%, 30%, 40%, and 50% for 12 (Figure 1(a)) and 24 h (Figure 1(b)) in the absence of ox-LDL, and the MTS assay revealed no significant difference in the viability of VSMCs treated with ADcs as compared with untreated cells.
e results indicated that ADcs had no cytotoxicity to VSMCs, and 20% ADcs was selected as an intervention concentration.

AD Inhibits ox-LDL-Induced VSMC Proliferation.
Incorporated BrdU of immunocytochemistry assay revealed that ox-LDL treatment increased the number of BrdU-positive cells (new proliferation of VSMCs) (p < 0.01, Figure 2); however, treatment with AD markedly reduced the number of BrdU-positive VSMCs at 12 and 24 h (p < 0.01, Figure 2), indicating that AD may suppress ox-LDL-induced VSMC proliferation.

AD Inhibits ERK1/2 Phosphorylation in ox-LDL-Induced
VSMCs. Since the ERK1/2 signaling pathway is associated with the proliferation of VSMCs, we assessed the effect of AD on the activation of the ERK1/2 pathway. Western blotting assay showed ox-LDL increased the expression of p-ERK1/2 in VSMCs (p < 0.05, Figure 3), and the treatment with AD and U0126 significantly suppressed p-ERK1/2 expression at 12, 24, and 48 h (p < 0.05 for AD and p < 0.01 for U0126, Figure 3). ese results suggested that AD inhibited ox-LDLinduced VSMC proliferation through downregulating the activity of the ERK1/2 signaling pathway.
Exploring the role of miR-17∼92a cluster in the VSMC proliferation, we searched for its possible target genes according to the TargetScan 7.2 database. Target genes of miR-17∼92a cluster are directly related to the expression of TIMP-2, TIMP-3, TIMP-4, p21, p21-activated kinases (PAks), and p27, which are associated with cell proliferation (Figure 4). TIMP-2, TIMP-3, and TIMP-4 are the important factors for the expression of MMP-2 and MMP-9. p21 and p27 show the negative correlation with cyclin D1 and cyclin E expressions, which are the target genes of miR-92a-1-5p and miR-18a-3p in the miR database.

Effects of AD and ERK1/2 Blocker U0126 on Cell-Cycle Regulatory Proteins in VSMCs.
As AD inhibited ox-LDLinduced VSMC proliferation, we hypothesized that AD may be associated with the expression of cell-cycle-associated proteins in VSMCs. Western blotting showed that ox-LDL increased expression of cyclin D1 and cyclin E in VSMCs (p < 0.05 for 12 h, p < 0.01 for 24 and 48 h) and decreased the p27 expression; AD and U0126 dramatically inhibited the overexpression of cyclin D1 and cyclin E (p < 0.01 for 12, 24, and 48 h). However, the expression of p27 was upregulated following the exposure to AD and U0126 (p < 0.05 for 12, 24, and 48 h, Figure 6).

Discussion
As an ancient classical traditional Chinese medicinal formula, AD (Alisma orientalis and Atractylodes macrocephala), was first described in the Eastern Han Dynasty and has been used for the treatment of cardiovascular and cerebral diseases in China for decades [16,17,20]. AD with the ED50 value of 8.04 g/mL and IC50 value of 3.15 or 5.14 g/mL were showed in cellular experiments [22,23]. e rats treated with AD (2.56 g/mL, 4 mL twice daily) were proved to be effective, and no side effects in our study were found. During atherogenesis, the proliferation and migration of arterial VSMCs from media to intima had been shown to contribute to neointimal VSMC accumulation [4], where ox-LDL  Figure 1: Effects of AD on the viability of VSMCs. Cells are exposed to AD at concentrations of 10%, 20%, 30%, 40%, and 50% for 12 h (a) and 24 h (b), respectively. MTS assay shows 20% AD had no cytotoxicity to VSMCs.   serves as a chemoattractant to VSMCs [24]. Although AD exhibits a wide range of bioactivities in diverse cells [8,20], its effect on ox-LDL-induced proliferation of VSMCs remain unknown until now. e present in vitro study was therefore designed to examine the potential role of AD through suppressing the proliferation of VSMCs stimulated by ox-LDL.
It has been proposed that the differentiation, proliferation, and apoptosis of cells are regulated by miRNAs. More than one-third of protein-coding genes are under miRNA translational control, with a single miRNA able to regulate hundreds of protein-coding genes. An miR-17∼92a cluster is a group of highly conserved miRNAs, located at chromosome 13q31-q32 [25], including miR-17, miR-18a, miR-19a/b, miR-20a, and miR-92a; miR-17∼92 is required for normal animal development and cell proliferation, individual member of the miR-17∼92 cluster appears to possess distinct function [26][27][28][29][30]. Several studies reported that miR-17∼92 played a key role in the regulation of cardiomyocyte proliferation [26] and miR-17 stimulated the proliferation of VSMCs, enhanced cell-cycle G1/S transition [11], and regulated p21 expression [30]. In the present study, we found that ox-LDL upregulated miR-17∼92a cluster expression and increased the number of new proliferations of VSMCs, AD treatment resulted in a significant reduction of miR-17∼92a expression, and the capacity of proliferation in ox-LDL-induced VSMCs was markedly inhibited following the treatment with AD. It has been recently confirmed that the miR-17∼92a cluster are mediated by the ERK pathway in endothelial cells [16] and miR-18a promotes cell proliferation via AKT and ERK pathways [31]. We found ox-LDL-induced the phosphorylation of ERK1/2 in VSMCs, AD inhibited the ERK1/2 TIMP-2, TIMP-3, and TIMP-4 are regulated by miR-20a-3p, miR-92a-1-5p, and miR-18a-3p. p21 and p27 were regulated by miR-92a-1-5p and miR-18a-3p, and the miR-17∼92a cluster acts on the p21 activated kinases (PAks) according to TargetScan7.2 database prediction.    Evidence-Based Complementary and Alternative Medicine pathway activation, and blocking ERK1/2 pathways dramatically suppressed miR-17∼92a cluster expression. We consider that AD may inhibit ox-LDL-induced VSMC proliferation, which is associated with the ERK1/2 pathway and miR-17∼92a cluster activation.
VSMC proliferation has shown an essential role in the progression of atherosclerotic lesions. p21 and p27 are regulated by miR-92a-1-5p and miR-18a-3p, and the miR-17∼92a cluster acts on p21-activated kinases (PAks) according to the TargetScan7.2 database prediction. As a  Figure 6: Effects of AD and ERK1/2 inhibitor U0126 on the expression of cell-cycle regulatory proteins in ox-LDL-induced VSMCs. Western blot analysis was performed to determine the expression of cyclin D1, cyclin E, and p27, while β-actin served as a loading control. e graphs represented the relative expression of these proteins for three independent experiments. * p < 0.05 ox-LDL vs. control group; * p < 0.01 ox-LDL vs. control group; # p < 0.05 AD or U0126 vs. the ox-LDL group; ## p < 0.01 AD or U0126 vs. the ox-LDL group. negative regulator of cyclins, p27, p21, and PAks are reported to play an important role in the cell-cycle arrest [5,32]. It is known that cyclin D, cyclin E, and p27 are important cyclins in cell-cycle progression, which are closely related to cell proliferation [33]. Progression through the G1 phase of the cell cycle is known to be controlled by cyclin D1 and cyclin E, and it is reported that cyclin D1 and cyclin E are necessary for transition from G1 to S phase, which is also known as the DNA synthesis phase. Physiologically, p27 maintains at a high level in the G0/G1 phases and allows cells at the G1 phase and quiescent [32]. Upon mitogenic stimulation, p27 is rapidly degraded, thus enabling cyclin D1 and cyclin E to promote cell proliferation. We detected that treatment with AD and ERK1/2 inhibitor U0126 increased p27 protein expression and reduced the expression of cyclin D1 and cyclin E proteins in VSMCs exposed to ox-LDL, suggesting that treatment with AD and ERK1/2 blocker upregulated p27 expression and inhibited cyclin D1 and cyclin E expressions, which may be linked to miR-17∼92a cluster expression.
It was estimated that MMP-2 and MMP-9 in VSMCs were crucial in the vascular remodeling [34] and the constitutive expression of MMP-2 and MMP-9 has been linked to VSMC proliferation [35,36]. Inhibition of MMP-2 and MMP-9 has been employed as a potential strategy to prevent atherosclerosis. In this study, AD suppressed the overexpression of MMP-2 and MMP-9 in ox-LDL-induced VSMCs. Similar effects were found in ox-LDL-treated VSMCs exposed to the ERK1/2 inhibitor U0126. Xu reported that the overexpression of miR-20a suppressed TIMP-2 expression and increased the expression of MMP-2 and MMP-9 [37]. It has been reported that the miR-92a expression in VSMCs is low in the quiescent state and miR-92a regulates the expression levels of MMP-9 and TIMP-3 in the hydrogen-peroxideinduced VSMCs [13]. According to the TargetScan7.2 database, TIMP-2, TIMP-3, and TIMP-4 are the target genes of miR-20a-3p, miR-92a-1-5p, and miR-18a-3p, and TIMPs is the negative regulator of MMPs. e data showed that AD regulated the miR-17∼92a cluster expression and further may affect the TIMP-MMP expression in the ox-LDL-induced VSMCs.
Several targets of miR-17∼92a in the process of atherosclerosis have been predicted. According to the Tar-getScan7.2 database, mir-17a-5p and mir-18a are associated with ATP-binding cassette transporter A1 (ABCA1) and ATP-binding cassette transporter G1 (ABCG1), which are the important factors in atherosclerosis. It has been demonstrated that mir-17a-5p is associated with ABCA1 expression [23,38] and lipid metabolism. Upregulation of ABC protein family (ABCA1 and ABCG1) can enhance the cholesterol efflux and reduce the formation of aortic plaques [39]. PPAR-alpha and PPAR-gamma activators induce the expression of ABCA1, which controls apoAI-mediated cholesterol efflux from macrophages [40]. PPAR-alpha is a major regulator of lipid metabolism and regulated by miR-92a-1-5p and miR-18a-3p. erefore, AD may regulate atherosclerosis associated with the lipid regulation of mir-17∼92a cluster.

Conclusion
e results of our study demonstrate that treatment with AD inhibits ox-LDL-stimulated VSMC proliferation, which is strongly associated with the inhibition of the ERK1/2 and miR-17∼92a activation. Our data provide new insights into the mechanisms of AD on atherosclerosis, and it provides a theoretical basis for the clinical application of AD.
Data Availability e data sets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Disclosure
Julian Shen and Wei Wei are co-first author.

Conflicts of Interest
e authors declare no conflicts of interest.