CircPMS1 promotes proliferation of pulmonary artery smooth muscle cells, pulmonary microvascular endothelial cells, and pericytes under hypoxia

Abstract

Background

Circular RNAs (circRNAs) have been recognized as significant regulators of pulmonary hypertension (PH); however, the differential expression and function of circRNAs in different vascular cells under hypoxia remain unknown. Here, we identified co‐differentially expressed circRNAs and determined their putative roles in the proliferation of pulmonary artery smooth muscle cells (PASMCs), pulmonary microvascular endothelial cells (PMECs), and pericytes (PCs) under hypoxia.

Methods

Whole transcriptome sequencing was performed to analyze the differential expression of circRNAs in three different vascular cell types. Bioinformatic analysis was used to predict their putative biological function. Quantitative real‐time polymerase chain reaction, Cell Counting Kit‐8, and EdU Cell Proliferation assays were carried out to determine the role of circular postmeiotic segregation 1 (circPMS1) as well as its potential sponge mechanism in PASMCs, PMECs, and PCs.

Results

PASMCs, PMECs, and PCs exhibited 16, 99, and 31 differentially expressed circRNAs under hypoxia, respectively. CircPMS1 was upregulated in PASMCs, PMECs, and PCs under hypoxia and enhanced the proliferation of vascular cells. CircPMS1 may upregulate DEP domain containing 1 (DEPDC1) and RNA polymerase II subunit D expression by targeting microRNA‐432‐5p (miR‐432‐5p) in PASMCs, upregulate MAX interactor 1 (MXI1) expression by targeting miR‐433‐3p in PMECs, and upregulate zinc finger AN1‐type containing 5 (ZFAND5) expression by targeting miR‐3613‐5p in PCs.

Conclusions

Our results suggest that circPMS1 promotes cell proliferation through the miR‐432‐5p/DEPDC1 or miR‐432‐5p/POL2D axis in PASMCs, through the miR‐433‐3p/MXI1 axis in PMECs, and through the miR‐3613‐5p/ZFAND5 axis in PCs, which provides putative targets for the early diagnosis and treatment of PH.

The expression of circPMS1 was upregulated in pulmonary artery smooth muscle cells (PASMCs), pulmonary microvascular endothelial cells (PMECs), and pericytes (PCs) under hypoxic conditions. CircPMS1 may upregulate DEPDC1 and POLR2D expression by targeting miR‐432‐5p in PASMCs, upregulate MXI1 expression by targeting miR‐433‐3p in PMECs, and upregulate ZFAND5 expression by targeting miR‐3613‐5p in PCs.

1. INTRODUCTION

Pulmonary hypertension (PH) is a fatal disease characterized by a continuous increase in pulmonary arterial pressure and pulmonary vascular resistance, ultimately leading to right heart failure and death. ¹ PH can be caused by chronic hypoxia, resulting in the hyperproliferation of pulmonary artery smooth muscle cells (PASMCs) and apoptosis‐resistant pulmonary microvascular endothelial cells (PMECs) ² , followed by the remodeling of pulmonary vessels. ³ , ⁴ Pericytes (PCs) appear in large numbers around remodeled vessels and exhibit a pro‐proliferative and pro‐migratory phenotype in vitro, indicating that PCs are also involved in pulmonary vascular remodeling ⁵ ; however, the pathogenesis of PH is complex and has not yet been fully elucidated. Therefore, there is a need to identify the molecular mechanisms underlying PH.

Circular RNAs (circRNAs) are a novel class of single‐stranded RNAs with a closed‐loop structure formed by a backsplicing event. ⁶ CircRNAs exhibit tissue, cell, and developmental stage‐specific expression patterns and are stable, endogenous, abundant, and conserved. ⁶ , ⁷ , ⁸ Importantly, circRNAs are present in vascular cells and related to their various functions as well as the development of PH. ² , ⁹ , ¹⁰ , ¹¹ Our previous study demonstrated that circular RNA‐gamma‐secretase‐activating protein (circGSAP) was downregulated in hypoxic PMECs. ¹¹ , ¹² , ¹³ Downregulation of circGSAP acted as microRNA‐942‐5p (miR‐942‐5p) to reduce the expression of SMAD family member 4 (SMAD4), thereby regulating cell cycle. ¹² We further found that circGSAP alleviated PMEC dysfunction via miR‐27a‐3p/bone morphogenetic protein receptor type 2 (BMPR2) axis. ¹³ Studies reported that circ_0010729 was a key regulator of vascular endothelial cell proliferation and apoptosis and functions via the miR‐186/hypoxia inducible factor 1 subunit alpha (HIF‐1α) axis. ¹⁴ Zhang et al. demonstrated that circ‐calm4 acts as a miR‐337‐3p sponge to regulate myosin 10 (myo10) and promote PASMC proliferation. ⁹ Circ‐calm4 could also regulate hypoxic PASMC pyroptosis via the circ‐calm4/miR‐124‐3p/programmed cell death 6 (PDCD6) axis. ¹⁰ Hsa_circ_0016070 was associated with vascular remodeling in PH by promoting proliferation of PASMCs via the miR‐942/cyclin D1 (CCND1) axis. ¹⁵ CircZNF532 was reported to regulate diabetes‐induced retinal pericyte degeneration and vascular dysfunction by sponging miR‐29a‐3p. ¹⁶ Undoubtedly, circRNAs can regulate the expressions of target genes, ultimately affecting the cellular phenotype of PMECs and PASMCs in PH. However, the similarities and differences in circRNA expression and the potential mechanisms of circRNAs in PASMCs, PMECs, and PCs under hypoxic conditions are unclear.

In this study, we performed whole transcriptome sequencing and revealed for the first time the circRNA expression profiles of three vascular cells on exposure to hypoxia. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were used to analyze the putative functions of the differentially expressed circRNAs. Among the identified circRNAs, circular postmeiotic segregation 1 (circPMS1) was the only co‐differentially expressed circRNA and exhibited increased expression in PASMCs, PMECs, and PCs after exposure to hypoxia. After the impact of co‐differentially expressed circPMS1 on PASMCs, PMECs, and PCs was investigated, we elucidated the underlying sponge mechanisms of circPMS1 in regulating the proliferation of these three different vascular cells under hypoxia.

2. MATERIALS AND METHODS

2.1. Cell culture

Human PASMCs (ScienCell) were cultured in smooth muscle cell medium (ScienCell, category number: 1101). Human PMECs (ScienCell) were cultured in endothelial cell medium (ScienCell, category number: 1001). Human PCs (ScienCell) were cultured in pericyte medium (ScienCell, category number: 1201). All basic culture mediums were supplemented with 10% fetal bovine serum (Gibco), 100 U/mL of penicillin, and 100 μg/mL of streptomycin. The cells were incubated at 37°C with 5% carbon dioxide (CO₂). For hypoxia exposure, PASMCs, PMECs, and PCs were cultured using the same medium as under normoxic conditions and exposed to hypoxic conditions (5% oxygen and 5% CO₂) for 24 h.

2.2. Whole transcriptome sequencing

Total RNA was prepared from normoxic and hypoxic PASMCs, PMECs, and PCs using TRIzol reagent (Invitrogen, 15596018). After ribosomal RNA was removed and libraries were constructed, whole transcriptome sequencing was performed by Shanghai OE Biotech Co.

2.3. Genomic DNA extraction

A genomic DNA (gDNA) isolation kit (Sangon Biotech, Shanghai, China) was used to extract gDNA according to the manufacturer's instructions.

2.4. Identification of differentially expressed circRNAs

Differentially expressed circRNAs from normoxic and hypoxic PASMCs or PCs were screened using |log2FC| > 1 and p‐value < 0.1. Differentially expressed circRNAs from normoxic and hypoxic PMECs were screened using |log2FC| > 2 and p‐value < 0.05.

2.5. GO and KEGG pathway analyses

The Database for Annotation, Visualization and Integrated Discovery (https://david.ncifcrf.gov/) (version 6.8) was used to perform GO and KEGG pathway analyses to identify the potential functions of linear transcripts of the differentially expressed circRNAs from three paired vascular cells. A p‐value < 0.05 was considered statistically significant.

2.6. Quantitative real‐time polymerase chain reaction

Total RNA of normoxic or hypoxic PASMCs, PMECs, and PCs was extracted using TRIzol reagent (Invitrogen, 15596018). For each sample, ReverTra Ace qPCR RT kit (FSQ‐101, TOYOBO) was used to convert 1 μg of total RNA into complementary DNA (cDNA). Quantitative real‐time polymerase chain reaction (qPCR) was performed using 2 × TOROGreen qPCR Master Mix (QST‐100, TOROIVD) according to the manufacturer's protocol. Glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) was used as an internal control for circRNA/messenger RNA (mRNA) quantification, and U6 was used as an internal control for microRNA (miRNA) quantification. The relative expression levels were calculated using the 2^−ΔΔCT method. The primers are presented in Table 1.

TABLE 1.

Primers and siRNA used in this study.

Primers
circPMS1	F: TCCTCATGAGCTTTGGTATCCT
	R: AAAGGAGTCGAACTGTTGCC
GAPDH	F: TCGTGGAAGGACTCATGACC
	R: ATGATGTTCTGGAGAGCCCC
U6	F: CTCGCTTCGGCAGCACA
	R: AACGCTTCACGAATTTGCGT
miR‐432‐3p	F: ACACTCCAGCTGGGTCTTGGAGTAGGTCA
	R: GTGCAGGGTCCGAGGT
miR‐433‐3p	F: ACACTCCAGCTGGGATCATGATGGGCTC
	R: GTGCAGGGTCCGAGGT
miR‐3613‐5p	F: ACACTCCAGCTGGGTGTTGTACTTTTTTTT
	R: GTGCAGGGTCCGAGGT
DEPDC1	F: TGCAGTGGAAAAACATCTTGA
	R: GCTCATCAAACTCCTGAGCA
FANCC	F: TGTGGCTCTTGGCCTTCTAC
	R: CTGCTACCGTCTGCAGGTC
POLR2D	F: TTGCCAGTGTTCGTAGCTTG
	R: TGCAGCTCCTCATCTTCAAA
MXI1	F: CTCAACAAAGCCAAAGCACA
	R: ACCCTGCAGCTGTTCCAGT
RSBN1	F: AGGTCCCAGAGTGATGATGG
	R: TGGTCCGAGGTAGGTACTGG
ZFAND5	F: CTCAGCCCAGTCCATCAGTT
	R: AAATTTCCACATCGGCAGTC
siRNA sequences
si‐circPMS1 #1	Sense (5′‐3′) UUUGUACAUAACAAGCUGCTT Antisense (5′‐3′) GCAGCUUGUUAUGUACAAATT
si‐circPMS1 #2	Sense (5′‐3′) UAACAAGCUGCUCUGUUAATT Antisense (5′‐3′) UUAACAGAGCAGCUUGUUATT

Abbreviations: CircPMS1, circular postmeiotic segregation 1; DEPDC1, DEP domain containing 1; FANCC, FA complementation group C; GAPDH, glyceraldehyde‐3‐phosphate dehydrogenase; MXI1, MAX interactor 1; POLR2D, RNA polymerase II subunit D; RSBN1, round spermatid basic protein 1; siRNA, small‐interfering RNA; ZFAND5, zinc finger AN1‐type containing 5.

2.7. Small‐interfering RNA transfection

Two small‐interfering RNAs (siRNAs) targeting circPMS1 were designed and synthesized by GenePharma (Shanghai, China). For circPMS1 silencing, PASMCs, PMECs, and PCs were transfected with 80 nM siRNA using Lipo2000 (11668019, Invitrogen, USA) in serum‐free medium. Serum‐free medium was replaced with serum‐containing medium after 6 h, and cells were cultured for another 24 h. The siRNA sequences are presented in Table 1.

2.8. Cell proliferation assay

PASMCs, PMEC, and PCs were seeded into 96‐well cell culture plates at a density of 2000 cells/well after transfection. A Cell Counting Kit‐8 (CCK8, CK04, Dojindo, Japan) was used to measure cell proliferation over a range of time points. Briefly, 10 μL of CCK8 was added to each well and incubated at 37°C for 2 h. The absorbance (optical density) value was then measured at 450 nm using a microplate reader.

2.9. EdU cell proliferation kit

PASMCs, PMEC, and PCs were seeded into 96‐well culture plates at a density of 5000 cells/well after transfection. EdU Cell Proliferation kit with Alexa Fluor 488 (CX002, Epizyme, China) was used to measure cell proliferation after 48 h. Briefly, 10 μM EdU was added to each well, and the cells were incubated at 37°C for 2 h. The cells were then fixed with 4% paraformaldehyde for 15 min (min) and the treated with 0.3% Triton X‐100 for 10 min. Subsequently, 50 μL of reaction cocktail was added to each well and incubated for 30 min, and 1× Hoechst 33342 was used to stain the cell nuclei for 10 min. Images were captured using a microscope. The percentage of EdU‐positive cells was defined as the proliferation rate.

2.10. ceRNA network analysis

CircInteractome (https://circinteractome.nia.nih.gov/) was used to predict the potential target miRNAs for the selected circRNAs. In addition, miRDB (http://mirdb.org/), TargetScan (https://www.targetscan.org/vert_80/), and miRWalk (http://mirwalk.umm.uni‐heidelberg.de/) were used to predict the target mRNAs of the selected miRNAs.

2.11. Statistical analysis

For qPCR analysis, the results are presented as mean ± standard error. The statistical difference between two independent groups was analyzed using an unpaired Student's t‐test. A one‐way analysis of variance was used to compare multiple groups. p‐Values < 0.05 were considered statistically significant. GraphPad Prism (version 9.0.1) was used to plot all graphs.

3. RESULTS

3.1. Differentially expressed circRNAs of PASMCs, PMECs, and PCs

Based on the screening criteria, 16 circRNAs were differentially expressed in PASMCs exposed to normoxic or hypoxic conditions. Of these, 10 circRNAs were highly expressed in hypoxia‐induced PASMCs, whereas 6 were downregulated. In total, 99 circRNAs in PMECs under hypoxic conditions showed differential expression, of which 69 were markedly upregulated and 30 exhibited decreased expression. In addition, 31 circRNAs, comprising 18 upregulated and 13 downregulated circRNAs, showed differential expression between hypoxic and normoxic PCs. A volcano plot and clustered heat map of the differentially expressed circRNAs in PASMCs, PMECs, and PCs are shown in Figure 1.

FIGURE 1.

Overview of circRNA (circular RNA) profiles in vascular cells. (A, C, E) Volcano plot of differentially expressed circRNAs in PASMCs (pulmonary artery smooth muscle cells), PMECs (pulmonary microvascular endothelial cells), and PCs (pericytes). (B, D, F) Clustered heat map of differentially expressed circRNAs in PASMCs, PMECs, and PCs.

3.2. Categories of differentially expressed circRNAs in PASMCs, PMECs, and PCs

The differentially expressed circRNAs in PASMCs, PMECs, and PCs were classified into diverse categories, which are shown in Figure 2A–F. In PASMCs, nine dysregulated circRNAs were derived from the sense strand, comprising seven sense‐overlapping, one intergenic, and one intronic circRNAs. There were seven antisense circRNAs in PASMCs, comprising three sense‐overlapping, three exonic, and one intergenic. The 99 dysregulated circRNAs in PMECs comprised 50 sense and 49 antisense circRNAs. The sense circRNAs comprised 45 sense‐overlapping, 2 exonic, and 3 intergenic circRNAs. The antisense circRNAs comprised 43 sense‐overlapping, 2 exonic, and 4 intergenic circRNAs. In the PCs, 15 differentially expressed circRNAs were from the sense strand, of which 14 circRNAs were sense‐overlapping and 1 was exonic. Moreover, 16 antisense circRNAs consisted of 14 sense‐overlapping and 2 exonic circRNAs.

FIGURE 2.

Characteristics of differentially expressed circRNAs (circular RNAs) in PASMCs (pulmonary artery smooth muscle cells), PMECs (pulmonary microvascular endothelial cells), and PCs (pericytes). (A–C) Categories of sense circRNAs in PASMCs, PMECs, and PCs. (D–F) Categories of antisense circRNAs in PASMCs, PMECs, and PCs. (G) GO (Gene Ontology) analysis of differentially expressed circRNAs in PASMCs, PMECs, and PCs. (H) KEGG (Kyoto Encyclopedia of Genes and Genomes) analysis of differentially expressed circRNAs in PASMCs, PMECs, and PCs.

3.3. Biological analyses of the linear transcripts of differentially expressed circRNAs in PASMCs, PMECs, and PCs

Based on the assumption that the biological functions of circRNAs are associated with the functions of their linear transcripts, GO and KEGG pathway analyses were used to determine the putative function of the differentially expressed circRNAs (Figure 2G,H). GO analysis revealed that the biological process of dysregulated circRNAs in PASMCs, PMECs, and PCs was mainly enriched in protein transport, endocytosis, cilium assembly, ubiquitin‐dependent protein catabolic process, and antigen processing and presentation processes. The cell components of the dysregulated circRNAs were associated with the cytosol, membrane, nucleoplasm, phagocytic vesicle membrane, and cytoplasm. The molecular process of the dysregulated circRNAs was significantly enriched in ATPase activity, ATP binding, ubiquitin‐protein transferase activity, histone binding, and small GTPase binding. KEGG pathway analysis suggested that the dysregulated circRNAs in PASMCs, PMECs, and PCs were associated with endocytosis, various types of N‐glycan biosynthesis, N‐glycan biosynthesis, viral myocarditis, and lysine degradation.

3.4. Identification of common differentially expressed circRNAs in PASMCs, PMECs, and PCs

The overlapping of the differentially expressed circRNAs in PASMCs, PMECs, and PCs showed that two circRNAs were dysregulated in both PMECs and PCs and two circRNAs were differentially expressed in both PASMCs and PCs, whereas only one circRNA was differentially expressed in both PASMCs and PMECs. Of note, circPMS1 (hsa_circ_0001083) is a common differentially expressed circRNA in PASMCs, PMECs, and PCs (Figure 3A), and it is located on human chromosome 2:190656515–190682906+ with a length of 602 nucleotides. The parental gene of circPMS1 is a mismatch repair (MMR) gene, postmeiotic segregation 1 (PMS1), which was backspliced at exons 2–5. The backsplicing junction of circPMS1 was validated using Sanger sequencing (Figure 3B).

FIGURE 3.

Common dysregulated circRNAs (circular RNAs) in PASMCs (pulmonary artery smooth muscle cells), PMECs (pulmonary microvascular endothelial cells), and PCs (pericytes). (A) Venn diagram of differentially expressed circRNAs in PASMCs, PMECs, and PCs. (B) Schema illustrating the production of circPMS1 (circular postmeiotic segregation 1). Sanger sequencing was used to validate the backsplicing junction of circPMS1. (C–E) Validation of circPMS1 expression in normoxic and hypoxic PASMCs, PMECs, and PCs (n = 3). (F–H) PCR (polymerase chain reaction) analysis of circPMS1 and linear GAPDH in cDNA (complementary DNA) and gDNA (genomic DNA). *p < 0.05.

Based on the sequencing analysis, circPMS1 was unregulated in hypoxic PASMCs and PMECs and downregulated in hypoxic PCs. However, qPCR results indicated that circPMS1 was increased in hypoxic PASMCs, PMECs, and PCs (Figure 3C–E). To further validate the sequence of circPMS1, we designed divergent and convergent primers to amplify circular and linear PMS1 products in cDNA and gDNA of PASMCs, PMECs, and PCs. Of note, circPMS1 was amplified using divergent primers in cDNA of PASMCs, whereas no amplification was observed in gDNA samples. However, the linear product of PMS1 was amplified in both cDNA and gDNA of PASMCs (Figure 3F). The same results were observed using cDNA and gDNA of PMECs and PCs as templates (Figure 3G,H).

3.5. CircPMS1 promotes cell proliferation in PASMCs, PMECs, and PCs

Because circPMS1 was overexpressed in hypoxia‐induced PASMCs, PMECs, and PCs, we designed an siRNA targeting circPMS1 to further study the potential function of circPMS1 in these cells under normoxic and hypoxic conditions. As shown in Figure 4A, circPMS1 was markedly downregulated in PASMCs after siRNA circPMS1 transfection under both normoxic and hypoxic conditions. Downregulation of circPMS1 reduced the proliferation of PASMCs under both normoxic and hypoxic conditions as assessed by EdU labeling and CCK8 assay (Figure 4B,G).

FIGURE 4.

Effect of circPMS1 (circular postmeiotic segregation 1) on the proliferation of PASMCs (pulmonary artery smooth muscle cells), PMECs (pulmonary microvascular endothelial cells), and PCs (pericytes). (A) Expression of circPMS1 in PASMCs after transfection of siRNA (small‐interfering RNA) circPMS1 under both normoxic and hypoxic conditions (n = 3). (B) EdU analysis of PASMCs under both normoxic and hypoxic conditions (n = 4). (C) Expression of circPMS1 in PMECs after transfection of siRNA circPMS1 under both normoxic and hypoxic conditions (n = 3). (D) EdU analysis of PMECs under both normoxic and hypoxic conditions (n = 4). (E) Expression of circPMS1 in PCs after transfection of siRNA circPMS1 under both normoxic and hypoxic conditions (n = 3). (F) EdU analysis of PCs under both normoxic and hypoxic conditions (n = 4). (G) CCK8 (Cell Counting Kit‐8) assay of PASMCs under both normoxic and hypoxic conditions (n = 5). (H) CCK8 assay of PMECs under both normoxic and hypoxic conditions (n = 5). (I) CCK8 assay of PCs under both normoxic and hypoxic conditions (n = 5). Scale bar: 50 μm. *p < 0.05, **p < 0.01, and ***p < 0.001.

In PMECs, transfection of siRNA circPMS1 knocked down the expression of circPMS1 significantly (Figure 4C). The low expression of circPMS1 reduced the proliferation of PMECs exposed to normoxia and hypoxia, as determined using EdU labeling and CCK8 assay (Figure 4D,H).

In PCs, transfection of siRNA circPMS1 efficiently reduced the expression of circPMS1 (Figure 4E). The EdU labeling and CCK8 assay results indicated that downregulated expression of circPMS1 reduced the proliferation of PCs under both normoxic and hypoxic conditions (Figure 4F,I).

3.6. Annotation of the circPMS1–miRNA–mRNA interaction

We predicted some potential miRNAs that may be sponged by circPMS1 (Table 2; Figure 5A). Of these, microRNA‐432‐5p (miR‐432‐5p) was downregulated in PASMCs isolated from idiopathic pulmonary arterial hypertension (IPAH) patients (GSE55427) (Figure 5B), whereas miR‐433‐3p was downregulated in tumor necrosis factor alpha (TNFα)–induced human umbilical vein endothelial cells (GSE92655) (Figure 5C). Furthermore, miR‐3613‐5p was upregulated in our previously established miRNA expression profile of PCs exposed to hypoxia (Figure 5D). Furthermore, qPCR results confirmed that the expression of miR‐432‐5p, miR‐433‐3p, and miR‐3613‐5p was decreased in hypoxic PASMCs, PMECs, and PCs, respectively (Figure 5B–D).

TABLE 2.

The predicted targeted miRNAs of circPMS1.

miRNA name	CircPMS1 (top) (5′‐3′)
	miRNA (bottom) pairing (3′‐5′)
miR‐1278	CUGUAAUGGCAAUGA AGUACU AC UAUCUACUAUACGUG UCAUGA U
miR‐1304	CAAUGAAGUACUACA CCUCAA AA GUGUAGAGUGACAUC GGAGUU U
miR‐1305	UUGUAAAAGAGCUUA UUGAAA AC AGAGAGGGUAAUCUC AACUUU U
miR‐145	CAAGCGUAGAUGUUA AACUGGA G UCCCUAAGGACCCUU UUGACCU G
miR‐224‐5p	UGGUAUCCUUAAACC UGACUU AA UUGCCUUGGUGAUC ACUGAA C
miR‐338‐3p	UUGAAAACUCCUUGGAUGCUGGU GUUGUUUUAGUGAC——UACGACCU
miR‐382	GAUCUUGAAAAUUUGACAACUUA GCUUAGGUGGUGCUUGUUGAAG
miR‐421	GGGAGGGUAUCAAGGCUGUUGAU CGCGGGUUAAUUACAGACAACUA
miR‐432‐5p	AUGAAAUAAAAAAGAUCCAAGAU GGUGGGUUACUGGAUGAGGUUCU
miR‐433‐3p	CUCAAAAAUAAAUAGUCAUGAAG UGUGGCUCCUCGGGUAGUACUA
miR‐488‐3p	GGCAACAGUUCGACUCCUUUCAA CUGGUUCUUUAUCGGAAAGUU
miR‐498	AAUAGUCAUGAAGAUCUUGAAAA CUUUUUGCGGGGGACCGAACUUU
miR‐507	GCAGUUUUACUCAACUGCAAAAA AAGUGAGGUUUUCCACGUUUU
miR‐557	GCAGUUUUACUCAACUGCAAAAA UCUGUUCCGGGUGGGCACGUUUG
miR‐515‐5p	GCGUAGAUGUUAAACUGGAGAAC GUCUUUCACGAAAGAAAACCUCUU
miR‐548I	GAUGGCAGUGGCCACAUACUUUC CUGUUUUGGGCGUUUAUGAAAA
miR‐576‐3p	UUUCUCAGAAACCUUCACAUCUU CUAAGGUUAAAAAGGUGUAGAA
miR‐578	GAGGUUUUAAUUACAACAAGAAC UGUUAGGAUCUCGUGUUCUUC
miR‐579	CUCUGUUAAAAGCGAAAAUGAAA UUAGCGCCAAAUAUGG—UUUACUU
miR‐589‐5p	UUCGACUCCUUUCAAGUUCUCAG GAGUCUCGUCUGCACCAAGAGU
miR‐600	UAUUUAAGAAUCUACCUGUAAGA CUCGUUCCGAGAACAGACAUUCA
miR‐634	UGAAAACUCCUUGGA–UGCUGGUG CAGGUUUCAACCCCACGACCAA
miR‐663b	GUUUUAGAUGGCAGUGGCCACAU GGAGUCCGUGCCGGCCCGGUGG
miR‐766	AGCGUAGAUGUUAAACUGGAGAA CGACUCCGACACCCCGACCUCA
miR‐885‐5p	GCUGUUGAUGCACCUGUAAUGGC UCUCCGUCCCAUCACAUUACCU
miR‐3613‐5p	GCUGAGGUUUUAAUUACAACA CUUGUUUUUUUUUUCAUGUUGU

Note: The bases highlighted in red represent interaction sites.

Abbreviations: CircPMS1, circular postmeiotic segregation 1; miRNA, microRNA.

FIGURE 5.

The predicted targeted miRNAs (microRNA) of circPMS1 (circular postmeiotic segregation 1). (A) The predicted circRNA–miRNA (microRNA) network of circPMS1. (B) Relative expression of miR‐432‐5p (microRNA‐432‐5p) in GSE55427 (left) and validation of miR‐432‐5p expression in normoxic and hypoxic PASMCs (pulmonary artery smooth muscle cells) (right) (n = 3). (C) Relative expression of miR‐433‐3p in GSE92655 (left) and validation of miR‐433‐3p expression in normoxic and hypoxic PMECs (right) (n = 3). (D) The relative expression of miR‐3613‐5p in the miRNA expression profile of PCs exposed to hypoxia (left) and validation of miR‐3613‐5p expression in normoxic and hypoxic PCs (right) (n = 3). *p < 0.05, **p < 0.01, and ***p < 0.001.

We also predicted the target mRNAs of miR‐432‐5p, miR‐433‐3p, and miR‐3613‐5p using miRDB, TargetScan, and miRWalk, respectively (Figure 6A–C). Of these, DEP domain containing 1 (DEPDC1), FA complementation group C (FANCC), and RNA polymerase II subunit D (POLR2D) were upregulated in PASMCs isolated from IPAH patients (GSE144274) (Figure 6D–F). MAX interactor 1 (MXI1) and round spermatid basic protein 1 (RSBN1) were upregulated in our previously established mRNA expression profile of PMECs exposed to hypoxia (Figure 6G,H). Zinc finger AN1‐type containing 5 (ZFAND5) was upregulated in lung tissues from PH patients (Figure 6I); however, qPCR indicated that DEPDC1 and POLR2D were upregulated in hypoxic PASMCs, whereas FANCC showed no significant difference between PASMCs under normal or hypoxic conditions (Figure 6J–L). In addition, qPCR indicated that MXI1 was upregulated in hypoxic PMECs, whereas RSBN1 exhibited no significant difference between PMECs under normal or hypoxic conditions (Figure 6M,N). The validation of ZFAND5 expression in normoxic and hypoxic PCs was consistent with that of the sequencing results (Figure 6O). Finally, we hypothesized a potential mechanism of circPMS1 in vascular cells, which is shown in Figure 7.

FIGURE 6.

The predicted mRNA (messenger RNA) expression. (A–C) miRDB, TargetScan, and miRWalk bioinformatics software predicts miR‐432‐5p (microRNA‐432‐5p), miR‐433‐3p, and miR‐3613‐5p target genes. (D–F) Relative expression of DEPDC1 (DEP domain containing 1), FANCC (FA complementation group C), and POLR2D (RNA polymerase II subunit D) in GSE144274 (n = 4). (G, H) Relative expression of MXI1 (MAX interactor 1) and RSBN1 (round spermatid basic protein 1) in the mRNA expression profile of PMECs (pulmonary microvascular endothelial cells) exposed to hypoxia (n = 4). (I) Relative expression of ZFAND5 (zinc finger AN1‐type containing 5) in GSE113439 (n = 11 or 15). (J–L) Validation of DEPDC1, FANCC, and POLR2D expression in normoxic and hypoxic PASMCs (pulmonary artery smooth muscle cells) (n = 3). (M, N) Validation of MXI1 and RSBN1 expression in normoxic and hypoxic PMECs (n = 3). (O) Validation of ZFAND5 expression in normoxic and hypoxic PCs (pericytes) (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

FIGURE 7.

Predicted mechanism of circPMS1 (circular postmeiotic segregation 1) in vascular cells.

4. DISCUSSION

Although there is already substantial evidence suggesting that circRNAs can contribute to the pathogenesis of PH by regulating the function of PASMCs or PMECs, this is the first study to establish circRNA expression profiles in vascular cells (PASMCs, PMECs, and PCs) under hypoxia, as well as to identify co‐differentially expressed circPMS1 that impacts the proliferation of these three vascular cells under hypoxia. Therefore, circPMS1 may be relevant to PH. The data indicated that circPMS1 was upregulated in hypoxic PASMCs and PMECs as well as PCs. Moreover, our results suggested that circPMS1 could potentially enhance cell proliferation in PASMCs via the miR‐432‐5p/DEPDC1 axis or miR‐432‐5p/POLR2D axis, in PMECs via the miR‐433‐3p/MXI1 axis, and in PCs via the miR‐3613‐5p/ZFAND5 axis, providing new evidence that circRNAs serve as significant mediators in regulating vascular remodeling under pathophysiological conditions of PH.

Recently, it was reported that noncoding RNAs (ncRNAs), including miRNAs, circRNAs, and long‐chain noncoding RNAs (lncRNAs), play a significant role in the pathogenesis of PH. With the development of high‐throughput sequencing technology, the functions of these ncRNAs in PH are being discovered. In a previous study, we analyzed the differential expression of circRNAs in peripheral blood mononuclear cells of IPAH and found that circGSAP levels were related to the occurrence and poor outcome of IPAH and may serve as a biomarker for IPAH. ¹¹ We also previously established the expression profiles of lncRNAs in PASMCs, PMECs, and PCs under hypoxic conditions and found that lncRNA taurine upregulated 1 (lncRNA TUG1) was upregulated in hypoxic PASMCs and PCs but downregulated in PMECs. ¹⁷ In the present study, we found 16, 99, and 31 differentially expressed circRNAs in PASMCs, PMECs, and PCs, respectively; however, none of these circRNAs have been reported in other PH‐related studies, which may be due to the differences in the samples that were sequenced. Of note, some of these differentially expressed circRNAs have been studied in cancer, ¹⁸ , ¹⁹ , ²⁰ which provides the foundation for the study of these circRNAs in PH. For example, hsa_circ_0006528 promotes human breast cancer progression by sponging miR‐7‐5p and activating the mitogen‐activated protein kinase (MAPK)/extracellular signal‐regulated kinase (ERK) signaling pathway. ¹⁸ Hsa_circ_0001801 functions as a sponge for miR‐224‐5p to promote glioma progression. ¹⁹ Furthermore, hsa_circ_0067997 promotes the progression of gastric cancer by inhibiting miR‐525‐5p and activating X chromosome–linked inhibitor of apoptosis. ²⁰ Further studies are needed to fully understand differentially expressed circRNAs in PH.

We also performed biological analyses to gain insight into the potential function of the differentially expressed circRNAs. GO analysis suggested that the circRNAs were significantly related to protein transport and cytosol and ATPase activity. KEGG pathway analysis indicated that the circRNAs were enriched in endocytosis, various types of N‐glycan biosynthesis, and N‐glycan biosynthesis.

Of all the differentially expressed circRNAs in PASMCs, PMECs, and PCs, circPMS1 exhibited a common differential expression pattern. CircPMS1 was found to be upregulated in the plasma of non‐small cell lung cancer (NSCLC) patients. ²¹ In the present study, circPMS1 was increased in hypoxic PASMCs, PMECs, and PCs. Using loss‐of‐function approaches, we found that circPMS1 enhanced the proliferation of PASMCs, PMECs, and PCs under hypoxic and normoxic conditions.

The parental gene of circPMS1 is PMS1, which is an MMR gene. MMR is a significant DNA repair mechanism that corrects mismatches occurring during DNA replication as well as apoptosis signaling in response to DNA damage. ²² , ²³ Previous studies have indicated that genetic variations in the MMR genes were attributed to survival and chemotherapeutic drug toxicity in various cancers. ²⁴ , ²⁵ PMS1 may interact with mutL homolog 1 to form a complex that functions as a matchmaker to signal mismatch recognition to downstream repair enzymes and contributes to this mutation avoidance pathway. ²⁶ In addition, PMS1 was downregulated in oral squamous cell carcinoma. ²⁷ In NSCLC, the rs5742933 variation in PMS1 may represent a candidate prognostic marker for clinical outcome. ²⁸ PMS1 polymorphisms may also affect the occurrence of dermatitis and prognosis in rectal cancer patients. ²⁹

CircRNAs may act as sponges to bind miRNAs, thereby promoting downstream mRNA expression. ³⁰ , ³¹ Previous studies have also reported the sponge mechanism of circRNAs in PH. ⁹ , ¹⁰ , ¹¹ Therefore, we predicted the downstream miRNAs of circPMS1 and the potential miRNA binding targets, including miR‐432‐5p, miR‐433‐3p, and miR‐3613‐5p. miR‐432‐5p was downregulated in PASMCs isolated from IPAH patients (GSE55427) and hypoxic PASMCs. miR‐432‐5p was reported to participate in cancer development. For example, knockdown of circular cysteine‐rich transmembrane BMP regulator 1 inhibits histone deacetylase 4 (HDAC4) to impede osteosarcoma proliferation, migration, and invasion and facilitate autophagy by targeting miR‐432‐5p. ³² miR‐433‐3p was downregulated in TNFα‐induced human umbilical vein endothelial cells (GSE92655) and hypoxic PMECs. miR‐433‐3p has also been widely studied in cancer. For example, circ_0011292 knockdown targeted miR‐433‐3p/checkpoint kinase 1 (CHEK1) axis to ameliorate progression and drug resistance in paclitaxel‐resistant NSCLC cells. ³³ Further studies are needed to determine the role of miR‐432‐5p and miR‐433‐3p in PH. In addition, miR‐3613‐5p was downregulated in our previous miRNA expression profile of PCs exposed to hypoxia. We further validated that miR‐3613‐5p was decreased in hypoxic PCs. miR‐3613‐5p was shown to be a tumor suppressor in pancreatic cancer, and patients with a higher lymph node metastasis rate showed reduced miR‐3613‐5p expression. Furthermore, miR‐3613‐5p downregulated cyclin‐dependent kinase 6 (CDK6) in repressing the metastatic capacity of pancreatic cancer cells in vitro and in vivo. ³⁴

Interestingly, miR‐145 is also a predicted target miRNA of circPMS1; however, miR‐145 was upregulated in PASMCs of PH mice exposed to hypoxia. ³⁵ The decreased expression of miR‐145 inhibited the development of PH. ³⁵ , ³⁶ Yue et al. demonstrated that miR‐145 promotes hypoxia‐induced proliferation and migration of PASMCs by regulating ATP binding cassette subfamily A member 1 (ABCA1) expression. ³⁷ This may occur because miRNAs may target multiple circRNAs.

There are studies on the crosstalk among the three cell types. ³⁸ , ³⁹ A previous study has reported that exosomes can carry functional miRNAs between cells. ³⁸ PASMCs can transfer miR‐143‐3p via exosomes to endothelial cell, thereby promoting the migration of endothelial cells. ³⁹ However, it remains to be further studied whether these three miRNAs (miR‐432‐5p, miR‐433‐3p, and miR‐3613‐5p) participate in the crosstalk among these three cell types.

We also predicted the downstream mRNAs targeted by miR‐432‐5p, miR‐433‐3p, and miR3613‐5p. DEPDC1 and POLR2D were upregulated in PASCMs exposed to hypoxia, which is consistent with the expression in PASMCs isolated from IPAH patients (GSE144274). DEPDC1 is highly expressed in lung adenocarcinoma and promotes tumor cell proliferation. ⁴⁰ DEPDC1 promotes hepatocellular carcinoma migration and invasion through the Wnt/β‐catenin signaling pathway and epithelial–mesenchymal transition. ⁴¹ POLR2D is indispensable for vertebrate development ⁴² and is upregulated in various cancers, such as prostate cancer ⁴³ and NSCLC. ⁴⁴ However, FANCC exhibited no difference between normoxic and hypoxic PASMCs. MXI1 was upregulated in PMECs exposed to hypoxia, which is consistent with the results of our previous mRNA expression profile of PMECs exposed to hypoxia. However, RSBN1 showed no difference between normoxic and hypoxic PMECs. Dong et al. demonstrated that MXI1 was significantly upregulated in PASMCs from hypoxic PH patients, and it promoted hypoxic PH through ERK/c‐Myc‐dependent proliferation of arterial smooth muscle cells. ⁴⁵ In addition, we demonstrated that ZFAND5 is upregulated in PCs exposed to hypoxia, which is consistent with our previous mRNA expression profile of PCs exposed to hypoxia. ZFAND5 was reported to be upregulated in perihilar cholangiocarcinoma and is an independent prognostic biomarker of perihilar cholangiocarcinoma. ⁴⁶

There are several limitations in our current study. First, the differential expression of circPMS1 has not been examined in patients with PH, nor has the biomarker function of circPMS1 been investigated for its potential use in the pathogenesis, diagnosis, and prognosis of PH. Second, we have not explored whether reducing circPMS1 could alleviate the progression of hypoxia‐induced PH. Third, although we validated the expression of circPMS1 and its downstream miRNAs and mRNAs in different vascular cells exposed to hypoxia, we have not validated the direct relationship between circPMS1 and miRNAs as well as miRNAs and mRNAs. Furthermore, we have confirmed only the impact of circPMS1 on cell proliferation; we have yet to determine whether the chosen miRNAs and mRNAs have any effect on cell proliferation. Additionally, although we observed the effect of circPMS1 on proliferation of PASMCs, PMECs, and PCs, it warrants further investigation on whether circPMS1 has an effect on other cell phenotypes, such as migration, apoptosis, and angiogenesis. Finally, we identified only the molecular mechanism of circPMS1 as a sponge to regulate the functions of different vascular cells; other molecular mechanisms, like binding RNA‐binding proteins, also need to be studied in vivo and in vitro to better understand the pathogenesis of PH.

5. CONCLUSION

The present study revealed a co‐differentially expressed circPMS1, which was highly expressed in all three vascular cells (PASMCs, PMECs, and PCs) under hypoxia. Furthermore, the data indicated that circPMS1 could serve as an miRNA sponge to promote cell proliferation in these three vascular cells. Further investigation is needed to fully understand the function of circPMS1 in PH with hypoxia signaling, as it has the potential to be a valuable therapeutic target or clinical marker.

AUTHOR CONTRIBUTIONS

Jinming Liu, Ping Yuan, and Xiaoyi Hu contributed to the conception and design of the study. Xiaoyi Hu, Shang Wang, and Hui Zhao performed statistical analysis. Xiaoyi Hu, Yaqin Wei, and Ruowang Duan wrote the first draft of the manuscript. Rong Jiang, Wenhui Wu, Qinhua Zhao, Sugang Gong, and Lan Wang wrote sections of the manuscript. All authors contributed to manuscript revision.

FUNDING INFORMATION

This work was supported by funding from the Program of Natural Science Foundation of Shanghai (21ZR1453800 and 201409004100), Fundamental Research Funds for the Central Universities (22120220562), Program of National Natural Science Foundation of China (81870044), and Program of Shanghai Pulmonary Hospital (FKLY20005 and fkzr2320).

ETHICS STATEMENT

This article did not involve human or animal experiments, there are no ethical issues.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Hu X, Wang S, Zhao H, et al. CircPMS1 promotes proliferation of pulmonary artery smooth muscle cells, pulmonary microvascular endothelial cells, and pericytes under hypoxia. Anim Models Exp Med. 2024;7:310‐323. doi: 10.1002/ame2.12332

DATA AVAILABILITY STATEMENT

The raw data supporting the conclusions of this article will be made available by the corresponding authors without undue reservation.