CircTM7SF3 contributes to oxidized low-density lipoprotein-induced apoptosis, inflammation and oxidative stress through targeting miR-206/ASPH axis in atherosclerosis cell model in vitro

Background Atherosclerosis (AS) is a chronic inflammatory disorder. The aim of our study was to explore the role of circular RNA (circRNA) transmembrane 7 superfamily member 3 (circTM7SF3) in AS progression. Methods Experiments were conducted using oxidized low-density lipoprotein (ox-LDL)-induced THP-1-derived macrophages and differentiated human monocyte-derived macrophages (hMDMs). Quantitative real-time polymerase chain reaction (qRT-PCR) was used to detect the expression of circTM7SF3, its linear form TM7SF3, microRNA-206 (miR-206) and aspartyl (asparaginyl) β-hydroxylase (ASPH) messenger RNA (mRNA). Cell viability and apoptosis were examined by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay and flow cytometry. Cell inflammation was analyzed by measuring the production of tumor necrosis factor α (TNF-α) and interleukin 6 (IL-6) using enzyme-linked immunosorbent assay (ELISA) kits. Cell oxidative stress was assessed through analyzing the levels of oxidative stress markers using their corresponding commercial kits. Dual-luciferase reporter assay and RNA-pull down assay were used to confirm the interaction between miR-206 and circTM7SF3 or ASPH. The protein level of ASPH was examined by Western blot assay. Results CircTM7SF3 level was markedly increased in the serum samples of AS patients and ox-LDL-induced THP-1-derived macrophages compared with their matching counterparts. ox-LDL induced-damage in THP-1 cells was partly attenuated by the interference of circTM7SF3. MiR-206 was a downstream molecular target of circTM7SF3. Si-circTM7SF3-mediated effects in ox-LDL-induced THP-1-derived macrophages were partly ameliorated by the addition of anti-miR-206. MiR-206 directly interacted with ASPH mRNA. CircTM7SF3 silencing reduced the expression of ASPH partly through up-regulating miR-206 in THP-1-derived macrophages. ASPH overexpression partly counteracted the effects induced by miR-206 overexpression in ox-LDL-induced THP-1-derived macrophages. Conclusion CircTM7SF3 contributed to ox-LDL-induced injury in AS cell model through up-regulating the expression of ASPH via targeting miR-206.


Background
Atherosclerosis (AS) is a chronic inflammatory disorder, and it causes cardiovascular diseases, including stroke and myocardial infarction [1,2]. Macrophages play crucial function in AS initiation and progression, and the abnormal apoptosis and failure in the scavenging of apoptotic body both contribute to the inflammatory response of AS [3]. The pathogenesis of AS needs to be explored to find novel effective targets for AS treatment. Oxidized low-density lipoprotein (ox-LDL) induces the inflammatory response of macrophages, eventually causing the apoptosis of macrophages [4]. We established the AS cell model through treating THP-1-derived macrophages by ox-LDL to study the molecular mechanism behind AS progression in vitro [5].
Aspartyl (asparaginyl) β-hydroxylase (ASPH) belongs to type II transmembrane protein family [20]. Recent studies have pointed out the crucial regulatory role of ASPH in the initiation and development of cancers. For instance, miR-135a was found to suppress the progression of endometrial cancer through targeting ASPH [21]. MiR-200a restrained the proliferation and motility of hepatocellular carcinoma cells through reducing ASPH level [22]. However, the interaction between miR-206 and ASPH has never been identified and the role of ASPH in AS remains to be disclosed.
Our purpose was to explore the biological function and working mechanism of circTM7SF3 in AS development through using ox-LDL-induced AS cell model. The interaction between miR-206 and circTM7SF3 or ASPH was identified for the first time in our study.

Sample collection
Thirty-nine peripheral blood samples (2 mL per subject) obtained from AS patients, along with 39 peripheral blood samples (2 mL per subject) obtained from healthy volunteers were collected at The First Hospital of Lanzhou University. Written informed consent was provided by each subject before sample collection.The protocol in this study was approved by the Ethics Committee of The First Hospital of Lanzhou University. Clinical and biological characteristics of human participants were shown in Table 1. Human monocyte-derived macrophages (hMDMs) were extracted from the peripheral blood of one volunteer using the methods described previously [24,25]. hMDMs were cultivated using RPMI-1640 (Gibco) added with 10% FBS (Gibco) and 200 nM PMA (Sigma) for 72 h. Differentiated hMDMs were washed using phosphate buffer saline (PBS; Sangon Biotech, Shanghai, China) for three times and then induced by 50 μg/mL ox-LDL for 24 h for further analysis.

Stability analysis of circTM7SF3
Transcription inhibitor Actinomycin D (Sigma; 2 mg/mL; Cat. no. SBR00013) was incubated with THP-1-derived macrophages for 0 h, 4 h, 8 h, 12 h or 24 h, and qRT-PCR was conducted to detect the expression of circTM7SF3 and TM7SF3 messenger RNA (mRNA).
A total of 3 μg RNA sample isolated from THP-1-derived macrophages was incubated with or without 9 units RNase R (Sigma; Cat. no. R6513) for 30 min at room temperature, and the levels of circTM7SF3 and TM7SF3 mRNA were examined by qRT-PCR.

Establishment of circRNA-miRNA-mRNA signal axis
The interaction between circTM7SF3 and miR-206 was predicted by starbase software. The mRNA targets of miR-206 were predicted by starbase, miRWalk and Tar-getScan softwares.

RNA-pull down assay
All procedures were conducted under the RNase-free condition. THP-1-derived macrophages were disrupted, and 2 μg cell lysate was incubated with 100 pmol biotinylated RNA (Bio-miR-NC, Bio-miR-206 and Bio-miR-206-MUT) followed by 100 μL streptavidin agarose beads incubation for 1 h. After boiling using sodium dodecyl sulfate (SDS) buffer, qRT-PCR was applied to detect the enrichment of circTM7SF3 and ASPH mRNA.

Statistical analysis
The statistical results from three independent experiments with three technical replications were displayed as mean ± standard deviation. The differences in two groups were analyzed by Student's t-test, and the comparison in multiple groups was analyzed by one-way analysis of variance (ANOVA) followed by Tukey's test. Linear correlation was analyzed by Spearman's correlation coefficient. Differences were statistically significant with the P value of less than 0.05.

CircTM7SF3 level is enhanced in THP-1-derived macrophages by ox-LDL treatment
According to the results of GSE107522, the five most up-regulated and down-regulated circRNAs in THP-1-derived macrophages with ox-LDL treatment compared with un-treated THP-1-derived macrophages were shown in Fig. 1a. CircTM7SF3 (hsa_circ_0007478) was one of the up-regulated circRNAs after ox-LDL treatment in THP-1-derived macrophages (Fig. 1a). The gene information of circTM7SF3 in Circular RNA Interactome database (https ://circi ntera ctome .nia.nih.gov) was shown in Fig. 1b. Serum samples from healthy volunteers (Control group; n = 39) and AS patients (n = 39) were collected to explore if circTM7SF3 was up-regulated in AS patients. A significant up-regulation in the expression of circTM7SF3 was observed in the serum samples of AS patients compared with that of control group (Fig. 1c). The clinical and biological characteristics of human participants were listed in Table 1, and patients with AS were significantly associated with hypertension (P < 0.001). The regulatory relationship between circT-M7SF3 and ox-LDL in THP-1-derived macrophages was subsequently tested. As shown in Fig. 1d and e, ox-LDL treatment time-or dose-dependently increased the expression of circTM7SF3 in THP-1-derived macrophages. 50 μg/mL ox-LDL treatment for 24 h was used in the following experiments. The stability of circT-M7SF3 was tested using transcription inhibitor Actinomycin D and exonuclease RNase R. Compared with its linear counterpart TM7SF3 mRNA, circTM7SF3 level remained stable after Actinomycin D treatment for 24 h (Fig. 1f ). TM7SF3 mRNA was significantly decreased with RNase R treatment, and circTM7SF3 was resistant to RNase R digestion (Fig. 1g).

Ox-LDL suppresses the viability and promotes the apoptosis, inflammation and oxidative stress of THP-1-derived macrophages through elevating circTM7SF3 level
We silenced circTM7SF3 in ox-LDL-induced THP-1-derived macrophages to explore if the up-regulation of circTM7SF3 was necessary for ox-LDL-induced effects in THP-1-derived macrophages. Three siRNAs targeting circTM7SF3 were designed to silence circT-M7SF3 in THP-1-derived macrophages. Among three siRNAs, the knockdown efficiency of si-circTM7SF3_3 was the highest (Fig. 2a), thus si-circTM7SF3_3 (written as si-circTM7SF3) was selected in the following loss-of-function experiments. ox-LDL down-regulated cell viability and promoted the apoptosis of THP-1-derived macrophages, while the interference of circT-M7SF3 largely recovered the viability and inhibited the apoptosis of THP-1-derived macrophages (Figu. 2 b and c). We also detected the concentrations of inflammation-associated cytokines to assess cell inflammation. ox-LDL-induced elevation of TNF-α and IL-6 was partly attenuated by the silencing of circTM7SF3 in THP-1-derived macrophages ( Fig. 2d and e). To further verify the effect of circTM7SF3 interference on ox-LDL-induced inflammation in AS, primary human monocyte-derived macrophages (hMDMs) were isolated from the peripheral blood of one volunteer, and we treated hMDMs with PMA followed by ox-LDL to establish AS cell model. As shown in Additional file 1: Figure 1A, ox-LDL-induced up-regulation in circT-M7SF3 expression was partly counteracted by the addition of si-circTM7SF3 in differentiated hMDMs. ox-LDL treatment promoted the secretion of inflammatory cytokines (TNF-α and IL-6), while these promoting effects were partly attenuated by the interference of circTM7SF3 in differentiated hMDMs (Additional file 1: Figure 1B and C), which demonstrated that ox-LDL induced inflammatory response in differentiated hMDMs partly through up-regulating circTM7SF3.
We also assessed the oxidative stress status of THP-1 cells through measuring the levels of several oxidative stress indicators (iNOS, ROS, SOD and MDA). ox-LDL-induced up-regulation of iNOS, ROS and MDA and down-regulation of SOD were partly overturned by circTM7SF3 knockdown in THP-1-derived macrophages ( Fig. 2f-i), suggesting that ox-LDL-induced oxidative stress in THP-1-derived macrophages was partly attenuated by the interference of circTM7SF3.

CircTM7SF3 directly interacts with miR-206 in THP-1-derived macrophages
Starbase software was used to seek the possible miRNA targets of circTM7SF3 (Fig. 3a). Among these candidate miRNA targets, five miRNAs, including miR-320b, miR-613, miR-425-5p, miR-206 and miR-34a-5p, have been reported to be down-regulated in AS, which exhibited to an opposite tendency to circTM7SF3, thus we focused on these five miRNAs. THP-1-derived macrophages were transfected with si-NC or si-circTM7SF3 for 24 h, and we found that the expression of miR-320b, miR-425-5p, miR-206 and miR-34a-5p was up-regulated with the interference of circTM7SF3, especially the miR-206 (Fig. 3b), thus we concentrated on the potential interaction between miR-206 and circTM7SF3. The putative binding sites between miR-206 and circTM7SF3 were shown in Fig. 3c. To test whether miR-206 could bind to circTM7SF3 in THP-1-derived macrophages, we conducted dual-luciferase reporter assay and RNA-pull down assay. THP-1-derived macrophages were co-transfected with miR-NC or miR-206 and circTM7SF3-WT or circTM7SF3-MUT. Compared with miR-NC and circT-M7SF3-WT co-transfected group, luciferase activity was markedly reduced in circTM7SF3-WT group when miR-206 was overexpressed (Fig. 3d). Furthermore, the luciferase activity was un-affected in circTM7SF3-MUT group when co-transfected with miR-NC or miR-206 (Fig. 3d), suggesting that miR-206 bound to circTM7SF3 via the predicted sites. MiR-206 was biotinylated to pull down its associated molecules. As shown in Fig. 3e, circTM7SF3 was enriched in Bio-miR-206 precipitation complex, which further verified the interaction between miR-206 and circTM7SF3. The expression of miR-206 in AS patients and healthy volunteers was explored. There was a significant reduction in miR-206 expression in the serum samples of AS patients (n = 39) compared with healthy volunteers (n = 39) (Fig. 3f ). Given the opposite expression tendency between miR-206 and circTM7SF3, we analyzed the linear correlation between the two molecules. CircTM7SF3 expression was negatively correlated with the level of miR-206 ( Fig. 3g; P<0.0001). MiR-206 level was decreased with ox-LDL treatment in THP-1-derived macrophages in a dose-or time-dependently manner ( Fig. 3h and i). MiR-206 level was notably increased with the interference of circTM7SF3 in THP-1-derived macrophages (Fig. 3j), suggesting the negative regulatory relationship between miR-206 and circTM7SF3.

Si-circTM7SF3-mediated effects in ox-LDL-induced THP-1-derived macrophages are partly counteracted by the addition of anti-miR-206
MiR-206 level was elevated in circTM7SF3 silencing group, and the co-transfection with anti-miR-206 decreased the expression of miR-206 in THP-1-derived macrophages (Fig. 4a). We conducted rescue experiments using similar treatments in ox-LDL-induced THP-1-derived macrophages to test if circTM7SF3 functioned partly through targeting miR-206. The protective effect of si-circTM7SF3 in the viability of ox-LDL-induced THP-1-derived macrophages was ameliorated by the introduction of anti-miR-206 (Fig. 4b). Also, the apoptosis was inhibited in ox-LDL and si-circTM7SF3 group, and miR-206 silencing promoted the apoptosis of THP-1-derived macrophages again (Fig. 4c). The release of pro-inflammatory cytokines was suppressed in circT-M7SF3 silencing group in ox-LDL-induced THP-1-derived macrophages, and the addition of anti-miR-206 further enhanced the concentrations of TNF-α and IL-6 ( Fig. 4d and e). The levels of iNOS, ROS and MDA were reduced in si-circTM7SF3 and ox-LDL group, while the expression of these molecules was partly up-regulated with the co-transfection of anti-miR-206 in THP-1-derived macrophages (Fig. 4f, g and  i). The level of SOD exhibited a reverse phenomenon to iNOS, ROS or MDA (Fig. 4h), demonstrating that circTM7SF3 interference suppressed the oxidative stress of ox-LDL-induced THP-1-derived macrophages partly through elevating miR-206 level.

ASPH is a direct target of miR-206 in THP-1-derived macrophages
We analyzed the aberrantly expressed molecules in GSE54666 and GSE54039 datasets, and eight genes were consistently reported to be up-regulated or downregulated in ox-LDL-induced THP-1-derived macrophages in two datasets (Fig. 5a). ASPH was found to be up-regulated after ox-LDL treatment in both datasets (Fig. 5a). Meanwhile, we predicted the downstream genes of miR-206 through using three bioinformatic softwares, including starbase, miRWalk and TargetScan softwares. A total of 409 genes were predicted to be putative targets of miR-206 by three softwares (Fig. 5b). In view of the expression pattern and the possible interaction with miR-206, we aimed to test the interaction between miR-206 and ASPH using dual-luciferase reporter assay and RNA-pull down assay. The complementary sites between miR-206 and ASPH predicted by starbase software were shown in Fig. 5c. Luciferase activity was dramatically reduced in miR-206 and ASPH 3'UTR-WT group compared with miR-NC and ASPH 3'UTR-WT group (Fig. 5d), proving the direct interaction between these two genes in THP-1-derived macrophages. The results of RNA-pull down assay suggested that ASPH mRNA was most enriched in Bio-miR-206 group than that in Bio-miR-NC or Bio-miR-206-MUT group (Fig. 5e), which further confirmed the interaction between miR-206 and ASPH. The expression of ASPH mRNA was significantly enhanced in the serum samples of AS patients (n = 39) in comparison with that in healthy volunteers (n = 39) (Fig. 5f ). The mRNA level of ASPH was elevated in THP-1-derived macrophages with ox-LDL treatment in a dose-or time-dependent manner ( Fig. 5g and h). ASPH expression was negatively correlated with miR-206 level (P<0.0001), while there The targets of miR-206 were analyzed using starbase, miRWalk and TargetScan softwares. There were 409 genes (including ASPH) that simultaneously predicted to be targets of miR-206 by these three bioinformatic softwares. c The putative binding sites between miR-206 and ASPH predicted by starbase software were shown. d and e The target relationship between miR-206 and ASPH was evaluated by dual-luciferase reporter assay and RNA-pull down assay (n = 3). The results in D were evaluated by Student's t-test, while the results in E were assessed by ANOVA followed by Tukey's test. f ASPH mRNA level in the serum samples of control healthy volunteers (n = 39) and AS patients (n = 39) was detected by qRT-PCR. The results were analyzed using Student's t-test. g and h After exposing to ox-LDL for different doses or treatment times, ASPH mRNA level was measured by qRT-PCR assay (n = 3). The results were assessed by ANOVA followed by Tukey's test. (I and J) The linear relation between the levels of ASPH mRNA and miR-206 or circTM7SF3 was analyzed by Spearman's correlation coefficient. k and l THP-1 cells were treated with si-NC, si-circTM7SF3, si-circTM7SF3 + anti-miR-NC or si-circTM7SF3 + anti-miR-206. The mRNA and protein levels of ASPH in THP-1 cells (n = 3) were detected by qRT-PCR and Western blot assay. The protein bands shown in Fig. 5L have been cropped. The original Western blot data in Fig. 5L could be find in Additional file 2. The results were assessed by ANOVA followed by Tukey's test. *: P < 0.05 was a positive correlation between the expression of ASPH and circTM7SF3 (P<0.0001) ( Fig. 5i and j). After verifying the interaction between miR-206 and circT-M7SF3 or ASPH, we aimed to test the regulatory relationship among these three genes in THP-1-derived macrophages. As shown in Fig. 5k and l, circTM7SF3 silencing down-regulated the mRNA and protein levels of ASPH, and its mRNA and protein levels were up-regulated in si-circTM7SF3 and anti-miR-206 cotransfected group, suggesting that circTM7SF3 silencing reduced ASPH expression partly via up-regulating miR-206 in THP-1-derived macrophages.

MiR-206 overexpression-mediated influences in ox-LDL-induced THP-1-derived macrophages are partly overturned by ASPH accumulation
We conducted rescue experiments through transfecting miR-206 alone or together with pcDNA-ASPH into THP-1-derived macrophages to test if miR-206 exerted its functions partly by down-regulating ASPH. MiR-206 accumulation reduced the mRNA and protein enrichment of ASPH, and its mRNA and protein levels were largely rescued by the addition of ASPH overexpression plasmid ( Fig. 6a and b). ox-LDL-induced reduction of cell viability and promotion of apoptosis were largely attenuated by the overexpression of miR-206 ( Fig. 6c and d), and these findings further confirmed that miR-206 protected THP-1-derived macrophages against ox-LDL-induced injury. Furthermore, ASPH overexpression ameliorated the protective role of miR-206 in ox-LDLinduced THP-1-derived macrophages again ( Fig. 6c and  d). MiR-206 overexpression diminished the inflammatory response and oxidative stress of THP-1-derived macrophages (Fig. 6e-j). The addition of ASPH overexpression plasmid promoted the release of TNF-α and IL-6 ( Fig. 6e, f ). Through analyzing the levels of iNOS, ROS and MDA and SOD in six groups, we found that miR-206-mediated suppression in the oxidative stress of ox-LDL-induced THP-1-derived macrophages was overturned by the addition of pcDNA-ASPH (Fig. 6g-j).

Discussion
We found that circTM7SF3 level was elevated with ox-LDL treatment in THP-1-derived macrophages, and it was also up-regulated in the serum of AS patients compared with healthy controls. CircTM7SF3/miR-206/ ASPH signal axis was identified in this study for the first time. CircTM7SF3 promoted ox-LDL-induced cell injury in AS cell model through increasing ASPH expression via sponging miR-206. AS is a polygenic and multifactorial inflammatory disorder [26]. Macrophages are identified as major contributors for the initiation and progression of AS [27]. In the current study, we established AS cell model through exposing THP-1-derived macrophages to ox-LDL in vitro. CircTM7SF3 level was found to be notably increased in AS patients in contrast to that in healthy volunteers. Besides, ox-LDL markedly enhanced circT-M7SF3 level in a dose-and time-dependent manner in THP-1-derived macrophages. Through silencing circT-M7SF3 in ox-LDL-treated AS cell model, we found that ox-LDL-mediated injury in THP-1-derived macrophages was largely ameliorated, suggesting that ox-LDL promoted the apoptosis, inflammation and oxidative stress and inhibited the viability of THP-1-derived macrophages partly through up-regulating circTM7SF3 level. The results of this study on THP-1-derived macrophages were further confirmed on a hMDMs AS cell model. We observed that ox-LDL also accelerated the inflammatory response of differentiated hMDMs via up-regulating circTM7SF3.
MiR-206 was identified as a tumor suppressor in many types of cancers. For instance, miR-206 hampered the proliferation and invasion of thyroid cancer cells through targeting RAP1B [14]. MiR-206 restrained the proliferation and motility of cervical cancer cells through targeting BAG3 [28]. MiR-206 has also been implicated in the regulatory mechanism of cardiovascular injury. Hu et al. claimed that lncRNA UCA1 promoted ox-LDL-induced injury in THP-1 cells through sponging miR-206 [19]. Accumulating studies have reported that circRNAs functioned through targeting miRNAs [29][30][31]. For instance, circITGA7 hampered the progression of colorectal cancer through sponging miR-3187-3p, thus elevating ASXL1 level [32]. Circ_0003204 suppressed the proliferation, migration and tube formation of endothelial cells in AS through regulating miR-370-3p [33]. Here, the direct interaction between circTM7SF3 and miR-206 was confirmed. MiR-206 level was negatively regulated by ox-LDL or circTM7SF3 in THP-1-derived macrophages. Further studies suggested that ox-LDL induced the injury of THP-1-derived macrophages through targeting circT-M7SF3/miR-206 signal axis.
ASPH is a member of α-ketoglutarate-dependent dioxygenase family [34]. ASPH is highly expressed in embryogenesis, and it accelerates cellular migration in organ development [35,36]. ASPH was involved in the regulation of a few cancers. Zou et al. found that the hydroxylase activity of ASPH accelerated the epithelial-to-mesenchymal transition of hepatocellular carcinoma cells [20]. Chen et al. demonstrated that miR-135a exerted its anti-tumor role in endometrial cancer through targeting ASPH [21]. Here, ASPH was found as a direct target of miR-206 in THP-1-derived macrophages. CircTM7SF3 elevated the expression of ASPH through sponging miR-206 in THP-1-derived macrophages. Furthermore, rescue experiments revealed that miR-206 exhibited a protective role in ox-LDL-induced THP-1-derived macrophages through down-regulating ASPH level.

Limitations
Human acute monocytic leukemia cell line THP-1 was used for most of the experiments. Because THP-1 was a tumorigenic cell line, the results of experiments from AS cell model using ox-LDL-induced THP-1-derived macrophages might be different from AS cell model using ox-LDL-induced primary macrophages.

Conclusion
Taken together, circTM7SF3 was a circular RNA generated from TM7SF3 gene that stably existed in THP-1-derived macrophages. CircTM7SF3 contributed to ox-LDL-induced apoptosis, inflammatory response and oxidative stress through targeting miR-206, thus causing the increase of ASPH expression (Fig. 7). The identification of circTM7SF3/miR-206/ASPH signal axis disclosed a novel insight into the role of circTM7SF3 in AS progression, providing promising therapeutic target for AS.