Introduction

Sepsis is one of the leading causes of death due to infection, with a high incidence and rapid disease progression, resulting in a heavy economic burden worldwide1. Sepsis was first defined in 1991 by the American College of Chest Physicians (ACCP) and the Society of Critical Care Medicine (SCCM)2. To this day, sepsis remains an ill-defined syndrome. Nevertheless, with recent developments in medical technology, the treatment effect of sepsis has improved. Although many clinical trials aiming to inhibit the inflammatory response in sepsis have failed, a deeper understanding of the pathophysiological mechanisms of sepsis and the development of corresponding precise treatment strategies are still important for alleviating or reversing the progression of sepsis. In sepsis, invading pathogens trigger an imbalance in the immune response, leading to a pathological syndrome characterized by persistent, excessive inflammation and immunosuppression. Changes in the immune system function play an irreplaceable role in the occurrence and development of sepsis, among which macrophages are important participants. During different immune stages of sepsis, macrophages undergo various phenotypic changes. For example, in the excessive inflammatory response stage, macrophage polarization is mainly of the M1 type3, whereas in the immunosuppression stage, macrophage polarization is mainly of the M2 type4,5. In addition, the phagocytic function of macrophages is significantly reduced, and the level of apoptosis is increased in sepsis. These changes increase the risk of secondary infections and may aggravate sepsis6,7,8.

An increasing number of researchers are attempting to uncover the complexity of sepsis through high-dimensional data analysis owing to advances in omics technologies that can simultaneously analyze multiple molecular components such as RNA, proteins, lipids, and metabolites9. CircRNAs are a newly identified class of non-coding RNAs (ncRNAs) characterized by their circular structure, high stability, and strong tissue-specific expression10. They participate in post-transcriptional regulation and other processes through various pathways, thereby regulating various physiological and pathological processes. At present, dozens of circRNAs, such as circC3P1, hsa_circRNA_104484, hsa_circRNA_104670, and circVMA21, have been found to change their expression levels in sepsis, influencing its occurrence and development11,12,13.

Exosomes are extracellular vesicles with diameters ranging from 30 to 150 nm that are produced through the endosomal pathway. They are secreted by various cell types and can be easily isolated from various biological fluids. The cargoes carried by exosomes include proteins, nucleic acids, lipids, and metabolites, which can reflect the origin and functional specificity of their parental cells14. After circularization in the nucleus, circRNAs are transported to the cytoplasm and can be selectively enriched in exosomes. Therefore, exosome detection is an effective method for identifying differentially expressed circRNAs in various diseases. Our previous research results showed that the circRNA expression profile in serum exosomes is altered in patients with sepsis compared with that in healthy individuals. The circRNA microarray results showed that the expression of 80 circRNAs was up-regulated and that of 52 circRNAs was down-regulated. Among them, hsa_circ_0001818 (hsa_circRNA_104670) is highly expressed in the serum exosomes of patients with sepsis and is of great diagnostic value12. Since exosomes can be produced by a variety of cell types and released into the circulatory system, there is a need to identify exosome donor cells with significantly altered circRNA expression levels. In sepsis, multiple cell types, including macrophages, epithelial cells, and endothelial cells, are activated and release exosomes15. Given the differential expression of hsa_circ_0001818 in sepsis, elucidating its regulatory role in cell function during sepsis has scientific and translational significance.

In this study, we identified the hsa_circ_0001818/miRNA/mRNA regulatory axis, which is involved in regulating septic macrophage apoptosis, phagocytosis, and the release of inflammatory cytokines, leading to excessive inflammation and immunosuppression in sepsis.

Results

Expression level of hsa_circ_0001818 was increased in the macrophage model of sepsis

LPS was used to stimulate BEAS-2B cells, HUVECs, and THP-1-derived macrophages for 24 h. The qRT-PCR results showed that the expression level of hsa_circ_0001818 increased in THP-1-derived macrophages after LPS stimulation (P < 0.05). Conversely, there was no significant change in the expression level of hsa_circ_0001818 in BEAS-2B cells and HUVECs (Fig. 1a). Subsequently, LPS was used to stimulate THP-1-derived macrophages at various time points (0, 4, 24, 48, and 72 h). The results showed that the expression of hsa_circ_0001818 increased after 24 h of LPS stimulation (P < 0.01) and gradually increased with increasing LPS stimulation time. The expression level of hsa_circ_0001818 in the 48 h LPS stimulation group was significantly higher than that in the control group (P < 0.001). Therefore, in subsequent experiments, the earliest time at which hsa_circ_0001818 showed significant changes after LPS stimulation (48 h) was selected as the LPS stimulation time (Fig. 1b). In addition, the cells were cultured in complete medium containing 500 ng/mL actinomycin D for 0, 4, 8, 12, 24, or 30 h. The results showed that the transcriptional inhibitory effect of actinomycin D on hsa_circ_0001818 was significantly lower than that on the linear parent gene UBR5, indicating that hsa_circ_0001818 was highly stable (P < 0.05; Fig. 1c). As shown in Fig. 1d, compared to UBR5, hsa_circ_0001818 tolerated RNase R digestion, suggesting that hsa_circ_0001818 has a closed-ring structure.

Fig. 1
figure 1

Expression of hsa_circ_0001818 in the macrophage model of sepsis. (a) Expression levels of hsa_circ_0001818 after LPS stimulation (1 µg/mL) of different cell types. (b) Expression levels of hsa_circ_0001818 in THP-1-derived macrophages at different times of LPS stimulation. (c) Stability detection of hsa_circ_0001818. (d) Identification of the ring structure of hsa_circ_0001818. (e-h) Effects of LPS stimulation on the morphology, secretion of inflammatory cytokines, apoptosis rate and phagocytic function of THP-1-derived macrophages. *P < 0.05, **P < 0.01, and ***P < 0.001.

Full size image

Microscopic observations showed that after LPS stimulation, the morphology of THP-1-derived macrophages changed. Cells without LPS stimulation were round or oval, whereas those stimulated with LPS exhibited extended pseudopodia in the form of spindles or irregular shapes (Fig. 1e). The expression levels of tumor necrosis factor -α (TNF-α), IL-6, and IL-1β, released into the cell culture supernatant by activated THP-1-derived macrophages, were significantly upregulated (P < 0.05; Fig. 1f). Moreover, the apoptosis rate was significantly increased (P < 0.05; Fig. 1g). However, as shown in Fig. 1h, DAPI-stained nuclei emitted blue fluorescence, and the microspheres phagocytosed into the cells emitted red fluorescence. In the activated THP-1-derived macrophages, the number of fluorescent microspheres phagocytosed was significantly reduced, and therefore, their phagocytic function was reduced(P < 0.05).

hsa_circ_0001818 promotes apoptosis and inflammatory cytokine production in sepsis macrophages, but inhibits their phagocytic function

THP-1-derived macrophages were stimulated with LPS for 48 h before being transfected with either si-NC (negative control) or one of three siRNAs targeting hsa_circ_0001818 (si-hsa_circ_0001818-1, si-hsa_circ_0001818-2, and si-hsa_circ_0001818-3), generating the negative control group (LPS + si-NC) and the low-expression groups (LPS + si-hsa_circ_0001818-1, -2, and − 3). Compared with the LPS + si-NC group, the expression level of hsa_circ_0001818 in the LPS + si-hsa_circ_0001818-1, LPS + si-hsa_circ_0001818-2, and LPS + si-hsa_circ_0001818-3 groups were all reduced (P < 0.05; Fig. 2a). Among them, the LPS + si-hsa_circ_0001818-2 group showed the highest silencing efficiency and was therefore selected for subsequent experiments. Compared with the empty vector group, the expression level of hsa_circ_0001818 in the hsa_circ_0001818 overexpression group significantly increased (P < 0.05; Fig. 2e). In addition, compared with the LPS + si-NC group, the secretion levels of inflammatory cytokines in the cell culture supernatant of the LPS + si-hsa_circ_0001818 group decreased (P < 0.05; Fig. 2b), the cell apoptosis rate decreased (P < 0.05; Fig. 2c), while the phagocytic ability of the cells was enhanced (P < 0.05; Fig. 2d). Compared with the empty vector group, the secretion level of inflammatory cytokines in the cell culture supernatant of the hsa_circ_0001818 overexpression group increased (P < 0.05; Fig. 2f), the cell apoptosis rate increased (P < 0.05; Fig. 2g), while the cell phagocytic function weakened (P < 0.05; Fig. 2h).

Fig. 2
figure 2

Regulatory effect of hsa_circ_0001818 on the function of THP-1-derived macrophages. (a) Silencing efficiency of hsa_circ_0001818 in THP-1-derived macrophages. (b-d) Effects of silencing hsa_circ_0001818 on the secretion of inflammatory cytokines (TNF-α, IL-6, and IL-1β), apoptosis rate, and phagocytic function of THP-1-derived macrophages. (e) Overexpression efficiency of hsa_circ_0001818 in THP-1-derived macrophages. (f-h) Effects of overexpression of hsa_circ_0001818 on the secretion of inflammatory cytokines (TNF-α, IL-6, IL-1β), apoptosis rate, and phagocytic function of THP-1-derived macrophages. *P < 0.05, **P < 0.01, and ***P < 0.001.

Full size image

hsa_circ_0001818 negatively regulates miR-17-3p, miR-433-3p and miR-642a-5p through sponge-like targeting

Arraystar software was used to predict the binding sites of hsa_circ_0001818 on miR-17-3p, miR-433-3p, and miR-642a-5p (Fig. 3a). Dual-luciferase reporter plasmids containing wild-type and mutant hsa_circ_0001818 (WT-hsa_circ_0001818 and MUT-hsa_circ_0001818, respectively) were constructed based on the predicted binding sites (Fig. 3b) and transfected into 293T cells to detect the binding of hsa_circ_0001818 to miRNAs. The results of the dual-luciferase assay showed that in cells transfected with WT-hsa_circ_0001818 plasmids, the luciferase activity of the co-transfection miR-17-3p mimic, miR-433-3p mimic, and miR-642a-5p mimic groups were significantly lower than that of the co-transfection mimic-NC group (P < 0.05). Conversely, in cells transfected with MUT-hsa_circ_0001818 plasmids, there were no significant differences in luciferase activity between the co-transfection miR-17-3p mimic, miR-433-3p mimic, and miR-642a-5p mimic groups and the mimic-NC group (P > 0.05; Fig. 3c). In subsequent experiments, hsa_circ_0001818 was silenced and overexpressed. Compared with the LPS + siNC group, the expression levels of miR-17-3p, miR-433-3p, and miR-642a-5p were upregulated in the LPS + si-hsa_circ_0001818 group (P < 0.05). Conversely, in the hsa_circ_0001818 overexpression group, these miRNAs were downregulated compared to the empty vector group (P < 0.05; Fig. 3d).

Fig. 3
figure 3

hsa_circ_0001818 acts as a molecular sponge, suppressing miR-17-3p, miR-433-3p and miR-642a-5p levels. (a) Bioinformatics prediction of the binding sites of hsa_circ_0001818 with miR-17-3p, miR-433-3p, and miR-642a-5p. (b) Construction of WT-hsa_circ_0001818 and MUT-hsa_circ_0001818 recombinant reporter gene plasmids. (c) Dual luciferase reporter gene assay to detect the binding of hsa_circ_0001818 with miR-17-3p, miR-433-3p and miR-642a-5p. (d) Regulatory effect of hsa_circ_0001818 on miR-17-3p, miR-433-3p, and miR-642a-5p. *P < 0.05, **P < 0.01, and ***P < 0.001.

Full size image

hsa_circ_0001818 increases the apoptosis rate of macrophages by targeting miR-17-3p/CASP3

Caspase-3, encoded by CASP3 mRNA, is a protein kinase involved in apoptosis signaling. CASP3 is also predicted to be regulated by miR-17-3p. We measured the the expression levels of miR-17-3p, CASP3 mRNA and protein in the sepsis macrophage model. The results showed that the expression level of miR-17-3p in THP-1-derived macrophages stimulated with LPS was lower than that in the control group (P < 0.05; Fig. 4b), and the expression levels of CASP3 mRNA and protein were higher than those in the control group (P < 0.05)(Fig. 4a, c, and S1). Subsequently, THP-1-derived macrophages were stimulated with LPS for 48 h, and the miR-17-3p mimic and mimic-NC groups were transfected to construct the miR-17-3p high-expression group. Additionally, the miR-17-3p inhibitor and inhibitor-NC groups were transfected to construct the miR-17-3p low-expression group. Compared with the LPS + mimic-NC group, the apoptosis rate was decreased in the LPS + miR-17-3p mimic group (P < 0.05). Conversely, compared to the inhibitor-NC group, the apoptosis rate was increased in the miR-17-3p inhibitor group (Fig. 4d).

Fig. 4
figure 4

hsa_circ_0001818 regulates apoptosis of THP-1-derived macrophages by targeting miR-17-3p/CASP3. (a) Expression level of CASP3 mRNA in THP-1-derived macrophages after LPS stimulation. (b) Expression level of miR-17-3p in THP-1-derived macrophages after LPS stimulation. (c) Expression level of CASP3 protein in THP-1-derived macrophages after LPS stimulation. (d) Regulatory effect of miR-17-3p on apoptosis level of THP-1-derived macrophages. (e) Bioinformatics prediction of the binding sites of miR-17-3p and CASP3. (f) Construction of WT-CASP3 and MUT-CASP3 recombinant reporter gene plasmids. (g) Dual luciferase reporter gene assay to detect the binding of CASP3 and miR-17-3p. (h) Expression level of CASP3 mRNA after overexpression and silencing of miR-17-3p. (i) Expression level of CASP3 protein after overexpression and silencing of miR-17-3p. (j) Apoptosis rate of THP-1-derived macrophages in the rescue experiment. (k) Expression level of CASP3 mRNA in the rescue experiment. (l) Expression level of CASP3 protein in the rescue experiment. *P < 0.05, **P < 0.01, and ***P < 0.001.

Full size image

TargetScan website was used to predict the binding sites of miR-17-3p and CASP3 (Fig. 4e). Based on the predicted binding sites, dual-luciferase reporter plasmids carrying wild type and mutant CASP3 (WT-CASP3 and MUT-CASP3, respectively) were constructed (Fig. 4f) and transfected into 293T cells to detect the binding of CASP3 and miR-17-3p. The results showed that in cells transfected with WT-CASP3 plasmids, the luciferase activity of the co-transfected miR-17-3p mimic group was significantly lower than that of the co-transfected mimic-NC group (P < 0.05), while in cells transfected with MUT-CASP3 plasmids, there was no significant difference in luciferase activity between the miR-17-3p mimic co-transfection group and the mimic-NC co-transfection group (P > 0.05; Fig. 4g). Compared with the LPS + mimic-NC group, the mRNA and protein expression levels of CASP3 in the cells of the LPS + miR-17-3p mimic group were downregulated (P < 0.05). Conversely, compared with the inhibitor-NC group, the mRNA and protein expression levels of CASP3 in the miR-17-3p inhibitor group were upregulated (P < 0.05; Fig. 4h, i, and S2).

To verify whether hsa_circ_0001818 regulated THP-1-derived macrophage apoptosis by targeting miR-17-3p/CASP3, a rescue experiment was designed. THP-1-derived macrophages were stimulated with LPS for 48 h and then co-transfected with si-hsa_circ_0001818 and miR-17-3p inhibitors. Cells were divided into four groups: LPS + si-NC, LPS + si-hsa_circ_0001818, LPS + si-hsa_circ_0001818 + inhibitor-NC, and LPS + si-hsa_circ_0001818 + miR-17-3p inhibitor. The results showed that compared with the LPS + si-NC group, the apoptosis rate of macrophages in the LPS + si-hsa_circ_0001818 group was reduced (P < 0.05). However, compared with the LPS + si-hsa_circ_0001818 + inhibitor-NC group, apoptosis was reversed in the LPS + si-hsa_circ_0001818 + miR-17-3p inhibitor group (P < 0.05; Fig. 4j). This suggests that the regulatory effect of hsa_circ_0001818 on macrophage apoptosis depends, to some extent, on miR-17-3p expression. Compared with the LPS + si-NC group, the expression levels of CASP3 mRNA and protein in macrophages in the LPS + si-hsa_circ_0001818 group were reduced (P < 0.05). In contrast, compared with the LPS + si-hsa_circ_0001818 + inhibitor-NC group, CASP3 mRNA and protein levels were restored in the LPS + si-hsa_circ_0001818 + miR-17-3p inhibitor group (P < 0.05; Fig. 4k, l, and S3). This suggests that the regulatory effect of hsa_circ_0001818 on the expression of CASP3 mRNA and protein in macrophages depends on miR-17-3p to a certain extent.

hsa_circ_0001818 inhibits macrophage phagocytic function by targeting miR-433-3p and miR-642a-5p

We measured the expression of miR-433-3p and miR-642a-5p in the macrophage model of sepsis. The results showed that the expression levels of miR-433-3p and miR-642a-5p were lower in the model group than in the control group (P < 0.05; Fig. 5a). To investigate their roles, THP-1-derived macrophages were stimulated with LPS for 48 h and then transfected with either mimic-NC, miR-433-3p mimic, or miR-642a-5p mimic to create high-expression groups, or with inhibitor-NC, miR-433-3p inhibitor, or miR-642a-5p inhibitor to create low-expression groups. Compared with the LPS + mimic-NC group, the phagocytic function of macrophages was enhanced in the LPS + miR-433-3p mimic and LPS + miR-642a-5p mimic groups (P < 0.05). Conversely, compared with the inhibitor-NC group, phagocytic function was reduced in the miR-433-3p and miR-642a-5p inhibitor groups (P < 0.05; Fig. 5b).

Fig. 5
figure 5

hsa_circ_0001818 regulates the phagocytic function of THP-1-derived macrophages by targeting miR-433-3p and miR-642a-5p. (a) Expression of miR-433-3p and miR-642a-5p in THP-1-derived macrophages after LPS stimulation. (b) Changes in macrophage phagocytic function after overexpression and silencing of miR-433-3p and miR-642a-5p. (c) Changes in THP-1-derived macrophage phagocytic function in the rescue experiment. *P < 0.05, **P < 0.01, and ***P < 0.001.

Full size image

To verify whether hsa_circ_0001818 regulates THP-1-derived macrophage apoptosis by targeting miR-433-3p and miR-642a-5p, we designed a rescue experiment. LPS was used to stimulate THP-1-derived macrophages for 48 h, followed by co-transfection with si-hsa_circ_0001818 and miRNA inhibitor, and the cells were divided into five groups: LPS + si-NC, LPS + si-hsa_circ_0001818, LPS + si-hsa_circ_0001818 + inhibitor-NC, LPS + si-hsa_circ_0001818 + miR-433-3p inhibitor, and LPS + si-hsa_circ_0001818 + miR-642a-5p inhibitor group. Compared with the LPS + si-NC group, macrophage phagocytic function was enhanced in the LPS + si-hsa_circ_0001818 group (P < 0.05). However, compared to the LPS + si-hsa_circ_0001818 + inhibitor-NC group, phagocytic function was reversed in the LPS + si-hsa_circ_0001818 + miR-433-3p inhibitor and LPS + si-hsa_circ_0001818 + miR-642a-5p inhibitor groups (P < 0.05)(Fig. 5c). This suggests that hsa_circ_0001818 regulates the phagocytic function of macrophages via miR-433-3p and miR-642a-5p.

hsa_circ_0001818 enhances the level of inflammatory cytokines producted by macrophages by regulating ZBTB20/NF-κB axis through miR-433-3p and miR-642a-5p

ZBTB20 is a member of the zinc finger and BTB domain-containing protein subfamily and has been reported to be involved in regulating the NF-κB signaling pathway16,17. After LPS stimulation, ZBTB20 mRNA, protein, and NF-κB pathway-related phosphorylated proteins were significantly upregulated compared with the control group (P < 0.05; Fig. 6a–c, S4, and S5). Cell immunofluorescence detection showed that the ZBTB20 protein labeled with cy3 dye emitted orange-red fluorescence, while the DAPI-stained nucleus emitted blue fluorescence. Orange-red fluorescence intensity increased significantly after LPS stimulation and was distributed throughout the cytoplasm. This suggested that the ZBTB20 protein expression increased after LPS stimulation and was localized in the cytoplasm (Fig. 6d).

Fig. 6
figure 6

miR-433-3p and miR-642a-5p can target and inhibit the expression of ZBTB20. (a) Expression level of ZBTB20 mRNA in THP-1-derived macrophages after LPS stimulation. (b) Expression level of ZBTB20 protein in THP-1-derived macrophages after LPS stimulation. (c) Expression level of NF-κB pathway-related phosphorylated proteins in THP-1-derived macrophages after LPS stimulation. (d) Localization and expression level of ZBTB20 protein in THP-1-derived macrophages after LPS stimulation. (e) Bioinformatics prediction of the binding sites of miR-433-3p and miR-642a-5p with ZBTB20. (f) Construction of WT-ZBTB20 and MUT-ZBTB20 recombinant reporter gene plasmids. (g) Dual luciferase reporter gene assay was used to detect the binding of miR-433-3p and miR-642a-5p with ZBTB20. *P < 0.05, **P < 0.01, and ***P < 0.001.

Full size image

Additionally, ZBTB20 was predicted to be a regulatory target of miR-433-3p and miR-642a-5p. We used the TargetScan website to predict the binding sites of miR-433-3p and miR-642a-5p on ZBTB20 (Fig. 6e). Dual-luciferase reporter plasmids carrying wild-type ZBTB20 (WT-ZBTB20) and mutant ZBTB20 (MUT-ZBTB20) were constructed based on the predicted binding sites (Fig. 6f). In cells transfected with WT-ZBTB20 plasmid, the luciferase activity of the co-transfected miRNA mimic group was significantly lower than that of the co-transfected mimic-NC group (P < 0.05), while in cells transfected with MUT-ZBTB20 plasmid, there was no significant difference in luciferase activity between the miRNA mimic co-transfection group and the mimic-NC co-transfection group (P > 0.05; Fig. 6g). THP-1-derived macrophages were stimulated with LPS for 48 h, followed by transfection with mimic-NC, miR-433-3p mimic, or miR-642a-5p mimic to create high-expression groups. Additionally, inhibitor-NC, miR-433-3p inhibitor, and miR-642a-5p inhibitor were used to create low-expression groups. Compared to the LPS + mimic-NC group, ZBTB20 mRNA and protein levels were downregulated in the LPS + miR-433-3p mimic and LPS + miR-642a-5p mimic groups (P < 0.05; Fig. 7a, b, and S6). Correspondingly, NF-κB pathway-related phosphorylated proteins (P < 0.05; Fig. 7c, and S7) and secreted inflammatory cytokines (P < 0.05; Fig. 7d) were also decreased. Conversely, compared to the inhibitor-NC group, ZBTB20 mRNA and protein levels were upregulated in the miR-433-3p inhibitor and miR-642a-5p inhibitor groups (P < 0.05; Fig. 7a, b). This was accompanied by increased levels of phosphorylated proteins related to the NF-κB pathway (P < 0.05; Fig. 7c and S7) and elevated inflammatory cytokine secretion (P < 0.05; Fig. 7d).

Fig. 7
figure 7

miR-433-3p and miR-642a-5p regulate the inflammatory response of THP-1-derived macrophages by targeting ZBTB20/NF-κB. (a) Expression level of ZBTB20 mRNA after overexpression and silencing of miR-433-3p and miR-642a-5p. (b) Expression level of ZBTB20 protein after overexpression and silencing of miR-433-3p and miR-642a-5p. (c) Expression level of phosphorylated proteins related to the NF-κB pathway after overexpression and silencing of miR-433-3p and miR-642a-5p. (d) Levels of inflammatory cytokines (TNF-α, IL-6, and IL-1β) secreted by THP-1-derived macrophages after overexpression and silencing of miR-433-3p and miR-642a-5p. *P < 0.05, **P < 0.01, and ***P < 0.001.

Full size image

To verify whether hsa_circ_0001818 regulates the inflammatory response of THP-1-derived macrophages by targeting miR-433-3p and miR-642a-5p, we designed a rescue experiment. LPS was used to stimulate THP-1-derived macrophages for 48 h, followed by co-transfection with si-hsa_circ_0001818 and miRNA inhibitor, and the cells were divided into five groups: LPS + si-NC, LPS + si-hsa_circ_0001818, LPS + si-hsa_circ_0001818 + inhibitor-NC, LPS + si-hsa_circ_0001818 + miR-433-3p inhibitor, and LPS + si-hsa_circ_0001818 + miR-642a-5p inhibitor group. Compared with the LPS + si-NC group, macrophages in the LPS + si-hsa_circ_0001818 group showed decreased secretion of inflammatory cytokines (TNF-α, IL-6, and IL-1β) (P < 0.05; Fig. 8a), along with reduced ZBTB20 mRNA and protein expression (P < 0.05; Fig. 8b, c, and S8) and lower levels of NF-κB pathway-related phosphorylated proteins (P < 0.05; Fig. 8d and S9). However, compared with the LPS + si-hsa_circ_0001818 + inhibitor-NC group, the secretion of inflammatory cytokines, ZBTB20 expression, and phosphorylated NF-κB proteins were restored in the LPS + si-hsa_circ_0001818 + miR-433-3p inhibitor and LPS + si-hsa_circ_0001818 + miR-642a-5p inhibitor groups (P < 0.05; Fig. 8a–d, and S9). This suggests that hsa_circ_0001818 regulates macrophage inflammatory cytokine secretion, ZBTB20 expression, and NF-κB pathway activation, at least partially, through miR-433-3p and miR-642a-5p.

Fig. 8
figure 8

hsa_circ_0001818 regulates the inflammatory response of THP-1-derived macrophages by targeting ZBTB20/NF-κB through miR-433-3p and miR-642a-5p. (a) Levels of inflammatory cytokines (TNF-α, IL-6, and IL-1β) secreted by THP-1-derived macrophages in the rescue experiment. (b) Expression level of ZBTB20 mRNA in the rescue experiment. (c) Expression level of ZBTB20 protein in the rescue experiment. (d) Expression level of phosphorylated proteins related to the NF-κB pathway in the rescue experiment. *P < 0.05, **P < 0.01, and ***P < 0.001.

Full size image

Discussion

Our previous research showed that hsa_circ_0001818 was significantly elevated in the serum exosomes of patients with sepsis. We speculated that this increase is due to production by activated cells, influencing cell function during sepsis. To test this, we stimulated various cell lines with LPS in vitro to simulate sepsis. LPS-stimulated BEAS-2B was used to simulate lung infection, the most common primary site of sepsis. In addition, vascular endothelial cells are a single layer of squamous epithelium lining the inner surface of blood vessels, which forms the inner wall of blood vessels. They are large in number and widely distributed. In sepsis, vascular endothelial cells are damaged and produce a large number of circRNAs, most of which are released into the peripheral blood. LPS-stimulated HUVECs were used to explore the regulatory effects of circRNA on vascular endothelial cell function in sepsis. Furthermore, macrophages are important part of innate immunity. In addition to engulfing, clearing foreign bodies and aging and damaged cells, they also serve as the main antigen presenting cells to capture and process antigens at the initial stage of the immune response. LPS-stimulated THP-1-derived macrophages were used to explore the regulatory effects of circRNA on macrophage function in sepsis. Our findings reveal a novel role for hsa_circ_0001818 in regulating macrophage function during sepsis. Specifically, LPS stimulation of human monocyte(THP-1)-derived macrophages increased hsa_circ_0001818 expression. This circRNA promotes inflammatory cytokine production and macrophage apoptosis while inhibiting phagocytic function. Bioinformatics analysis revealed that hsa_circ_0001818 targets and binds to multiple immune function-related miRNAs. Therefore, we speculated that hsa_circ_0001818 may participate in regulating the immune response of macrophages in sepsis through a competing endogenouse RNA (ceRNA) mechanism. Further studies showed that hsa_circ_0001818 partially promotes CASP3 expression by sponging miR-17-3p, thereby promoting macrophage apoptosis. It also acts as a sponge for miR-433-3p and miR-642a-5p, facilitating inflammatory cytokine production through the ZBTB20/NF-κB axis and suppressing macrophage phagocytic function. These findings reveal a previously unknown mechanism by which hsa_circ_0001818 alters macrophage function, potentially offering a new therapeutic target for sepsis.

Macrophages are important components of innate and adaptive immunity, and are highly heterogeneous and plastic. When the microenvironment of tissues changes, they polarize into different phenotypes with varying immunoregulatory effects. During sepsis, the invasion of many pathogens activates macrophages and releases a variety of proinflammatory cytokines, which may cause organ damage or even death by causing excessive inflammatory responses3,18. Concurrently, pathogen-induced macrophage reprogramming reduces phagocytic function, impairing pathogen and cell debris clearance. This increases the risk of secondary infections, thereby aggravating the severity of sepsis6,7. In addition, pathogens induce macrophage apoptosis, hindering cell proliferation and weakening the host’s defense against infection19,20. Therefore, a deeper understanding of the entire molecular process of macrophage functional transformation in sepsis will help divide patients into subgroups with common features and improve the effects of therapies targeting specific pathophysiological mechanisms.

Studies have found that the epigenetic regulation of genes contributes to immune cell reprogramming mainly through histone modification and ncRNA regulation21. Among these, circRNAs are ncRNAs with special ring structures. In many diseases, circRNAs promote or delay disease progression by regulating macrophage polarization. Notably, circRNAs can promote macrophage M1 polarization and aggravate inflammatory responses. For example, circ_0000518 binds to the RNA binding protein FUS and promotes macrophage M1 polarization through the CaMKKβ/AMPK pathway, thereby aggravating multiple sclerosis22. circN4bp1 mediates M1 polarization of macrophages and promotes sepsis-induced ARDS through the miR-138-5p/EZH2 axis23. Additionally, circRNA Cdyl promotes the formation of abdominal aortic aneurysms by inducing the M1 polarization of macrophages and inflammatory responses24. In contrast, circRNAs participate in anti-inflammatory responses and tumor development by regulating the M2 polarization of macrophages25,26,27. Tumor-associated macrophages (TAMs) have an M2-like phenotype in tumors. In tumors, circRNAs cause immunosuppression by promoting M2 polarization or M1-to-M2 transformation of macrophages, leading to proliferation, metastasis, and drug resistance28,29,30,31,32,33,34,35,36,37,38,39,40. In addition, circRNAs are crucial to the regulation of macrophage apoptosis, pyroptosis, and autophagy. For example, circRNA_0050486 promotes oxLDL-induced apoptosis of atherosclerotic cells by targeting miR-127041. circNFIC promotes inflammation and apoptosis in macrophages via the miR-30e-3p/DENND1B axis42. Exosomal circRNA11:120406118|12,040,782 aggravates NLRP3-mediated pyroptosis of macrophages during the development of silicosis by sponging miR-30b-5p43. hsa_circ_0045474 is downregulated in tuberculosis macrophages and induces autophagy through the miR-582-5p/TNKS2 axis, suggesting a potential therapeutic target for active pulmonary tuberculosis44.

Several studies have demonstrated that the expression profiles of circRNAs in macrophages are altered during sepsis. High-throughput sequencing and analysis of circRNAs revealed that the expression profile of circRNAs changed in the pulmonary macrophages of mice with acute lung injury caused by sepsis, with 11 circRNAs being significantly upregulated and 126 circRNAs being significantly downregulated. Bioinformatics analysis has suggested that they may be involved in regulating the differentiation and polarization of pulmonary macrophages45. In a recent study, circRNA microarray technology was used to compare the expression of circRNAs in macrophages under two different polarization conditions. The results showed that 419 circRNAs were differentially expressed in M1 and M2 macrophages. Among these, circ-Cdr1as can increase the transcription of anti-inflammatory mediators and the percentage of CD206-positive cells, thus playing a key role in regulating the anti-inflammatory phenotype of macrophages46. Another study employing RNA sequencing analysis also found that the expression level of circRNAs differs between macrophages in different polarization states and may be involved in regulating the polarization process of macrophages47.

Among the predicted miRNAs targeted by hsa_circ_0001818, miR-17-3p plays an anti-apoptotic role in various diseases. For example, in colon cancer, miR-17-3p inhibits apoptosis by targeting PLCD1 and promotes tumor progression48. In intervertebral disc degeneration, reduced miR-17-3p levels decrease its inhibitory effect on Ang-2, leading to increased nucleus pulposus cell apoptosis49. Methyltransferase-like 3 (METTL3) improves cardiomyocyte proliferation and inhibits cardiomyocyte apoptosis after myocardial infarction by upregulating miR-17-3p50. Our results confirmed the direct binding of hsa_circ_0001818 to miR-17-3p. In addition, the expression of miR-17-3p is reduced in sepsis, hsa_circ_0001818 negatively regulates the expression of miR-17-3p, and miR-17-3p inhibits macrophage apoptosis. Rescue experiments showed that silencing miR-17-3p partially reversed the reduction in macrophage apoptosis mediated by silencing hsa_circ_0001818. Thus, hsa_circ_0001818 may promote apoptosis by targeting miR-17-3p. Previous studies have shown that miR-433-3p and miR-642a-5p are involved in regulating the functions of monocytes and macrophages. miR-433-3p promotes the M2 polarization of macrophages and exerts anti-inflammatory effects51,52, whereas miR-642a-5p can inhibits the LPS-induced inflammatory response of monocytes53. Our results confirmed the direct binding of hsa_circ_0001818 to miR-433-3p and miR-642a-5p. Subsequent experiments showed that the expression of miR-433-3p and miR-642a-5p was reduced in sepsis, miR-433-3p and miR-642a-5p could promoted macrophage phagocytosis, and hsa_circ_0001818 negatively regulated the expression of miR-433-3p and miR-642a-5p. Rescue experiments showed that simultaneous silencing of hsa_circ_0001818 and miR-433-3p/miR-642a-5p partially reversed the enhancement of macrophage phagocytosis mediated by hsa_circ_0001818 silencing. Therefore, our results confirmed that hsa_circ_0001818 inhibits the phagocytic function of macrophages in sepsis to a certain extent, which is dependent on miR-433-3p and miR-642a-5p. Further exploration is needed on how miR-433-3p and miR-642a-5p regulate the phagocytic function of macrophages. In addition, miR-433-3p and miR-642a-5p exert anti-inflammatory effects in a variety of diseases. For example, miR-433-3p can inhibit NF-κB signaling by targeting TLR1054, and dendrobium officinale polysaccharides can reduce intestinal inflammation by increasing the miR-433-3p content in small extracellular vesicles52. In ulcerative colitis, miR-642a-5p can inhibit inflammation by inhibiting the TLR4 signaling pathway in THP-1 cells and reducing the levels of LPS-induced inflammatory cytokines53. The results of this study showed that miR-433-3p and miR-642a-5p inhibit the secretion of inflammatory cytokines by macrophages. Rescue experiments showed that simultaneous silencing of hsa_circ_0001818 and miR-433-3p/miR-642a-5p partially reversed the reduction in macrophage secretion of inflammatory cytokines mediated by hsa_circ_0001818 silencing. These results indicated that the ability of hsa_circ_0001818 to promote the release of inflammatory cytokines by macrophages was partially dependent on miR-433-3p and miR-642a-5p.

We predicted the target mRNAs of miR-17-3p using bioinformatics. Among them, caspase-3, the translation product of CASP3 mRNA, is currently known to be a key executor of cell apoptosis, which causes cell apoptosis by cleaving multiple key proteins, such as the nuclear enzyme poly ADP-ribose polymerase (PARP)55. Therefore, we explored the regulatory mechanism of hsa_circ_0001818 in macrophage apoptosis. The results showed that the expression level of CASP3 was increased in sepsis, and that miR-17-3p could directly target and negatively regulate the expression of CASP3. Simultaneous silencing of hsa_circ_0001818 and miR-17-3p partially reversed the downregulation of CASP3 mediated by hsa_circ_0001818 silencing. Furthermore, bioinformatics technology was used to predict potential targets of miR-433-3p and miR-642a-5p, among which ZBTB20 has been confirmed to be a regulator of inflammatory response. The BTB/POZ family comprises nuclear DNA-binding transcription factors that participate in various biological processes56. ZBTB20, initially identified in human dendritic cells (DC) and named DPZF, is a member of this family. It features an N-terminal BTB/POZ domain and a C-terminal DNA-binding zinc finger domain, functioning primarily as a transcriptional repressor57. Studies by Liu et al. showed that ZBTB20 can act as a transcriptional inhibitor of the inhibitory gene IκBα, promote the activation of the NF-κB signaling pathway, and enhance the production of cytokines triggered by TLRs. After knocking out ZBTB20 gene, macrophages in septic mice secreted fewer inflammatory cytokines and were more resistant to septic shock16. In addition, studies by Qiu et al. also confirmed that ZBTB20 promoted titanium particle (TiP)-induced macrophage inflammatory responses by inhibiting transcription of IκBα and activating the NF-κB pathway17. Our results showed that miR-433-3p and miR-642a-5p inhibited the expression of ZBTB20 and NF-κB pathway-related phosphorylated proteins. Rescue experiments showed that silencing of miR-433-3p and miR-642a-5p could partially reverse the downregulation of the expression of ZBTB20 and NF-κB pathway-related phosphorylated proteins caused by silencing hsa_circ_0001818. Therefore, in sepsis, hsa_circ_0001818 may promote the production of inflammatory cytokines by macrophages by targeting ZBTB20/NF-κB through miR-433-3p and miR-642a-5p.

In conclusion, our findings reveal a novel pathophysiological mechanism involving hsa_circ_0001818, a newly identified ncRNA. This circRNA regulates the apoptosis, phagocytosis, and the secretion of inflammatory cytokines in septic macrophages through the miRNA/mRNA axis. This represents a new mechanism for sepsis progression. Due to their unique closed loop structure, enhanced stability, and highly tissue- and disease-specific expression patterns, circRNAs theoretically offer significant advantages as ideal therapeutic targets or tools. The high specificity of circRNAs in developing targeted therapies is expected to significantly reduce off-target effects, thereby improving the safety and precision of treatments. The regulatory mechanism of hsa_circ_0001818 is shown in Fig. 9. Furthermore, considering the mechanism by which exosomes serve as important vehicles for intercellular communication, it is possible that hsa_circ_0001818 expression in macrophages is elevated in sepsis and that it may be selectively packaged into exosomes and released into the extracellular microenvironment. Once these exosomes carry hsa_circ_0001818 in the bloodstream, they may act on macrophages or other cell types through autocrine or paracrine pathways, thereby mediating or amplifying its biological effects. Exosomal circRNAs may also become potential targets for precision targeted therapy. This will be one of our key research directions, aiming to further elucidate its regulatory network and clinical relevance. However, since hsa_circ_0001818 is a circRNA molecule unique to humans and no identical circRNA molecules have been found in animals, this study cannot be verified in vivo and cannot be verified in a mouse macrophage cell line. Therefore, larger-scale patient sample validation is needed to analyze the correlation between the expression level of hsa_circ_0001818 in macrophages of septic patients and the severity and prognosis of sepsis, and to analyze its correlation with currently known molecules related to the pathophysiological stage of sepsis.

Fig. 9
figure 9

Schematic diagram of the mechanism of hsa_circ_0001818 in regulating macrophage function in sepsis.

Full size image

Materials and methods

Cell culture and treatment

The normal human lung epithelial cell line BEAS-2B was obtained from the Medical Research Center of the Second Hospital of Jilin University and cultured in RPMI 1640 medium containing 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin solution. Human umbilical vein endothelial cells (HUVECs) cell line was collected from the Medical Research Center of the Second Hospital of Jilin University and cultured in high glucose Dulbecco’s modified Eagle’s medium (DMEM) containing 10% FBS and 1% penicillin-streptomycin solution. The human monocyte cell line THP-1 was procured from Wuhan Pricella Biotechnology Co. Ltd. (Wuhan, China) and cultured in RPMI 1640 medium containing 10% FBS, 1% penicillin-streptomycin solution, and 0.05 mM β-mercaptoethanol. All cell lines were cultured at 37 ℃ and 5% CO2.

THP-1 cells were differentiated into the M0 phenotype using 15 ng/mL phorbol 12-myristate 13-acetate (PMA) (M4647; Abmole, Shanghai, China) for 18 h. After the cells adhered to the culture dish, the PMA-containing medium was removed, and the cells were then cultured in fresh medium without PMA for an additional 24 h. Subsequently, the induced THP-1 cells were stimulated with lipopolysaccharide (LPS) (L4391; Sigma-Aldrich, Germany) at a concentration of 1 µg/mL to construct a human macrophage inflammation model.

Plasmids construction

The siRNAs used for cell transfection were designed and synthesized by RiboBio Co., Ltd. (Guangzhou, China), and the overexpression plasmid was synthesized by GenePharma Co., Ltd. (Suzhou, China) using the pcDNA3.1(+) vector. The mimic-NC, miRNA mimics, inhibitor-NC, and miRNA inhibitors used for cell transfection in this study were synthesized by RiboBio Co., Ltd. (Guangzhou, China), and cell transfection was conducted using the Lipofectamine™ 3000 transfection reagent (Invitrogen, USA).

RNA isolation and quantitative real‑time polymerase chain reaction (RT‑qPCR)

Total RNA was extracted from the cells using the TRIzol Reagent (Invitrogen, USA) according to the manufacturer’s protocol. Total RNA from each sample was quantified using a NanoDrop 2000 ultra-micro spectrophotometer (Thermo Fisher Scientific, USA). For circRNA and mRNA quantification, RNA was reverse-transcribed into complementary DNA (cDNA) using a PrimeScript RT reagent kit (Takara, Japan) according to the manufacturer’s protocol. Real-time quantitative PCR reactions was performed on a real-time PCR system (LightCycler480, Roche, Switzerland) using TB Green Premix Ex Taq II (Takara, Japan). The primer sequences for circRNAs and mRNAs used in the experiment are shown in Table 1. RT-qPCR was performed using a miDETECT A Track miRNA qRT-PCR Starter Kit (RiboBio, Guangzhou, China). The miRNA qPCR primers were purchased from Guangzhou RiboBio Corporation. β-tublin was used as a reference gene for circRNAs and mRNAs, and U6 was used as an reference gene for miRNAs. The 2−△△CT value reflects the relative expression level of RNAs.

Table 1 Primers designed for qRT-PCR analysis of circrna and mRNAs.
Full size table

Western blotting

RIPA lysis buffer was used to extract proteins from the cells. Protein concentration was quantified using a BCA Protein Assay Kit (Beyotime, Shanghai, China). A total of 20 µg of protein was loaded for electrophoresis on gels of varying concentrations, depending on the target protein’s molecular weight, and then transferred to a PVDF membrane (Millipore, USA). After blocking in 5% BSA for 2 h at 25 ℃, the membranes were incubated with primary antibodies (specific for p-IκBα, IκBα, p-p65, p65, caspase-3, cleaved caspase-3, ZBTB20, and β-tublin) overnight at 4 ℃. Primary antibodies were detected using horseradish peroxidase-conjugated secondary antibodies (SA00001-1 or SA00001-2; Proteintech, USA). Finally, the ECL chemiluminescence agent (Thermo Fisher Scientific, USA) was used to visualize the protein bands, which were analyzed and quantified using Image J software.

Flow cytometry analysis

The apoptotic rate of cells was determined 48 h after transfection. The cells were centrifuged and resuspended in phosphate-buffered saline (PBS) for flow cytometry. The apoptosis rate of the cells was determined using an Annexin V-FITC/PI apoptosis kit (E-CK-A211; Elabscience, Wuhan, China) according to the manufacturer’s instructions. The collected cells were incubated with Annexin V and PI staining solutions for 15 min at 25 ℃ in the dark, and then analyzed by flow cytometry. Cells double-stained with Annexin V and PI were considered apoptotic.

Phagocytosis assay

Fluorescent microspheres (1%) were added to the cell culture medium containing 10% FBS and incubated at 37 ℃ in the dark for 30 min for preconditioning. A total of 1.0 × 106 THP-1 cells were seeded into 6-well plates with transparent bottoms, and then the preconditioned fluorescent microspheres were added to each well and incubated at 37 ℃ in the dark for 30 min. Phagocytosis of the fluorescent microspheres was then observed and recorded using a fluorescence microscope (Olympus, Japan).

ELISA

The expression levels of inflammatory cytokines TNF-α, IL-6, and IL-1β in the cell culture supernatant were detected using ELISA detection kits (KE00154, KE00139, KE00021; Proteintech, Wuhan, China), and ELISA detection was excuted according to the manufacturer’s instructions.

Luciferase assay

TargetScan was used to predict the miRNA binding sites of circRNAs and mRNA. Target sequences containing the binding sites and their mutants were cloned into the psi-check2 plasmid vector and transfected into 293T cells. Mimic-NC/miRNA mimics were co-transfected into 293T cells. Firefly (RLU1) and Renilla (RLU2) luciferase fluorescence intensities were detected using the Dual-Luciferase Reporter Assay System (1910; Promega, USA). Renilla luciferase was used as the main reporter gene, and firefly luciferase as the internal reference reporter gene. Luciferase activity was calculated as the ratio of RLU2/RLU1.

Statistics

ImageJ software was used to quantitatively detect the gray scale of the western blot band and the fluorescence intensity of the fluorescent microsphere phagocytosis experiment. Graphpad Prism 9 software was used for statistical analysis and graphing, and the Biorender website was used for graphing. Each experiment was repeated at least three times. Student’s t-test was used to analyze the differences between the two independent samples, one-way/two-way ANOVA were performed to compare multiple groups, and the results were expressed as mean ± standard deviation, with P < 0.05 considered statistically significant.