Administration of a combination of COX-2/TGF-β1 siRNAs induces hypertrophic scar fibroblast apoptosis through a TP53 mediated caspase pathway Authors: Rao Fu, Sizheng Zhou, Chuanqi Liu, Jia Zhou, Qingfeng Li Published: 2024 ABSTRACT Hypertrophic scar (HTS) formation is a pathological fibrotic skin disease, with no satisfactory treatments available currently. Inducing apoptosis of HTS-derived fibroblasts (HSFs) are becoming promising approaches. In this research, we aim to improve the technology with co-delivery COX-2 and TGF-β1 siRNAs and further investigate the underlying mechanism. Firstly, the HSFs were transfected with 1 µg/ml COX-2 and/or TGF-β1 siRNAs, and proved that the apoptosis of HSFs was greater induced by COX-2/TGF-β1 siRNAs than either COX-2 or TGF-β1 siRNA alone by flow cytometry. To investigate the impact of co-silencing TGF-β1 and COX-2 mRNA expression in vivo, we established HTSs model in rat tails. Our results confirmed that co-silencing of TGF-β1 and COX-2 mRNA expression could significantly alleviate the HTS formation in vivo. Furthermore, we explored the potential molecular mechanism and revealed that the protein levels of TP53, Bcl-2 and Caspase-3 were downregulated while Bax and Cleaved Caspase-3 were upregulated in the COX-2/TGF-β1 siRNA groups compared with HKP group. Taken together, our results demonstrated that simultaneous silencing of COX-2 and TGF-β1 expression by siRNAs induced HSF apoptosis through a TP53 mediated caspase pathway. Therefore, COX-2/TGF-β1 siRNAs might serve as a novel and effective therapeutic alternative for HTSs treatments.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-024-77756-1. Introduction Hypertrophic scars (HTSs) are outcomes of the pathological wound healing after dermal injuries caused by trauma, burn, surgery, and infection1. Individuals with HTSs suffer from itching and functional disability, which seriously impact quality of life and cause substantial financial burdens 2. HTSs are often characterized by the accumulation of inflammatory cells, excessive proliferation of myofibroblasts and deposition of extracellular matrix (ECM), mainly including collagen type I (Col I) and collagen type III (Col III)3,4. Current treatments for HTSs include compression therapy, steroids, surgery, radiation, 5-fluorouracil (5-FU) and silicone-based camouflage, but the efficacy is limited, and the recurrence rate is high5,6. Although the specific molecular pathway underlying the formation of HTSs is still unclear, previous studies indicate that excessive proliferation and decreased apoptosis of myofibroblasts are related to HTS formation7,8. Therefore, strategies targeting the promotion of myofibroblast apoptosis might be effective ways to treat HTS. It has been reported that the TGF family plays a critical role in regulating cell proliferation and promoting the formation of ECM9. Among the members of this family, TGF-β1 is a principal isoform in humans, and it plays dual roles in inhibiting and promoting cell growth10. HSFs have been shown to have elevated expression of TGF-β1, which leads to a positive feedback loop that further exacerbates fibrosis, resulting in HTS formation11. Therefore, targeting TGF-β1 through different pathways can potentially prevent HTS formation12. In addition, Cyclooxygenase (COX) is a membrane-associated enzyme located in the endoplasmic reticulum or nuclear membrane that can be further classified into two isoforms, cyclooxygenase-1 (COX-1) and COX-213. COX-1 is constitutively expressed in almost all tissues and cells and plays an important “housekeeping” role to maintain regular cellular physiology14. However, COX-2 promotes the production of prostaglandins (PGs), which are involved in inflammatory responses, in response to several stimuli, such as IL-1β and TGF-β. COX-2 can regulate the early inflammatory response, and its persistent production can result in chronic wound healing after injury15. Thus, the COX-2 targeting therapy may provide new insight into the control of inflammation during HTS formation13. COX-2 is an inducible enzyme that is upregulated during inflammation and injury, catalyzing the conversion of arachidonic acid to prostaglandins16. Among these, prostaglandin E2 (PGE2) is particularly significant in the context of wound healing and fibrosis. Studies have shown that elevated COX-2 expression in fibroblasts and keratinocytes leads to increased levels of PGE2, which can modulate various cellular processes involved in scar formation, including fibroblast proliferation, migration, and extracellular matrix synthesis17. Although RNA interference technology with siRNA has been applied widely in disease treatments in recent years18, their simultaneous delivery has not been previously explored in HTS treatment to our knowledge. Development of effective and safe siRNA formulation is critical for the siRNA therapies. Compared to small-molecule inhibitors, siRNA offers a broader range of applications. While small-molecule inhibitors typically target to limited, well-defined therapeutic binding sites, siRNA can address diseases previously deemed “undruggable” and can be employed in combination to achieve synergistic effects. Additionally, the prolonged use of small-molecule inhibitors often leads to the development of drug resistance19. In contrast, siRNA can minimize the risk of drug resistance because it directly degrades target mRNA, preventing the synthesis of target proteins20. Our previous research indicated that using a cationic carrier to package both COX-2 and TGF-β1 siRNAs can ameliorate human HTS formation through the induction of fibroblast apoptosis, although the specific mechanism is still unclear21. Here, we conducted a continued research to explore the potential molecular mechanism underlying this induction of apoptosis. Based on our previous study and literature, we hypothesized that dual-targeting siRNA therapeutics can inhibit the proliferation, migration and invasion of human HSFs. In this study, for the first time, we report that treatment with siRNA targeting the COX-2 and TGF-β1 genes successfully induced HSF apoptosis through a TP53 mediated caspase pathway, and alleviated HTS formation in vivo. Results Silencing TGF-β1 and COX-2 expression downregulated the mRNA and protein levels of Col I, Col III, and α-SMA in HSFs in vitro We transfected human HSFs with siRNA sequences that have previously been proven efficacy21. When the HSFs were transfected with selected siRNAs targeting either TGF-β1 or COX-2 individually or in combination, we observed that the siRNAs significantly knocked down the expression of the corresponding target gene (Fig. 1A). Notably, downregulation of COX-2 expression was also observed in TGF-β1 siRNA-transfected cells (Fig. 1A). To investigate the effect of silencing TGF-β1 and/or COX-2 expression on the expression of fibrosis-related molecules in HSFs, the mRNA and protein levels of Col I, Col III, and α-SMA were assessed by qRT–PCR and Western blotting. The results showed that silencing TGF-β1 or COX-2 gene expression could downregulate both the mRNA and protein levels of Col I and Col III (Fig. 1A, B). Furthermore, the combination of the COX-2/TGF-β1 siRNAs resulted further inhibition (P < 0.01, Fig. 1A, B, Supplementary Fig. 1), indicating a significant dual effect. In the cells transfected with the COX-2/TGF-β1 siRNAs, the mRNA and protein expression of α-SMA was downregulated, but not in the COX-2 siRNA-transfected cells (Fig. 1A, B, Supplementary Fig. 1). Immunocytofluorescence staining also verified that silencing the expression of COX-2 and TGF-β1 markedly downregulated the expression of α-SMA (Fig. 1C). Moreover, the HSFs presented a much lower cell density and a narrow morphology in the COX-2/TGF-β1 siRNA-transfected group (Fig. 1C). Fig. 1Effects of silencing TGF-β1 and/or COX-2 mRNA expression on the expression of fibrosis-related molecules. (A) qRT–PCR analysis of the mRNA levels of PTGS2, TGF-β1, ACTA2, Col I and Col III in HSFs after transfection with HKP-TGF-β1 and/or HKP-COX-2 siRNAs for 48 h (1 µg/ml). β-actin served as the control. (B)Left, Western blotting analysis of the protein levels of COX-2, TGF-β1, α-SMA, Col I, and Col III in HSFs after transfection with HKP-TGF-β1 and/or HKP-COX-2 siRNAs for 48 h (1 µg/ml). β-actin served as the control. Right, Quantitation of fibrosis-related protein levels by densitometric analysis using ImageJ (n = 3; compared with HKP-treated HSFs). (C) Immunofluorescence staining of α-SMA in HSFs transfected with TGF-β1 and/or COX-2 siRNAs for 48 h (1 µg/ml). α-SMA is indicated by green fluorescence. Nuclei were stained with DAPI and are indicated by blue fluorescence. Silencing TGF-β1 expression alone and in combination with silencing COX-2 expression markedly downregulated the expression of α-SMA, and the HSFs presented a much lower cell density and a narrow morphology. Scale bar = 100 μm. Each bar represents the mean ± SD. (n = 3). NC: Negative control (nontargeting siRNA). HKP: Vehicle control without scrambled siRNA (HKP alone). *P < 0.05, **P < 0.01, ***P < 0.001, each group compared to the HKP group. The gels showed in B were cropped, original gels were presented in Supplementary Fig. 1. Effects of silencing TGF-β1 and/or COX-2 mRNA expression on the expression of fibrosis-related molecules. (A) qRT–PCR analysis of the mRNA levels of PTGS2, TGF-β1, ACTA2, Col I and Col III in HSFs after transfection with HKP-TGF-β1 and/or HKP-COX-2 siRNAs for 48 h (1 µg/ml). β-actin served as the control. (B)Left, Western blotting analysis of the protein levels of COX-2, TGF-β1, α-SMA, Col I, and Col III in HSFs after transfection with HKP-TGF-β1 and/or HKP-COX-2 siRNAs for 48 h (1 µg/ml). β-actin served as the control. Right, Quantitation of fibrosis-related protein levels by densitometric analysis using ImageJ (n = 3; compared with HKP-treated HSFs). (C) Immunofluorescence staining of α-SMA in HSFs transfected with TGF-β1 and/or COX-2 siRNAs for 48 h (1 µg/ml). α-SMA is indicated by green fluorescence. Nuclei were stained with DAPI and are indicated by blue fluorescence. Silencing TGF-β1 expression alone and in combination with silencing COX-2 expression markedly downregulated the expression of α-SMA, and the HSFs presented a much lower cell density and a narrow morphology. Scale bar = 100 μm. Each bar represents the mean ± SD. (n = 3). NC: Negative control (nontargeting siRNA). HKP: Vehicle control without scrambled siRNA (HKP alone). *P < 0.05, **P < 0.01, ***P < 0.001, each group compared to the HKP group. The gels showed in B were cropped, original gels were presented in Supplementary Fig. 1. Simultaneous silencing of TGF-β1 and COX-2 expression affected the proliferation of HSFs in vitro A CCK-8 assay was performed to analyze the influence of silencing TGF-β1 and/or COX-2 mRNA expression on HSF proliferation. As shown by proliferation curves, silencing TGF-β1 or COX-2 mRNA expression alone inhibited the proliferation of HSFs in a dose-dependent manner (2 µg/ml to 8 µg/ml siRNA, Fig. 2A). However, when TGF-β1 and COX-2 mRNA expression was simultaneously silenced, the inhibitory effect was observed even when the cells were treated with 0.5 µg/ml siRNA (Fig. 2A). Due to the strong inhibitory effect on cell proliferation, we chose a concentration of 1 µg/ml to investigate the effect of silencing TGF-β1 and/or COX-2 mRNA expression in the following experiments. These results were also further supported by the EdU assay of HSFs. The COX-2/TGF-β1 siRNA-transfected cells exhibited a more evident effect on proliferation inhibition (Fig. 2B). A highly significant decrease in the numbers of EdU-positive cells was observed in COX-2/TGF-β1 siRNA-transfected groups compared to the HKP and NC groups, whereas the COX-2 siRNA- and TGF-β1siRNA-treated group showed no significant difference (Fig. 2C, P < 0.01). Fig. 2Silencing TGF-β1 and COX-2 mRNA expression profoundly inhibited the proliferation of HSFs. (A) Proliferation curves of HSFs after transfection with 0, 0.5, 1, 2, 4–8 µg/ml TGF-β1 and/or COX-2 siRNAs for 48 h as determined by CCK-8 assays. Silencing TGF-β1 or COX-2 mRNA expression inhibited the proliferation of HSFs in a dose-dependent manner, and the combination of COX-2/TGF-β1 mRNAs further inhibited proliferation. (n = 3; ***P < 0.001, the COX-2/TGF-β1 siRNA group compared with the HKP group; ^^^P < 0.001, the COX-2 siRNA group compared with the HKP group; ###P < 0.001, the TGF-β1 siRNA group compared with the HKP group.) (B) Representative images of the EdU assay after transfection with 1 µg/ml TGF-β1 and/or COX-2 siRNAs for 48 h. The number of EdU-positive cells in TGF-β1- and/or COX-2-knockdown HSFs was markedly lower than that in the HKP and NC groups. Scale bar = 100 μm. (C) Quantification of the EdU-positive cells in (B). The data are presented as the mean ± SD. (n = 3; **P<0.01). Silencing TGF-β1 and COX-2 mRNA expression profoundly inhibited the proliferation of HSFs. (A) Proliferation curves of HSFs after transfection with 0, 0.5, 1, 2, 4–8 µg/ml TGF-β1 and/or COX-2 siRNAs for 48 h as determined by CCK-8 assays. Silencing TGF-β1 or COX-2 mRNA expression inhibited the proliferation of HSFs in a dose-dependent manner, and the combination of COX-2/TGF-β1 mRNAs further inhibited proliferation. (n = 3; ***P < 0.001, the COX-2/TGF-β1 siRNA group compared with the HKP group; ^^^P < 0.001, the COX-2 siRNA group compared with the HKP group; ###P < 0.001, the TGF-β1 siRNA group compared with the HKP group.) (B) Representative images of the EdU assay after transfection with 1 µg/ml TGF-β1 and/or COX-2 siRNAs for 48 h. The number of EdU-positive cells in TGF-β1- and/or COX-2-knockdown HSFs was markedly lower than that in the HKP and NC groups. Scale bar = 100 μm. (C) Quantification of the EdU-positive cells in (B). The data are presented as the mean ± SD. (n = 3; **P<0.01). Co-silencing of TGF-β1 and COX-2 expression inhibited the migration and invasion abilities and induced the apoptosis of HSFs in vitro To investigate the migration ability of the mRNA-knockdown HSFs, we performed wound healing and Transwell assays. Figure 3A shows that the wound widths of the TGF-β1-knockdown HSFs and the COX-2/TGF-β1 dual- knockdown HSFs were wider at 12, 24 and 36 h than those of the HKP and NC HSFs. The quantitative analysis of the percentage of the wound healing area in each group also showed a significant decrease when COX-2/TGF-β1 mRNA expression was silenced compared to the COX-2 knockdown cells, especially at 24 h (P < 0.01; Fig. 3A); these results indicated that the migration ability was significantly inhibited by co-silencing COX-2/TGF-β1 mRNA expression. Furthermore, crystal violet staining demonstrated that fewer HSFs migrated after transfection with TGF-β1 siRNA alone or with COX-2/TGF-β1 siRNAs, and the relative numbers of invading cells were significantly decreased compared with that in the HKP group (P < 0.001; Fig. 3B). Notably, treatment with the COX-2/TGF-β1 siRNA combination resulted in a significant decrease in invasion compared to treatment with single siRNA alone (P < 0.01; Fig. 3B). In addition, flow cytometry analysis were performed to evaluate the ability of siRNA to induce HSF apoptosis. The apoptosis rate of HSFs was increased when transfected with corresponding siRNA compared to treatment with HKP and NC, and a further increase in apoptosis was observed after COX-2/TGF-β1 siRNA transfection (Fig. 3C). Furthermore, the dual-konckdown group resulted in a more significant increase in apoptosis compared to that of single siRNA transfection (P < 0.05, Fig. 3D). However, when double the dosage of single siRNA-transfected group, the apoptosis rate of HSFs was not increased, indicating that COX-2/TGF-β1 siRNAs showed significant greater dual-effect on inducing apoptosis (Supplementary Fig. 2A-C). Fig. 3Simultaneously silencing TGF-β1 and COX-2 mRNA expression significantly affected the migration ability and induced the apoptosis of HSFs. (A)Left, Representative images and quantification of the wound healing in HSFs after transfection with 1 µg/ml TGF-β1 and/or COX-2 siRNAs for 48 h. Wound healing was markedly inhibited when TGF-β1 and COX-2 expression was simultaneously silenced. Right, Quantitative analysis of the cell migration area after 12 and 24 h (n = 3). Scale bar = 500 μm. (B) Representative images of the Transwell assay with HSFs after transfection with 1 µg/ml TGF-β1 and/or COX-2 siRNAs for 24 h. Inhibition the expression of both TGF-β1 and COX-2 remarkedly inhibited the migration ability of HSFs. Scale bar = 200 μm. (C) Flow cytometry analysis of apoptosis in HSFs after transfection with 1 µg/ml TGF-β1 and/or COX-2 siRNAs for 48 h. Apoptotic activity was profoundly induced when TGF-β1 and COX-2 expression was both silenced. (D) Quantification of the cell apoptosis ratio in (C). The data are presented as the mean ± SD. (n = 3); *P<0.05, **P<0.01 and ***P<0.001. Simultaneously silencing TGF-β1 and COX-2 mRNA expression significantly affected the migration ability and induced the apoptosis of HSFs. (A)Left, Representative images and quantification of the wound healing in HSFs after transfection with 1 µg/ml TGF-β1 and/or COX-2 siRNAs for 48 h. Wound healing was markedly inhibited when TGF-β1 and COX-2 expression was simultaneously silenced. Right, Quantitative analysis of the cell migration area after 12 and 24 h (n = 3). Scale bar = 500 μm. (B) Representative images of the Transwell assay with HSFs after transfection with 1 µg/ml TGF-β1 and/or COX-2 siRNAs for 24 h. Inhibition the expression of both TGF-β1 and COX-2 remarkedly inhibited the migration ability of HSFs. Scale bar = 200 μm. (C) Flow cytometry analysis of apoptosis in HSFs after transfection with 1 µg/ml TGF-β1 and/or COX-2 siRNAs for 48 h. Apoptotic activity was profoundly induced when TGF-β1 and COX-2 expression was both silenced. (D) Quantification of the cell apoptosis ratio in (C). The data are presented as the mean ± SD. (n = 3); *P<0.05, **P<0.01 and ***P<0.001. Inhibition of both TGF-β1 and COX-2 mRNA expression alleviated HTS formation in vivo To investigate the impact of simultaneously silencing TGF-β1 and COX-2 mRNA expression in vivo, we established HTSs model in rat tails (Supplementary Fig. 3A-B) and treated the HTSs with the indicated siRNAs via injection. Fifteen days after treatment, the HTSs in the COX-2/TGF-β1 siRNA group were significantly ameliorated compared to those in the NC group and single siRNA groups based on a gross analysis (Fig. 4A). The quantitative analysis showed that the scar area was significantly reduced in the COX-2/TGF-β1 siRNA group compared to the single siRNA groups and NC group (Fig. 4B, P < 0.05). H&E staining was performed on rat tail scar samples, which revealed numerous fibroblasts and inflammatory cells in the NC group, along with abundant coarsely arranged collagen fibers (Fig. 4C, above; Supplementary Fig. 4A-B). However, after treated with HKP-COX-2/TGF-β1 siRNA, the collagen fibers were more neatly arranged, even compared to the single siRNA treatment group (Fig. 4C, below; Supplementary Fig. 4A-B, the H&E staining of normal rat tail was provided in Supplementary Fig. 4C). We then performed SEI analysis, and the COX-2/TGF-β1 siRNA group demonstrated the lowest SEI, which was close to 1 and consistent with the overall gross appearance (P < 0.05; Fig. 4D). To evaluate the effect of COX-2/TGF-β1 siRNAs on apoptosis in HTS tissue in vivo, we performed a TUNEL assay on the HTS samples. TUNEL staining showed that the number of positively stained cells in the COX-2/TGF-β1 siRNA group was obviously increased compared to that in the other three groups, indicating that simultaneous silencing of COX-2/TGF-β1 siRNA expression could induce apoptosis in HTS tissues in vivo and alleviate the HTSs formation (P < 0.05; Fig. 4E, F; Supplementary Fig. 5A-C). Fig. 4Simultaneously silencing TGF-β1 and COX-2 mRNA expression inhibited HTS formation in rat tails. (A) Representative photographs of HTS formation on rat tails after treatment with the indicated siRNAs for 12 weeks. (B) Quantification of the scar area in (A) using ImageJ (n = 6). (C) H&E staining of scar tissue at 12 weeks after the indicated siRNA treatments. Black arrows indicate scar thickness, while red arrows indicate normal skin thickness. Scale bar = 200 μm. (D) The quantification of scar elevation in (C) by ImageJ and the scar elevation index was significantly lower in the COX-2/TGF-β1 siRNA group than in the NC and single-silenced groups. (E) TUNEL analysis of scar tissue at 12 weeks after the indicated siRNA treatments. Black arrows indicate TUNEL-positive cells. Scale bar = 20 μm. (F) Quantification of TUNEL-positive cells in (E). The data are presented as the mean ± SD. (n = 6); *P<0.05, **P<0.01 and ***P<0.001. Simultaneously silencing TGF-β1 and COX-2 mRNA expression inhibited HTS formation in rat tails. (A) Representative photographs of HTS formation on rat tails after treatment with the indicated siRNAs for 12 weeks. (B) Quantification of the scar area in (A) using ImageJ (n = 6). (C) H&E staining of scar tissue at 12 weeks after the indicated siRNA treatments. Black arrows indicate scar thickness, while red arrows indicate normal skin thickness. Scale bar = 200 μm. (D) The quantification of scar elevation in (C) by ImageJ and the scar elevation index was significantly lower in the COX-2/TGF-β1 siRNA group than in the NC and single-silenced groups. (E) TUNEL analysis of scar tissue at 12 weeks after the indicated siRNA treatments. Black arrows indicate TUNEL-positive cells. Scale bar = 20 μm. (F) Quantification of TUNEL-positive cells in (E). The data are presented as the mean ± SD. (n = 6); *P<0.05, **P<0.01 and ***P<0.001. Simultaneous silencing of TGF-β1 and COX-2 expression induced the apoptosis of HSFs through caspase pathway As we proved above, simultaneous silencing of TGF-β1 and COX-2 mRNA expression caused a large amount of cell apoptosis both in vitro and in vivo. Therefore, we firstly investigated the apoptosis-related molecular pathway that was affected in vitro in HSFs. Analysis of the expression of TP53, Bcl-2, Bax and Caspase-3 was performed by qRT–PCR and Western blotting. According to the Western blotting results, simultaneously silencing TGF-β1 and COX-2 expression significantly increased the expression of cleaved Caspase-3, compared with other treatments (Fig. 5A, Supplementary Fig. 6). Moreover, the levels of the apoptosis inhibitors TP53 and Bcl-2 were decreased (P < 0.05; Fig. 5A, Supplementary Fig. 6). Although the expression levels of TP53 were also decreased by silencing TGF-β1 expression, the expression of cleaved Caspase-3/Caspase 3 was not statistically significant (Fig. 5A, B, Supplementary Fig. 6). Notably, the expression of the apoptotic factor Bax was markedly increased in the COX-2/TGF-β1 siRNA group compared with the single siRNA groups (P < 0.001; Fig. 5A, B, Supplementary Fig. 6). qRT–PCR analysis yielded similar results: the expression of TP53 and Bcl-2 was significantly downregulated, and the expression of the apoptotic factors Bax and Caspase-3 was upregulated in the COX-2/TGF-β1 siRNA-transfected group (P < 0.05; Fig. 5C). Therefore, we speculated that simultaneous silencing of COX-2 and TGF-β1expression induced apoptosis in HSFs through the p53-mediated caspase cascade, as illustrated in Fig. 5D. Fig. 5Simultaneously silencing TGF-β1 and COX-2 mRNA expression induces apoptosis through caspase pathway. (A) Western blotting results of TP53, Bcl-2, Bax, Caspase-3 and Cleaved Caspase-3 expression in HSFs. The protein expression of Cleaved Caspase-3 and Bax was markedly increased when TGF-β1 and COX-2 expression was simultaneously silenced compared to the other groups, while the expression of TP53, Bcl-2 and Caspase-3 were decreased. β-actin served as the control. (B) Quantitation of apoptosis-related protein levels by densitometric analysis using ImageJ (n = 3; compared with HKP-treated HSFs). (C) qRT–PCR results of TP53, Bcl-2, Bax and CASP3 expression in HSFs. Simultaneous silencing of TGF-β1 and COX-2 expression significantly increased the expression of Caspase-3 and Bax, which are components of the apoptosis pathway, compared to HKP and NC. The expression of the apoptosis inhibitors TP53 and Bcl-2 was decreased when TGF-β1 and COX-2 expression was simultaneously silenced. (D) Diagram of the molecular pathway demonstrating the possible mechanism of apoptosis induction by the simultaneous silencing of TGF-β1 and COX-2 expression. The data are presented as the mean ± SD. (n = 3); *P<0.05, **P<0.01 and ***P<0.001. The gels showed in A were cropped, original gels were presented in Supplementary Fig. 6. Simultaneously silencing TGF-β1 and COX-2 mRNA expression induces apoptosis through caspase pathway. (A) Western blotting results of TP53, Bcl-2, Bax, Caspase-3 and Cleaved Caspase-3 expression in HSFs. The protein expression of Cleaved Caspase-3 and Bax was markedly increased when TGF-β1 and COX-2 expression was simultaneously silenced compared to the other groups, while the expression of TP53, Bcl-2 and Caspase-3 were decreased. β-actin served as the control. (B) Quantitation of apoptosis-related protein levels by densitometric analysis using ImageJ (n = 3; compared with HKP-treated HSFs). (C) qRT–PCR results of TP53, Bcl-2, Bax and CASP3 expression in HSFs. Simultaneous silencing of TGF-β1 and COX-2 expression significantly increased the expression of Caspase-3 and Bax, which are components of the apoptosis pathway, compared to HKP and NC. The expression of the apoptosis inhibitors TP53 and Bcl-2 was decreased when TGF-β1 and COX-2 expression was simultaneously silenced. (D) Diagram of the molecular pathway demonstrating the possible mechanism of apoptosis induction by the simultaneous silencing of TGF-β1 and COX-2 expression. The data are presented as the mean ± SD. (n = 3); *P<0.05, **P<0.01 and ***P<0.001. The gels showed in A were cropped, original gels were presented in Supplementary Fig. 6. We further assessed whether the caspase pathway was activated in the in vivo HTS samples by qRT–PCR. After the injection of COX-2 siRNA, the expression levels of both COX-2 and TGF-β1 were significantly downregulated compared to those in the NC group (Fig. 6A). Consistent results were also observed in the single TGF-β1 or COX-2 siRNA injection group, indicating the successful knockout of the corresponding gene expression in vivo (Fig. 6B). Notably, the expression of the apoptosis inhibitor TP53 was significantly downregulated, and the expression of the apoptosis-promoting genes Caspase-3 and Caspase-9 was upregulated in the COX-2/TGF-β1 siRNA group (Fig. 6C-E). These results supported the increased numbers of apoptotic cells observed in TUNEL staining, and were consistent with the in vitro results (Fig. 4E). Although the expression of TP53 was also decreased in the COX-2 siRNA and TGF-β1 siRNA injection groups compared to the NC group, increased expression of either Caspase-3 or Caspase-9 was not observed. Fig. 6Simultaneously silencing TGF-β1 and COX-2 expression upregulated the expression of apoptosis pathway in vivo. (A-E) qRT–PCR results of COX-2, TGF-β1, TP53, Caspase-3 and Caspase-9 expression in the HTSs of rat tails. The expression of COX-2 and TGF-β1 was inhibited, while TP53, Caspase-3 and Caspase-9 expression was consistent with the in vitro results. The data are presented as the mean ± SD. (n = 3); *P<0.05, **P<0.01 and ***P<0.001. Simultaneously silencing TGF-β1 and COX-2 expression upregulated the expression of apoptosis pathway in vivo. (A-E) qRT–PCR results of COX-2, TGF-β1, TP53, Caspase-3 and Caspase-9 expression in the HTSs of rat tails. The expression of COX-2 and TGF-β1 was inhibited, while TP53, Caspase-3 and Caspase-9 expression was consistent with the in vitro results. The data are presented as the mean ± SD. (n = 3); *P<0.05, **P<0.01 and ***P<0.001. Moreover, the protein levels of PI3K and p-AKT were found to be decreased in the COX-2/TGF-β1 siRNA group, although the downregulation of PI3K had no statistical significance (Fig. 7A, B, Supplementary Fig. 7). In addition, we also observed that the expression of downstream molecule of AKT, namely MEK and ERK was significantly downregulated in the COX-2/TGF-β1 siRNA group, whereas the expression of MEK and ERK was slightly affected in the single siRNA groups (Fig. 7A, B, Supplementary Fig. 7). These results indicated that the MEK/ERK pathway might be activated and contribute to the induced apoptosis when TGF-β1 and COX-2 expression levels were simultaneously silenced. Fig. 7Simultaneously silencing TGF-β1 and COX-2 mRNA expression inhibited the MEK/ERK pathway. (A) The protein levels of PI3K, p-AKT p-MEK and p-ERK1/2 were significantly downregulated when TGF-β1 and COX-2 expression was simultaneously silenced in vitro, while this effect was not observed in the single-knockdown groups. (B) Quantitation of protein levels in (A) by densitometric analysis using ImageJ (n = 3; compared with HKP-treated HSFs). The data are presented as the mean ± SD. (n = 3); *P<0.05, **P<0.01 and ***P<0.001. The gels showed in A were cropped, original gels were presented in Supplementary Fig. 7. Simultaneously silencing TGF-β1 and COX-2 mRNA expression inhibited the MEK/ERK pathway. (A) The protein levels of PI3K, p-AKT p-MEK and p-ERK1/2 were significantly downregulated when TGF-β1 and COX-2 expression was simultaneously silenced in vitro, while this effect was not observed in the single-knockdown groups. (B) Quantitation of protein levels in (A) by densitometric analysis using ImageJ (n = 3; compared with HKP-treated HSFs). The data are presented as the mean ± SD. (n = 3); *P<0.05, **P<0.01 and ***P<0.001. The gels showed in A were cropped, original gels were presented in Supplementary Fig. 7. Discussion Hypertrophic scar tissue (HTS) formation is generally attributed to reduced apoptosis of myofibroblasts and excessive collagen accumulation, leading to both cosmetic concerns and reduced quality of life for patients22,23. Despite the need for effective treatments, current therapeutic strategies for HTS remain limited, often yielding unsatisfactory results. HTS is frequently characterized by heightened inflammation, excessive collagen accumulation in the extracellular matrix (ECM), and fibroblast hyperproliferation. These pathological features are commonly linked to the overexpression of COX-2 and TGF-β1, which play crucial roles in the inflammatory and fibrotic processes underlying HTS. The overexpression of COX-2 is strongly associated with inflammation, and its inhibition has been shown to alleviate the inflammatory phase of HTS formation24. Besides, TGF-β1 is a multifunctional cytokine involved in cell cycle regulation, differentiation, and ECM accumulation25. Therefore, targeting TGF-β1 and COX-2 presents a promising therapeutic approach for addressing HTS. siRNA therapeutics offer the potential to deliver targeted treatments safely and effectively, with the capability to address a wide range of diseases26,27. Although siRNA therapy has demonstrated considerable promise and progress in translational applications for HTS treatment, outcomes have not met expectations28,29. In response, we have developed a siRNA cocktail designed to simultaneously silence the expression of TGF-β1 and COX-2 mRNAs. This approach demonstrated significant advantages in treating HTS, as the combination of two different siRNA sequences enables a more effective knockdown of undesirable proteins than single siRNAs alone, as reported in previous studies30,31. In this study, we successfully silenced TGF-β1 and COX-2 mRNA expression using siRNAs, proposing a novel strategy for HTS treatment. Our results showed a significant inhibition of the proliferation, migration, and invasion of hypertrophic scar fibroblasts (HSFs). Our findings further confirmed that simultaneous silencing of these targets induces apoptosis in HSFs through the p53-mediated caspase cascade, as illustrated in Fig. 5D. To our knowledge, there are no existing reports on the co-delivery of siRNAs targeting TGF-β1 and COX-2 for HTS treatment. Apoptosis, or programmed cell death, is a critical mechanism that ensures the controlled elimination of cells with minimal impact on surrounding tissues32. This process involves two main pathways: intrinsic and extrinsic, which work together to remove defective cells from the body33. The intrinsic pathway, also known as mitochondrial outer membrane permeabilization (MOMP), usually leads to caspase activation and acts as one of the decisive steps in apoptotic cell death34. To elucidate the apoptotic mechanisms triggered by the simultaneous silencing of TGF-β1 and COX-2 mRNA expression, we analyzed the expression levels of executioner caspase-3 both in vitro and in vivo, along with genes related to MOMP. The Bcl-2 family of proteins is vital in regulating apoptosis, with members like Bax and Bak playing essential roles in MOMP and the release of cytochrome c, which activates the caspase cascade35. Our findings revealed that the expression of Caspase-3 and Bax, key components of the apoptosis pathway, was significantly upregulated in the dual-knockdown group compared to groups transfected with single siRNAs. Conversely, the levels of apoptosis inhibitors TP53 and Bcl-2 were markedly downregulated upon the co-silencing of TGF-β1 and COX-2. This suggested that the simultaneous silencing of these genes induces apoptosis via the p53-mediated caspase pathway, highlighting the potential effectiveness of this strategy in promoting cell death in target tissues. TGF-β1 is a pivotal signaling molecule that significantly influences various cellular biological functions, also known as a potent inducer of fibroblast differentiation into myofibroblasts and plays a crucial role in modulating key cellular processes such as proliferation, migration, and apoptosis36. However, we observed that downregulating TGF-β1 expression did not result in the expected increase in the expression of pro-apoptotic factors BAX and cleaved-Caspase-3. There are few reports on TGF-β1 inducing fibroblast apoptosis in hypertrophic scars. However, through literature review, we found that in certain cell populations, TGF-β1 may exert its apoptotic effects through alternative pathways rather than the classical caspase-mediated mechanism, such as but the upstream or downstream pathways of caspase pathways37. Moreover, our investigation revealed that knocking down COX-2 also did not significantly upregulate BAX and cleaved-Caspase-3 expression. As a critical component in prostaglandin E2 (PGE2) synthesis, COX-2 plays an essential role in hypertrophic scar (HTS) formation through the COX-2/PGE2 cascade38. Prostaglandins act in a paracrine manner, influencing both the cells that produce them and neighboring cells, and can directly induce pro-inflammatory changes39. PGE2 has been demonstrated to stimulate the expression of Bcl-240, and our study corroborated this by showing a downregulation of Bcl-2 in the COX-2 siRNA group. Interestingly, the upregulation of BAX and cleaved-Caspase-3 proteins was only observed in the dual-knockdown group, indicating that the concurrent suppression of COX-2 and TGF-β1 is necessary to trigger significant apoptotic responses. While upregulation of COX-2 may promote cell migration and proliferation, contributing to increased scar tissue formation, its downregulation may not directly affect cell survival. Instead, the impact on apoptosis is mediated through PGE2, highlighting a complex regulatory role that warrants further investigation17. The pronounced apoptosis of HSFs following the simultaneous downregulation of COX-2 and TGF-β1 pathways may indicate the involvement of distinct, yet potentially overlapping, signaling pathways in the regulation of these genes. Existing studies have reported that PGE2 signaling can vary significantly depending on the presence or absence of TGF-β1, and similarly, the cellular environment can influence PGE2 signaling dynamics. Additionally, it is plausible that the distribution of transcriptional coactivators might differ with or without PGE2, further influencing cellular responses41. While we observed substantial fibroblast apoptosis in the dual-knockdown group, increasing the dose of siRNA in the single knockdown groups also resulted in a higher number of apoptotic cells. Nevertheless, the proportion of apoptotic cells remained significantly greater in the dual-knockdown group. This indicates that the concurrent knockdown of COX-2 and TGF-β1 exerted a synergistic effect, which amplifies the induction of HSFs apoptosis. We investigated the underlying mechanisms of apoptosis induced by the co-silencing of TGF-β1 and COX-2 expression42. The phosphatidylinositol 3-kinase (PI3K)/protein kinase B (Akt) signaling pathway is associated with various biological activities, including proliferation, migration, and apoptosis43. The COX-2 and TGF-β1 pathways are intricately connected to the Akt signaling pathway, playing a significant role in cellular processes. It is documented that COX-2 may inhibit the Akt pathway via the secretion of PGE244. Furthermore, research suggested that COX-2 can modulate adhesion-dependent signaling pathways, such as STAT-3, FAK, and Akt, which are crucial in TGF-β1-induced myofibroblast differentiation, through PGE2 secretion41. This raises the possibility that, upon dual silencing of COX-2 and TGF-β1, the downstream effectors of the PI3K/Akt pathway may be activated, thereby triggering apoptosis in HSF cells45. This potential mechanism warrants further investigation to understand the complex interplay between these pathways and their impact on cellular fate. As a pathway that cross-communicates with the PI3K/Akt pathway, the MEK/ERK pathway was significantly inhibited when TGF-β1 and COX-2 mRNA expressions were silenced simultaneously. Activation of the MEK/ERK pathway leads to the accumulation of p53, promoting proliferation and inhibiting apoptosis. The MEK/ERK pathway has been implicated in the proliferation and apoptosis of fibroblasts46. Studies have shown that ERK is closely linked to various fibrotic diseases and the expression of α-SMA, resulting in excessive collagen secretion and fibroproliferative disorders47. Overall, our study revealed that simultaneously silencing TGF-β1 and COX-2 expression led to significant downregulation of MEK/ERK protein expression compared to single-knockdown groups. The inhibition of the MEK/ERK protein was further involved in activating the expression of p53, resulting in massive numbers of apoptotic cells, which is consistent with present studies in fibrotic diseases48. To further assess the effects of simultaneously silencing TGF-β1 and COX-2 mRNA expression in vivo, we utilized a cutaneous scarring model in rat tails, as described in previous studies49,50. At present, there is no ideal animal model that precisely replicates human hypertrophic scars or keloids. The most commonly employed animal models either induce fibrosis to varying extents in species that do not naturally develop fibrosis or involve the transplantation of human hypertrophic scar samples into animals51. To address these limitations, our research utilized a rat tail model, which more closely resembles the biological characteristics of human hypertrophic scars. This model provides a more accurate representation, enabling better understanding and investigation of the underlying mechanisms of hypertrophic scar formation. In this model, a 6 × 6 mm wound was created on the tail, and customized stainless steel rings were used to minimize wound contraction and induce hypertrophic scar (HTS) formation during re-epithelialization. In our current study, the in vivo findings from the tail model were consistent with in vitro results, confirming that apoptosis in hypertrophic scar fibroblasts (HSFs) was induced by activating the caspase cascade. Our data also indicated that transfection with either TGF-β1 or COX-2 siRNA alone slightly induced HSF apoptosis, though the combined COX-2/TGF-β1 siRNA treatment exhibited a stronger effect. According to our research, this is the first study to reveal that simultaneously silencing TGF-β1 and COX-2 mRNA expression induces HTSs apoptosis, through the p53-activated caspase pathway, with MEK/ERK pathway involved. The dual-targeted siRNA approach further exhibited significantly higher efficiency in silencing the expression of multiple genes simultaneously and provides a novel viewpoint for the treatment of HTS in the clinic. Materials and methods Cell culture and treatment All HTSs were obtained from patients undergoing HTS resection surgery at the Department of Plastic and Reconstructive Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine. The study protocol was approved by the ethics committee of our institute (ethics number: SH9H-2020-A126-1). A total of 6 subjects with HTSs were enrolled in this study, and the characteristics of these patients are shown in Table 1. HTS tissues were minced, immersed in 0.25% Dispase II (Roche, Switzerland), and incubated at 37 °C for 1 h. The epidermis was then separated and digested with 0.2% NB4 (Roche, Switzerland) at 37 °C for 2–3 h to isolate the HSFs. The collected HSFs were cultured in complete high-glucose Dulbecco’s modified Eagle’s medium (CM) (DMEM, Gibco, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA) and 100 U/ml penicillin and streptomycin (Gibco, USA) and incubated at 37 °C in 5% CO2. The medium was changed every 3 days. All experiments were conducted using cells from the third to sixth passages (P3-P6), pooled from six different samples to ensure consistency and reliability. The siRNA oligonucleotides and HKP (a biodegradable polypeptide cationic molecule) were provided by Sirnaomics (Suzhou, China). HSFs were seeded in the wells of 6-well plates at a density of 106 cells in 3 ml media per well. Six hours later, the media was replaced with fresh media supplemented with HKP alone, HKP (NC siRNA), HKP (COX-2/TGF-β1 siRNAs), HKP (TGF-β1 siRNA), or HKP (COX-2 siRNA) at concentrations of 4 ug/ml HKP loaded with 1 ug/ml siRNA and incubated for 48 h unless otherwise stated. Table 1The characteristics of each sample.Patient numberSexAgeBiopsy siteTreatment before surgery1Male25ArmNone2Male54NeckNone3Male35ChestNone4Female26AbdomenNone5Male15ArmNone6Male64AbdomenNone Table 1The characteristics of each sample.Patient numberSexAgeBiopsy siteTreatment before surgery1Male25ArmNone2Male54NeckNone3Male35ChestNone4Female26AbdomenNone5Male15ArmNone6Male64AbdomenNone The characteristics of each sample. Cell counting kit-8 (CCK-8) assay The proliferation of HSFs was measured by CCK-8 assay. HSFs were seeded into the wells of 96-well plates at a density of 3 × 104 cells in 100 µl media per well. After adherence, the media were replaced with fresh media supplemented with HKP alone, HKP (NC siRNA), HKP (0.5, 1, 2, 4, 8 µg/ml TGF-β1 siRNA), HKP (0.5, 1, 2, 4, 8 µg/ml COX-2 siRNA) or HKP (1, 2, 4, 8, 16 µg/ml COX-2/TGF-β1 siRNAs). Five replicates were established for each concentration. After treatment for 48 h, 10 µl CCK-8 solution (Dojindo, Japan) was added to each well and incubated for 2 h at 37 °C. Cell viability was quantified by measuring the absorbance at 450 nm with a microplate reader (Synergy H (1), Biotek, Vermont, USA). Three biological replicates were performed for the CCK-8 assay. Quantitative real-time PCR (qRT–PCR) After HSFs were transfected, total RNA was extracted by TRIzol Reagent (Invitrogen, USA). A spectrophotometer (NanoDrop2000, USA) was then used to assess the purity of the RNA. cDNA was synthesized by the reverse transcription of 1000 ng RNA according to the manufacturer’s protocols (Takara, Japan). Gene expression levels of COX-2, TGF-β1, Col I, Col III, α-SMA, TP53, Bcl-2, Bax, Caspase-3 and Caspase-9 were measured by qRT–PCR using the TB GREEN Premix Ex Taq™ kit (Takara, Japan), and gene expression was normalized to β-actin expression on an ABI QuantStudio6 Pro (Applied Biosystems, USA). The primer sequences used in this study are presented in Table 2. Five biological replicates were performed for the qRT–PCR. Table 2Primer sequences used in the study for qRT–PCR analyses.GeneForward sequenceReverse sequencehACTB5′-ACCAACTGGGACGACATGGAGAAA-3′5′-TAGCACAGCCTGGATAGCAACGTA-3′hTGFβ15′-GAGCCTGAGGCCGACTACTA-3′5′-CGGAGCTCTGATGTGTTGAA-3′hPTGS25′-ATTCCCTTCCTTCGAAATGC-3′5′-GGGGATCAGGGATGAACTTT-3′hACTA25′-GGCATTCACGAGACCACCTAC-3′5′-GGGGCGATGATCTTGATCTT-3′hCOL1A15′-AGGGCCAAGACGAAGACATC-3′5′-GTCGGTGGGTGACTCTGAGC-3′hCOL3A15′-TGAAGGGCAGGGAACAACT-3′5′-GGATGAAGCAGAGCGAGAAG-3′hCASP35′-CCAAAGATCATACATGGAAGCG-3′5′-CTGAATGTTTCCCTGAGGTTTG-3′hTP535′-TTCCTGAAAACAACGTTCTGTC-3′5′-AACCATTGTTCAATATCGTCCG-3′hBCL25′-GACTTCGCCGAGATGTCCAG-3′5′-GAACTCAAAGAAGGCCACAATC-3′hBAX5′-CGAACTGGACAGTAACATGGAG-3′5′-CAGTTTGCTGGCAAAGTAGAAA-3′rACTB5′-AGCCACCAATCCACACAG-3′5′-CCTCTATGCCAACACAGT-3′rTGFβ15′-CTAATGGTGGACCGCAACAAC-3′5′-CACTGCTTCCCGAATGTCTGA-3′rPTGS25′-AACACCTGAGCGGTTACCACT-3′5′-GAGGCAATGCGGTTCTGATAC-3′rTP535′-GTCAGGGACAGCCAAGTCT-3′5′-CGTCTCACGACCTCAGTCAT-3′rCASP35′-ATAAAAATGACCTTACTCGTGAA-3′5′-AGTAGTCGCCTCTGAAGAAAC-3′rCASP95′-GGACTGTCTTTTGCCTGTATC-3′5′-GACCTTGGAACACAGAGAAATAC-3′ Table 2Primer sequences used in the study for qRT–PCR analyses.GeneForward sequenceReverse sequencehACTB5′-ACCAACTGGGACGACATGGAGAAA-3′5′-TAGCACAGCCTGGATAGCAACGTA-3′hTGFβ15′-GAGCCTGAGGCCGACTACTA-3′5′-CGGAGCTCTGATGTGTTGAA-3′hPTGS25′-ATTCCCTTCCTTCGAAATGC-3′5′-GGGGATCAGGGATGAACTTT-3′hACTA25′-GGCATTCACGAGACCACCTAC-3′5′-GGGGCGATGATCTTGATCTT-3′hCOL1A15′-AGGGCCAAGACGAAGACATC-3′5′-GTCGGTGGGTGACTCTGAGC-3′hCOL3A15′-TGAAGGGCAGGGAACAACT-3′5′-GGATGAAGCAGAGCGAGAAG-3′hCASP35′-CCAAAGATCATACATGGAAGCG-3′5′-CTGAATGTTTCCCTGAGGTTTG-3′hTP535′-TTCCTGAAAACAACGTTCTGTC-3′5′-AACCATTGTTCAATATCGTCCG-3′hBCL25′-GACTTCGCCGAGATGTCCAG-3′5′-GAACTCAAAGAAGGCCACAATC-3′hBAX5′-CGAACTGGACAGTAACATGGAG-3′5′-CAGTTTGCTGGCAAAGTAGAAA-3′rACTB5′-AGCCACCAATCCACACAG-3′5′-CCTCTATGCCAACACAGT-3′rTGFβ15′-CTAATGGTGGACCGCAACAAC-3′5′-CACTGCTTCCCGAATGTCTGA-3′rPTGS25′-AACACCTGAGCGGTTACCACT-3′5′-GAGGCAATGCGGTTCTGATAC-3′rTP535′-GTCAGGGACAGCCAAGTCT-3′5′-CGTCTCACGACCTCAGTCAT-3′rCASP35′-ATAAAAATGACCTTACTCGTGAA-3′5′-AGTAGTCGCCTCTGAAGAAAC-3′rCASP95′-GGACTGTCTTTTGCCTGTATC-3′5′-GACCTTGGAACACAGAGAAATAC-3′ Primer sequences used in the study for qRT–PCR analyses. Western blotting analysis The HSFs were lysed in radioimmunoprecipitation assay (RIPA) lysis buffer supplemented with 1 mM phenylmethyl sulfonyl fluoride (PMSF) for 30 min on ice and then centrifuged. The protein concentrations of the supernatants were measured with a BCA Protein Assay Kit (Thermo Fisher Scientific, USA). The extracted proteins were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and transferred to 0.45 μm polyvinylidene difluoride (PVDF) membranes (Millipore, USA). After blocking with 5% bovine serum albumin (BSA) for 1 h at room temperature, the membranes were incubated with primary antibodies at 4 °C overnight. The primary antibodies were anti-β-actin (ab16039; Abcam, UK), anti-type I collagen (Col I, ab34710, Abcam, UK), anti-type III collagen (Col III, ab7778, Abcam, UK), anti-α-SMA (#48938, Cell signaling technology, USA), anti-TGF-β1 (#84912, Cell signaling technology, USA), anti-COX-2 (#4842, Cell signaling technology, USA), anti-TP53 (#2527, Cell signaling technology, USA), anti-Bcl-2 (#4223, Cell signaling technology, USA), anti-BAX (#5023, Cell signaling technology, USA), anti-Caspase-3 (#9662; Cell Signaling Technology, USA), anti-cleaved Caspase-3 (#9664; Cell Signaling Technology, USA), anti-phosphor-MEK (AF3385, Affinity, USA), anti-phosphor-ERK (AF1015, Affinity, USA), anti-AKT (#4060; Cell Signaling Technology, USA) and anti-PI3K(#4255; Cell Signaling Technology, USA) antibodies. Horseradish peroxidase (HRP)-conjugated secondary antibodies (#7074, Cell Signaling Technology, USA) were added on the second day and incubated at room temperature for 1 h. The protein expression levels were analyzed with Odyssey V3.0 image scanning (Li-COR. Inc., Lincoln, NE, USA). The reference gene β-actin (Abcam, UK) was used as a loading control. Western blot assays were conducted on three biological replicates. Wound healing assay Briefly, HSFs were seeded in 6-well plates at a density of 6 × 104 cells/cm2 and cultured with CM until confluence. A scratch wound was then made in the middle of each well with 1 ml pipette tips. Floating cells were removed by washing with phosphate-buffered saline (PBS) three times. The HSFs in each well were transfected with the corresponding siRNAs, and the width of the scratch was recorded in five random fields at 0, 12, 24 and 36 h. Five biological replicates were performed for the wound healing assay. Transwell assay The indicated transfected HSFs were added to the upper chambers of Millicell Hanging Cell Culture Inserts with 8-µm pores (Millipore, USA) at a density of 1 × 105 cells/ml in 200 µl serum-free DMEM, and 500 µl of DMEM supplemented with 10% FBS was added to the lower chambers. After 24 h of incubation, the chambers were removed and fixed with 4% paraformaldehyde (PFA) for 5 min, gently washed and stained with crystal violet for 10 min. The migrated cells that had entered the lower chamber were quantified in each field of images captured under a microscope (Nikon, Japan). The transwell assay was performed on three biological replicates. EdU assay HSFs were seeded in the wells of 96-well plates at a density of 3 × 104 cells in 100 µl media per well. EdU (RiboBio, China) was dissolved in CM at a concentration of 50 µM, and 100 µl was added to each well. After incubation for 2 h and washed, the cells were fixed in 4% PFA for 30 min. The cells were incubated with 100 µl 0.5% Triton X-100 (Sigma, USA) for 10 min and washed twice. For the EdU click reaction, 100 µl of RiboBio’s staining solution was added, and incubated for 30 min in the dark then stained the nuclei with Hoechst (Invitrogen, USA). A total of 5 images were randomly selected and high-magnification views were used to calculate the percentage of positive cells with ImageJ software (National Institutes of Health, USA). The data were analyzed based on five biological replicates. Immunofluorescence analysis The transfected HSFs were fixed with 4% PFA and permeabilized with 0.5% Triton X-100 (Sigma, USA) for 20 min. After blocking with 5% goat serum (Gibco, USA) for 1 h at room temperature, the slides were incubated with a rabbit monoclonal anti-α-SMA antibody (ab5694, Abcam, UK) at 4 °C overnight. The slides were incubated with an Alexa Fluor 488-conjugated goat anti-rabbit IgG secondary antibody (ab150077, Abcam, UK) for 1 h nextday. Finally, the slides were sealed with DAPI Fluoromount-G, and the fluorescence was analyzed by confocal microscopy (Nikon, Japan). Immunofluorescence analysis was conducted on at least three biological replicates. Annexin V/PI staining The transfected cells were digested with 0.25% trypsin (Gibco, USA) and washed with cold PBS. Cell apoptosis was assessed using a FITC Annexin V Apoptosis Detection Kit (BD, USA) according to the experimental instructions. The stained cells were analyzed by flow cytometry (Beckman, USA), and the data were analyzed by FlowJo software (Treestar, USA) based on at least three biological replicates. Animal model For the animal model, HTSs were generated in the tails of 8-week-old male Sprague–Dawley rats as we previously described50. All methods were carried out in accordance with relevant guidelines, and were adhered to the ARRIVE guidelines to ensure ethical and methodological rigor. The animal experiment procedures were approved by the ethics committee of Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine (ethics number: SH9H-2020-A126-SB). The rats were anesthetized by inhaling isoflurane before surgery. Briefly, 6 × 6-mm wounds were created on the tails. After allowing re-epithelialization to occur for approximately 2 weeks, the wounds were stretched with continuous strain to induce the formation of HTSs for another two weeks (Supplementary Fig. 3A-B). Twelve weeks after being subjected to mechanical strain, the scar formed on the rat tails and the rats were randomly divided into four groups: the HKP (NC siRNA), HKP (COX-2 siRNA), HKP (TGF-β1 siRNA), and HKP (COX-2/TGF-β1 siRNAs) groups. HKP (4 µg/ml) loaded with the indicated siRNA (1 µg/ml) was injected into the scar every five days for 15 days. Postoperatively, the rats were housed in separate cages with free access to food and water. Upon completion of the procedure, the rats were humanely euthanized using an overdose of anesthesia (Isoflurane, from Shanghai Yuyan Instruments Co., Ltd), ensuring a painless and humane end to their involvement in the study. Scar samples were harvested after treatment for further evaluation. We used the scar elevation index (SEI), which has been reported in many studies on HS, to quantify scar formation33,52,53. The SEI ratio was calculated as follows: SEI = the length of the highest point of the scar to the surface / the length of the highest point of the adjacent normal skin to the surface of the tail. An SEI > 1 indicates HS formation. The data was analyzed based on at least three biological replicates. Histological staining After fixing the harvested samples with 4% PFA for 24 h, the tissues were paraffin-embedded and cut into 5-µm-thick sections. Tissue sections were stained with hematoxylin & eosin (H&E, Sigma–Aldrich) and observed by microscopy (Nikon, Japan). ImageJ software (National Institutes of Health, USA) was then used to measure and analyze the gross area of the generated scar. Cell apoptosis of the tail HTS samples was assessed with a terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay using the In Situ Cell Death Detection Kit (Roche, Switzerland) following the manufacturer’s instructions. Briefly, the sections were deparaffinized, blocked, and then stained with TUNEL test solution. After washing twice with PBS, the sections were incubated with an HRP-conjugated goat anti-rabbit secondary antibody at room temperature. The percent of TUNEL-positive cells was quantified by ImageJ software (National Institutes of Health, USA) based on six biological replicates. Statistical analysis The results are presented as the mean ± standard deviation (SD). The data between two groups were analyzed by Student’s t test to determine significance. For multivariate comparisons, one-way ANOVA with Tukey’s multiple comparison test was performed. P < 0.05 was considered statistically significant. Statistical analysis was performed with GraphPad Prism 9.0 software (GraphPad Software, USA). Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1