Curcumin (65% purity) and lipopolysaccharide (99% purity) were purchased from Sigma-Aldrich (#C1386 and #L2630, St. Louis, MO, United States), respectively. Recombinant human transforming growth factor-β (TGF-β1, PeproTech, NJ, United States). Raw 264.7 cells were obtained from the American Type Culture Collection (ATCC) (China). Neonatal rat cardiac fibroblasts (NRCFs) were isolated from P0–P3 neonatal rats using the Neonatal Heart Dissociation Kit mouse and rat (#130-098-373) (Miltenyi, United States) in accordance with the manufacturer’s protocols. Dulbecco’s modified Eagle’s medium (DMEM) (#8121348) was obtained from Gibco (Thermo Fisher Scientific, Inc.) Fetal bovine serum (FBS) was obtained from BI (#2053264) (Biological Industries, Inc.), phosphate buffered saline (PBS) (#2104140103), Trypsin (#BC-CE-005-100 ml) and Penicillin/Streptomycin (#BC-CE-007-100 ml) were obtained from (Nanjing BioChannel Biotechnology Co., Ltd, China) and In Situ Cell Death Detection Kit, Fluorescein was obtained from Roche Diagnostics (Roche Applied Science, Indianapolis, IN, United States). Annexin V-FITC Apoptosis Detection Kit (#CA1020) was supplied by Beijing Solarbio (Beijing, China). Antibodies against Cleaved-caspase 3 (#9661), p-SMAD2 (#3108), and SMAD2/3 (#8685) were purchased from Cell Signaling Technology, Inc. EDA-Fibronectin (#6328), Periostin (#14041), Vimentin (#92547), α-SMA (#5694), p-SMAD3 (#52903), CD68 (#125212), CD3 (#16669) and Troponin I (#47003)were purchased from Abcam, Inc. ECL western blot detection kits (#FD8020). were purchased from FD bio Science Biological Technology Co., Ltd (China). A cell counting kit-8 (CCK-8) dye was purchased from Biosharp, Inc. Isoflurane was obtained from RWD Life Science Co., Ltd. Picro Sirius Red (#BP-DL030) was obtained from Sbjbio (Nanjing SenBeiJia Biological Technology Co., Ltd.). ELISA kit was used to quantify IL-18 (#ml002816) (Shanghai Enzyme-linked Biotechnology Co., Ltd.).

Forty male C57/BL/6J mice weighing 20-25 g were purchased from the Zhejiang Experimental Animal Center (Hangzhou, China). All experiments were approved by the Ethical Committee of Zhejiang University and all surgical procedures were performed by experienced technician sin a blinded manner. Mice were acclimatized to the standard conditions with 12 h lighting cycle, 25 ± 2°C temperature, free access to water and standard chow for 1 week. Then, they were sorted into four groups of 10 mice per group. Group 1 (sham) received dimethylsulfoxide (DMSO-saline) for 28 days intragastrically (i.g.) as a vehicle for curcumin. Group 2 (MI + DMOS) was treated with MI.Group 3 was treated with curcumin 50 mg/kg/day i.g. for 28 days after MI. Group 4 was treated with curcumin 100 mg/kg/day i.g. for 28 days after MI. The appropriate dose (50 mg/kg and 100 mg/kg) was selected (Wang et al., 2019; He et al., 2020). MI was induced by ligation of the left anterior descending coronary artery. The MI model was established as described previously (Hu et al., 2008).

During 4 weeks treatment, mice was anaesthetized (1.5–2% isoflurane mixed with 98% air condition) and transthoracic echocardiography was performed with a 40 MHz transducer (Vevo 2100 Imaging System, VisualSonics, FUJIFILM, Canada). Two-dimensional B-mode and M-mode measurements in the long-axis view level include left ventricular end-diastolic dimension (LVID,d), left ventricular end-systolic dimension (LVID,s), interventricular septal wall thickness in diastole (IVS,d) and in systole (IVS,s), and left ventricular posterior wall thickness in diastole (LVPW,d) and in systole (LVPW,s). Left ventricular ejection fraction and fractional shortening were automatically calculated by the echocardiographic system (Xiao et al., 2018).

Mouse hearts 7 days post-MI were dehydrated in 30% sucrose solution, embedded in Tissue-Tek OCT compound, snap-frozen in dry ice, and then cut into 7 μm sections. The sections were then stained with a In Situ Cell Death Detection Kit (Roche Applied Science, CH), CD3, CD68, Troponin I (Abcam, United Kingdom), and DAPI (Vector Laboratories, Burlingame, CA, United States).

Mouse hearts 28 days post-MI were fixed in 10% formalin-PBS, then paraffin embedded, and cut into 3 μm sections. After deparaffinization, rehydration and tissue antigen recovery, the sections were stained with Periostin (R&D, United States), Troponin I (Abcam, United Kingdom), and DAPI (Vector Laboratories, Burlingame, CA, United States).

After deparaffinization, sections were stained with Sirius Red. Tissue damage was scored by calculating the scar circumference, including both internal and external scar diameters. The sirius red–stained sections were scanned with a microscope digital camera (Olympus Instrument, United States), and Biotechnologies, China). The percentage of fibrotic area was calculated as the mean value of the endocardial and epicardial length of the whole fibrotic area in proportion to the mean length of the endocardial and epicardial left ventricle using Image using Image Pro Plus software version 6.0.

Mouse macrophage-like Raw 264.7 cells were acquired from the ATCC and cultured in DMEM with 10% FBS and 1% Penicillin/Streptomycin at 37°C with 5% CO₂. After 2 days, the medium was replaced, and nonadherent cells were discarded.

Raw 264.7 cell viability was examined using the CCK-8 assay (Bio-sharp, China) in accordance with the manufacturer’s protocols. Cells were treated with liposaccharides (LPS 1 μg/ml) or LPS with curcumin at different concentrations (0.1, 1, 10, 20, and 50 μM) and the appropriate dose (10μM, 20 μM) was selected refer the previous studies (Gao et al., 2015; Chun-Bin et al., 2020). Then Cells were seeded in a 96-well plate at a density of 8×10³ cells/well. Following treatment, 10 μL of CCK-8 solution was added to each well and incubated for 2 h. Survival Ratio was calculated according to the following equation: cell survival = [(As-Ab)/(Ac-Ab)] ×100%, where As = treated group, Ac = normal group, and Ab = vehicle control group. The absorbance of each well was measured at 450 and 630 nm using a microplate reader (Spark TECAN, Switzerland). All data were calculated from triplicate samples.

Raw 264.7 cells were treated with 1 μg/ml LPS or LPS with curcumin (10 and 20 μM). Cell apoptosis was evaluated by flow cytometry with Annexin V-FITC and PI staining (CytoFlex, Beckman Coulter, Germany). The flow cytometry assay was examined in accordance with the manufacturer’s protocols.

NRCFs were isolated from P0–P3 neonatal rats using the Neonatal Heart Dissociation Kit mouse and rat (Miltenyi, United States) in accordance with the manufacturer’s protocols. Raw 264.7 cells were first seeded in 0.4 μm Transwell chamber (#3412), after treatment in low-glucose DMEM with serum deprivation overnight, NRCFs were stimulated with 10 ng/ml TGF-β and then co-cultured with 1 μg/ml LPS or 1 μg/ml LPS with 20 μM curcumin, respectively.

NRCFs in each group were seeded in a 6-well plate at 2 × 10⁵ cells per well and cultured in FBS free medium overnight. NRCFs were transfected for 48 h with 2 μg of the IL18 mimic (IL-18 rat, Shanghai GenePharma Co., Ltd, China) to overexpress IL18 using X-tremeGENE HP DNA Transfection Reagent according to the manufacturer’s instructions. Real-time fluorescent quantitative polymerase chain reaction (RT-qPCR) and enzyme-linked immunosorbent assay (ELISA) were used to quantify IL-18 transfection efficiency. Conditioned medium was collected from transfected cell medium and analyzed according to the manufacturer’s instructions.

Total RNA was isolated from three replicates per group after treatment using the TRIzol method, 2 μg of total RNA was reverse-transcribed to cDNA using Evo M-MLV RT Premix for qPCR method according to the manufacturer’s instructions (AG). And resultant cDNA samples were subjected to qPCR on LightCycler^® 480 PCR System (Roche) using SYBR^® Green Premix Pro Taq HS qPCR Kit (AG). The upstream and downstream primers of IL6, IL-1β, and TNF-α were designed and synthesized by Tsingke Biotechnology Co., Ltd. The primer sequences used were as follows:

Mouse-IL-6, forward (5′-GAG​GAT​ACC​ACT​CCC​AAC​AGA​CC-3′) and reverse (5′-AAG​TGC​ATC​ATC​GTT​GTT​CAT​ACA-3′); mouse-IL-1β, forward (5′-CCA​GCT​TCA​AAT​CTC​ACA​GCA​G-3′) and reverse (5′-CTT​TGG​GTA​TTG​CTT​GGG​ATC-3′); mouse-TNFα, forward (5′-CGG​AGT​CCG​GGC​AGG​T-3′) and reverse (5′- GCT​GGG​TAG​AGA​ATG​GAT​GAA​CA-3′); mouse-GAPDH, forward (5′-TGG​CCT​TCC​GTG​TTC​CTA​C-3′) and reverse (5′- GAG​TTG​CTG​TTG​AAG​TCG​CA-3′); rat-IL-18, forward (5′-TCA​GAC​CAC​TTT​GGC​AGA​CT-3′) and reverse (5′- GAT​TCG​TTG​GCT​GTT​CGG​TC-3′) rat-GAPDH, forward (5′-AAA​GGG​TCA​TCA​CCC​GCC-3′) and reverse (5′- AGT​GAT​GGC​ATG​GAC​TGT​GG-3′).

The cells seeded in the 6-well plates were collected, lysed in 2.5×sodium dodecyl sulfate (SDS) gel loading buffer (30 mM Tris-HCl, pH 6.8, 1% SDS, 0.05% bromphenol blue, 12.5% glycerol, and 2.5% mercaptoethanol) and boiled for 30 min; then, 20 μL of the protein was loaded onto 8% concentrated gel for protein electrophoresis separation. After that, the protein was transferred to the PVDF membrane (Millipore, Boston, MA), which was then blocked using 5% skim milk powder for 1 h at room temperature. Then, the membranes were incubated with primary antibodies specific for Cleaved-caspase 3, p-SMAD2, SAMD2/3(Cell Signaling Technology), EDA-Fibronectin, Periostin, Vimentin, α-SMA, p-Smad3 (Abcam), gapdh (#KC-5G5), and β-actin (#KC-5A08) (KangChen, Shanghai) at 4°C overnight, washed with PBST, and further incubated with a secondary antibody for 1 h. Finally, the immunoreactive protein was detected by a chemiluminescence assay (AI680RGB, GE HealthCare, United States) using the FDbio-Dura ECL kit (FDbio Science Biological Technology Co., LTD, China).

All results were expressed as value ±standard error of the mean (SEM). Significant differences between two groups were determined by the Student’s t-test, and One-way ANOVA test was used for multiple comparisons, two-way ANOV test was conducted in multiple group with different time points. p < 0.05, p < 0.01, was considered statistically significant present *, **, respectively. Statistical calculations were carried out using GraphPad Prism 9.0. The sample size was ≥5 in each group for in vivo animal studies, and ≥3 in each group for in vitro studies.

To test the effect of curcumin on MI, we performed MI surgery by permanent ligation of the left anterior descending artery followed by intragastric administration of curcumin for 7 days or 28 days. In parallel, a placebo post-MI group was also established. At 7 days post-MI surgery, we discovered significantly inhibited inflammation activation reflected by reduced CD68⁺ and CD3⁺ cells detected in the peri-infarcted area in the curcumin group compared with the placebo group, while no difference in CD68⁺ and CD3⁺ cell counts was observed between the high-dose curcumin group (100 mg/kg) and the low-dose curcumin group (50 mg/kg) (Figures 1A–D). In addition, cardiac function in the MI + curcumin group exhibited significant improvement compared with the placebo group 1 month after MI, with no obvious change observed 7 days post-MI between the two groups, as demonstrated by ejection fraction (EF%) and fraction shortening (FS%) (Figures 1E–G), in addition, the chamber of left ventricle in systolic phase was dramatically decreased in MI + curcumin group compared with MI group (LVID s), while no significant change were detected between groups in diastolic phase (LVID d) (Figures 1H,I). To summarize, the administration of curcumin after MI significantly ameliorated inflammation in the acute phase; however, curcumin exerted a protective effect by preserving long-term cardiac function only after MI, which suggested that the reduced inflammation activation might be related to adverse cardiac remodeling mediated by curcumin intake.

As previously indicated, curcumin delivery significantly improved long-term cardiac function after MI in C57 mice. To determine what yields this beneficial effect, we conducted TUNEL and picro sirius red staining and, surprisingly, discovered no significant reduction in anti-apoptotic effect on cardiomyocyte within border area in the curcumin-treated group compared with the placebo group 7 days after MI (Supplementary Figure S1A,B). However, we observed dramatically reduced scar formation reflected by picro sirius red staining in both scar circumference and infarct size dimension in curcumin group compared with the placebo group ( Figures 2A–C ). More importantly, we discovered significant ameliorated fibrosis within non-infarcted area in MI + curcumin group compared with MI group using the dosage of 100 mg/kg (Figures 2D,E), indicating the robust anti-fibrotic role of curcumin post MI. In addition, we conducted immunostaining on MI segments using periostin to verify cardiac myofibroblast enrichment and collagen deposition, revealing significantly decreased periostin⁺ cell expression within MI segment in Curcumin group (Figures 2F,G), western blot results also confirmed the therapeutic utilization of Curcumin in inhibiting pro-fibrotic protein expression as displayed by ameliorated Periostin, Vimentin, and α-SMA expression (Figure 2H–K). The results presented above indicate that the use of curcumin following MI significantly improved cardiac function by inhibiting excessive collagen deposition and scar formation.

It is known that LPS significantly activate inflammation and mobilize macrophage proliferation as well as M1 polarization. To further investigate the mechanism of curcumin in anti-inflammatory processes, we performed LPS stimulation on macrophage in vitro for 24 h followed by Curcumin administration, CCK8 test revealed dramatic inhibited macrophage proliferation when dosage was 20 uM, indicating Curcumin might exerted anti-proliferation or pro-apoptotic effect in macrophages under LPS stimulation (Figure 3A). Flow cytometry was conducted to determine whether curcumin promoted macrophage apoptosis under LPS stimulation. As shown in Figures 3B,C, curcumin significantly increased macrophage apoptosis under LPS stimulation compared with LPS only also using 20 uM dosage. In addition, the cleaved-caspase three protein level, which acts as a marker of apoptosis, was also significantly up-regulated in the curcumin-treated macrophages with LPS group compared with LPS only reflected by western blot (Figures 3D,E). Furthermore, macrophage activation could induce the secretion of pro-inflammatory cytokines such as IL-6, IL1β, and TNF-α to initiate an immune response to ischemic injury, thus aggravating cardiomyocyte apoptosis; however, curcumin treatment significantly inhibited pro-inflammatory cytokine secretion in macrophages with LPS group compared with LPS only, as reflected by reduced IL-6, IL-1β, and TNF-α secretion especially at 20 uM concentration (Figures 3F–H). These data indicate that the administration of curcumin on macrophages under LPS stimulation exerted both pro-apoptotic effects in macrophages and anti-inflammatory effects by reducing cytokine release from macrophages, which could be greatly beneficial post-MI.

As previously mentioned, curcumin has also been shown to regulate macrophage polarization by increasing anti-inflammatory cytokine IL-10 levels and decreasing the M1 phenotype marker CD86 along with the pro-inflammatory cytokines TNF-α and IL-6. To validate whether curcumin exerted an inhibitory effect on cardiac fibrosis by alleviating inflammation-induced fibrosis or by directly suppressing cardiac fibroblast trans-differentiation and collagen secretion, we co-cultured macrophages under LPS stimulation with isolated primary neonatal rat cardiac fibroblasts in vitro followed by TGF-β stimulation in NRCF for 24 h, Surprisingly, we discovered that curcumin administration in macrophages with LPS stimulation significantly mitigated collagen synthesis from co-cultured NRCF, which was revealed by pro-fibrotic protein expression (α-SMA, Vimentin, Periostin, and EDA-Fibronectin) (Figure 4A,C–F). In contrast, we did not observe a significant change in fibrosis protein expression when curcumin was directly added to NRCF treated with TGF-β for 24 h compared with the TGF-β only group (Figure 4B,G–J). Mechanistically, we discovered decreased phosphorylation of SMAD2/3 in the curcumin-macrophage co-cultured NRCF group compared to NRCF without curcumin treatments, as reflected by the p-SMAD2/3 to total SMAD2/3 ratio (Figure 4K–M), importantly, we also discovered down-regulated phosphorylation of SMAD2/3 in hearts from post-MI with curcumin treated group compared with MI group, indicating the consistent role in anti-phosphorylation of curcumin towards SMAD2/3(Figure 4N–P). In summary, the administration of curcumin only inhibited LPS-stimulated macrophage-fibroblast crosstalk induced excessive collagen deposition, this effect is mediated by the inhibition of SMAD2/3 phosphorylation, while in NRCF with TGF-β stimulation, curcumin delivery was unable to reverse established pro-fibrotic protein expression.

To elucidate the detailed mechanism of the identified curcumin-mediated anti-fibrosis effect, we performed bulk mRNA sequencing in NRCF co-cultured with curcumin-treated macrophages and NRCF co-cultured with PBS-treated macrophages; 776 genes were downregulated and 1,467 genes were upregulated in the curcumin group compared with the PBS group (Figures 5A,B). Gene enrichment from GO pathway analysis revealed that both immune and defense response signaling were mostly activated and altered between the two groups and that these two signaling pathways were classified and regulated by defense response regulation, indicating the pivotal utilization of curcumin in alleviating an overreactive immune response (Figures 5C,D), the top 10 up-regulated and down-regulated genes enriched in immune response signaling are listed in Table 1. Among all altered genes enriched in immune response signaling, IL-18 was found to be the top-ranked gene that was significantly down-regulated in the curcumin group, considering the established pro-fibrotic effect mediated by inflammosome-secreted IL-18 and NLRP3 activation. We speculated that the anti-fibrosis effect exerted by curcumin was mediated by mitigating IL-18 expression and secretion in cardiac fibroblasts.

Previously, we identified IL-18 as a central molecule in mediating LPS-treated macrophage-activated myofibroblast trans-differentiation on NRCF, while curcumin administration significantly ameliorated this process. Hence, to test whether compensation of IL-18 in NRCF after co-culture reverses the anti-fibrotic effect of curcumin, we constructed an IL-18 overexpression plasmid and transfected it into TGF-β-stimulated NRCF after co-culture with LPS and curcumin double-treated macrophages. We discovered significantly upregulated IL-18 gene expression and considerable transfection efficacy, as demonstrated by RT-PCR and immunofluorescence (Supplementary Figure S2A,B). Furthermore, IL-18 protein expression levels were determined by ELISA in the supernatant collected from NRCF, indicating successful overexpression after plasmid transfection (Supplementary Figure S2C). As expected, IL-18 overexpression in NRCF significantly reversed the curcumin-mediated anti-fibrotic effect, as shown by the expression of fibrotic genes (i.e., EDA-Fibronectin, Periostin, Vimentin, and α-SMA) were detected by western blot analysis (Figure 6A–E) more importantly, SMAD2/3 phosphorylation, which was inhibited by curcumin treatment, rebounded by IL-18 overexpression in NRCF (Figure 6F–H). Furthermore, IL-18 content was also discovered to be significantly down-regulated in plasma from MI + curcumin group compared with MI group using ELISA, which is consistent with the outcome from unbiased transcriptome analysis as well as in vitro findings above (Figure 6I). From these findings, we conclude that IL-18-p-SMAD2/3 signaling plays a critical role in the curcumin-mediated anti-fibrosis mechanism in TGF-β-stimulated NRCF co-cultured with LPS-treated macrophages.