Nasal polyp samples from CRSwNP patients were obtained through functional endoscopic sinus surgery at the Department of Otolaryngology Head & Neck Surgery, West China Hospital, Sichuan University. Turbinate tissues were collected from patients who were undergoing endoscopic skull base surgery or septoplasty. We diagnosed CRSwNP according to EPOS 2020, and ECRSwNP met the requirement of more than 10 eosinophils in each high power field (HFP) of three random fields. Otherwise, it is nECRSwNP (3). Patients who were younger than 18 years old, treated with corticosteroids, antihistamines, or antibiotics before surgery and patients with antrochoanal polyps, autoimmune disease, ciliary dysfunction, fungal rhinosinusitis, and inverted papilloma were excluded. The skin prick test was applied to evaluate the atopic status of the patient. Patient comorbidities and basic demographic data were documented preoperatively. This study was approved by the Medical Ethics Committee of the West China Hospital of Sichuan University, and informed consent was obtained prior to the study. The clinical characteristics of the control and CRSwNP groups are shown in Supplementary Table 1 .

Tissues were embedded in paraffin following ethyl alcohol dehydration and then sectioned to 4 μm thickness. Immunohistochemistry (IHC) staining was performed with a universal detection kit (PV-6000, ZSGB-Bio, Beijing, China) and DAB Detection System (ZLI-9017, ZSGB-Bio, Beijing, China). The sections were air-dried overnight at 37°C, followed by deparaffinization, hydration, antigen retrieval and endogenous peroxidase removal. Then, sections were incubated with primary antibodies against NOS2 (1:400, ABclonal, Wuhan, China), 3-nitrotyrosine (3-NT) (1:200, Abcam, Cambridge, UK), HO-1 (1:400, ABclonal, Wuhan, China) and SOD2 (1:400, ABclonal, Wuhan, China)at 4°C overnight, and incubated with enzyme-labeled sheep anti-mouse/rabbit IgG polymer at 37°C for 20 min. Next, slides were stained with freshly prepared DAB, counterstained with hematoxylin and finally imaged under a 400-fold microscope.

SDS−PAGE gels were used to separate the proteins, after which the proteins were transferred to PVDF membranes (Millipore, MA, USA). Membranes were blocked in TBST containing 5% nonfat milk for 1 h at room temperature (RT). Then, membranes were incubated with primary antibodies against NOS2 (1:1000, ABclonal, Wuhan, China), NOX1 (1:1000, ABclonal, Wuhan, China), HO-1 (1:1000, ABclonal, Wuhan, China), SOD2 (1:1000, ABclonal, Wuhan, China), Kelch-like ECH-associated protein 1 (KEAP1) (1:1000, Proteintech, Wuhan, China), nuclear factor κB (NF-κB) P65 (1:1000, Cell Signaling Technology, Danvers, USA), GAPDH (1:10000, Proteintech, Wuhan, China), Histone H3 (1:4000, HUABIO, Hangzhou, China) and Tublin-α (1:10000, abcam, USA) at 4 °C overnight followed by incubation with secondary antibody (1:5000, Proteintech, Wuhan, China) for 1 h at RT. Finally, membranes were visualized by NcmECL Ultra (NCM Biotech, Shanghai, China).

Animal Total RNA Isolation Kit (FOREGENE, Chengdu, China) and Cell Total RNA Isolation Kit (FOREGENE, Chengdu, China) were used to extract total RNA from nasal tissues and cell, respectively. Total RNA was reverse transcribed into cDNA using HiScript III All-in-one RT SuperMix Perfect for qPCR (Vazyme Biotech, Nanjing, China) according to the manufacturer’s protocols. qPCR was achieved with synthetic primers and Taq pro Universal SYBR qPCR Master Mix (Vazyme Biotech, Nanjing, China). The related primers used are shown in Supplementary Table 2 .

Paraffin sections were prepared with deparaffinization, hydration, and antigen retrieval, and HNEpC were fixed in 4% paraformaldehyde for 15 min at RT. Both tissues and cells were permeabilized in PBS supplemented with 0.1% Triton X-100. Then, 5% BSA supplemented with 0.01% Triton X-100 was used to block nonspecific binding for 2 h at RT. Sections were incubated overnight at 4°C with primary antibodies against CD68 (1:2000, Proteintech, Wuhan, China), NOS2 (1:400, ABclonal, Wuhan, China), HO-1 (1:400, ABclonal, Wuhan, China), SOD2 (1:400, ABclonal, Wuhan, China) and nuclear factor E2-related factor 2 (NRF2) (1:200, Proteintech, Wuhan, China), and cells were incubated with primary antibodies against NRF2 and NF-κB P65 (1:400, Cell Signaling Technology, Danvers, USA), followed by incubation with fluorophore-conjugated secondary antibodies. Finally, sealing tablets containing DAPI were added to the slides, which were later covered by cover glasses.

Peripheral blood mononuclear cells (PBMCs) were isolated from healthy controls by Ficoll-Paque PREMIUM density gradient (GE Healthcare, USA) centrifugation. The CD14+ monocytes were separated by positive magnetic selection (Miltenyi Biotec, Germany). PBMCs were labeled with CD14 MicroBeads for 20 min at 4°C in the dark. Then, the cells were washed and immediately sorted on a MACS Separator (Miltenyi Biotec, Germany) to obtain CD14+ monocytes.

The CD14+ monocytes were cultured in serum-free RPMI-1640 medium for 2 h. After the cells adhered to the wall, the medium was changed to serum-free medium (LONZA, Switzerland) containing 20 ng/ml macrophage colony-stimulating factor (M-CSF) (PeproTech, USA). The medium was changed every 3 days and supplemented with M-CSF. The cells were induced to M0 macrophages (M0) after one week. To polarize into M1 macrophages (M1), cells were treated with 100 ng/ml LPS and 20 ng/ml interferon-γ (IFN-γ) (Novoprotein, Shanghai, China) for 24 h, and cells were stimulated with 20 ng/ml interleukin (IL)-4 (Novoprotein, Shanghai, China) for 24 h to polarize into M2 macrophages (M2).

The HNEpC line was kindly donated by the First Affiliated Hospital of Sun Yat-sen University. The culture medium used for HNEpC was RPMI-1640 medium containing 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin (Gibco, Paisley, UK). The cells were stimulated by either lipopolysaccharide (LPS) (Sigma−Aldrich, USA) or staphylococcal enterotoxin B (SEB) (Toxin Technology, Sarasota, FL, USA) with or without crocin (target mol, USA) when they reached 70-80% confluency.

Fresh nasal polyp tissues were obtained from CRSwNP patients during surgery, washed with RPMI-1640 medium three times, and then cut into smaller pieces weighing approximately 40 mg each. Then, the tissue was passed through a mesh (pore size 0.9 mm²) to acquire tissue fragments, which were resuspended in a 12-well plate and cultured in RPMI-1640 medium containing 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin. Explants were treated with either SEB or LPS for 24 h with or without crocin.

Statistical analyses were performed by using GraphPad Prism 7.0 (GraphPad Software, San Diego, USA). The Kruskal−Wallis test was used for comparisons between multiple groups in nasal tissues and the Spearman correlation coefficient was applied to determine variable relationships in nasal polyp tissues. Cell culture data are presented as the mean ± standard deviation (M ± SD) and were analyzed using one-way analysis of variance (ANOVA). Asterisks indicate statistical significance (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).

Patients were divided into the control group, ECRSwNP group and nECRSwNP group according to the number of infiltrated eosinophils in the tissue sections ( Supplementary Figure 1 ). Furthermore, oxidase expression and distribution were determined by IHC, and the results showed that NOS2 was more highly expressed in the epithelial and submucosal cells of the NP than in those of the control mucosa and that NOX1 was more strongly expressed in the submucosal cells of nasal tissues ( Figure 1A ). Increased levels of the NOS2 and NOX1 proteins were detected in samples from ECRSwNP patients compared with control samples ( Figures 1B, C ). Moreover, the mRNA expression of NOS2 and NOX1 was significantly greater in ECRSwNP patients than in control subjects ( Figure 1D ). As a product of protein oxidation, 3-nitrotyrosine (3-NT) is a marker of oxidative damage (25). 3-NT expression was greater in the epithelial cells and submucosal cells of NPs than in those of UPs from control subjects ( Supplementary Figure 2 ). Collectively, our data revealed an increase in oxidative stress in CRSwNP patients, especially in ECRSwNP patients.

Immunostaining showed that HO-1 was mainly expressed in the submucosal cells of NPs from CRSwNP patients ( Figure 2A ). SOD2 was distributed primarily in epithelial cells, submucosal cells and submucosal glands in nasal tissues and was increased in the submucosal cells of CRSwNP NPs compared with control NPs ( Figure 2A ). In contrast to previous studies (20, 26), we found that HO-1 and SOD2 protein levels were significantly greater in CRSwNP patients than in controls ( Figures 2B, C ). In addition, we found that HO-1 mRNA levels were significantly greater in ECRSwNP patients than in nECRSwNP patients and control subjects, and SOD2 mRNA expression was significantly greater in ECRSwNP patients than in control subjects ( Figure 2D ). In summary, we observed that the expression of antioxidant enzymes increased with increasing oxidative stress in ECRSwNP patients.

To elucidate the role of oxidases and antioxidant enzymes in CRSwNP, we examined the relationship between these enzymes and inflammatory cytokines, such as IL-6, IL-8, IL-5, IL-13 and IFN-γ, in NPs from patients with CRSwNP. NOS2 mRNA levels were positively correlated with IL-6 mRNA expression ( Figure 3A ). The mRNA levels of IL-5 were positively correlated with NOX1 mRNA expression ( Figure 3B ). In addition, the mRNA expression of HO-1 had a negative relationship with that of IL-8, IL-13 and IFN-γ ( Figure 3C ). We also found that SOD2 expression was negatively correlated with IL-5, IL-13, and IFN-γ levels ( Figure 3D ). In general, oxidase levels may be positively correlated with inflammatory cytokine expression, while antioxidant enzyme levels can be negatively correlated with cytokine expression. We speculate that the increase in oxidative stress may lead to an increase in inflammatory cytokine levels, whereas increased antioxidant enzymes may inhibit inflammation.

Previous studies have shown that macrophages are the main cellular source of reactive oxygen species (ROS)/reactive nitrogen species (RNS) in the lungs and can produce NO through NOS2 (27). We investigated the expression of oxidase and antioxidant enzymes in macrophages labeled with CD68 by dual immunofluorescence staining of nasal tissue sections. Our results showed that NOS2 (green), NOX1 (green), HO-1 (green), and SOD2 (green) could be coexpressed with CD68 (red) in nasal tissue from patients with CRSwNP ( Figures 4A–D ). Therefore, macrophages are likely the main source of free radicals.

Real-time quantitative PCR showed that IL-4 promoted M2 macrophage polarization by increasing the mRNA expression of CCL24 and MRC1 ( Figure 5A ). The expression of NOS2 and NOX1 was not increased in M2 macrophage ( Figures 5B, D ) but was increased in M1 macrophages ( Supplementary Figure 3B ), which means that M1 activation induced oxidative stress. M2 polarization suppressed the expression of the antioxidant enzymes HO-1 and SOD2 ( Figures 5C, D ). Interestingly, crocin strongly inhibited the expression of M2 markers ( Figure 5A ), increased HO-1 and SOD2 expression ( Figures 5C, D ) and decreased NOS2, NOX1 and KEAP1 expression in M2 macrophages ( Figures 5B, D ). Moreover, crocin decreased M1 macrophages polarization ( Supplementary Figure 3A ). In summary, we speculated that M1 polarization results in increased oxidative stress, but M2 polarization impairs antioxidant capacity. In addition, crocin simultaneously inhibited M1 polarization by directly reducing the level of oxidative stress and inhibited M2 polarization by improving the antioxidant capacity.

As oxidase and antioxidant enzymes are expressed in the epithelial cells of NPs from patients with CRSwNP, we explored the relationship between epithelial cells and oxidative stress. Our results showed that the nuclear localization of NF-κB was increased in the SEB treatment group ( Figures 6A, C ). In addition, the mRNA and protein levels of NOS2 were increased in the SEB-treated group ( Figures 6D, E ), and the level of IL-33 was increased in the SEB-treated group ( Figure 6E ). These findings indicated that SEB treatment activated oxidative injury and cell inflammation in HNEpCs. Crocin has been proven to ameliorate cardiotoxicity via the KEAP1-NRF2/HO-1 pathway (28). Our results revealed that NRF2 translocated into the nucleus of HNEpCs treated with crocin ( Figure 6B ). SEB treatment decreased HO-1 protein and mRNA levels ( Figures 6D, E ). Interestingly, SEB treatment did not change SOD2 protein levels ( Figure 6D ). Furthermore, crocin induced NF-κB translocation from the nucleus to the cytoplasm ( Figure 6A ), and the expression of NOS2, IL-33 was inhibited by crocin ( Figures 6D, E ). These results provide further evidence that crocin attenuates oxidative injury and inflammation in HNEpCs through the KEAP1-NRF2/HO-1 pathway. We also found that LPS treatment activated inflammation and crocin reversed this effect ( Supplementary Figure 4 ).

Nasal polyp explants have the advantage of replicating the in situ mucosal environment, so we confirmed the antioxidant effect of crocin on nasal polyps based on in vitro explant models. SEB significantly increased NOS2 and NOX1 expression ( Figures 7A, B ). Moreover, HO-1 expression was decreased, while SOD2 expression was elevated by SEB ( Figures 7A, C ). In contrast, crocin decreased NOS2 expression and NOX1 expression ( Figures 7A, B ) and increased the expression of HO-1 and SOD2 ( Figures 7A, C ). SEB also increased IL-5, IL-13, IL-6, IL-8, IL-25, IL-33, IFN-γ and IL-1β mRNA expression, and crocin treatment reversed this effect ( Figure 7D ). In addition, LPS stimulation increased NOX1 mRNA expression as well as that of SOD2, IL-6, IL-8, IL-25, and IL-33 ( Figures 7E, F ). Crocin reduced NOX1, IL-6, IL-8, IL-25, and IL-33 mRNA expression but increased HO-1 and SOD2 expression ( Figures 7E, F ). Together, these results indicated that crocin exerts an anti-inflammatory effect on nasal polyp explants by improving the antioxidant capacity of cells.