Fig 2: SAA1 promotes the proliferation and metastasis of ovarian cancer in vivo and in vitro. (A–C) Transwell assay to evaluate the migration and invasion capacity of A2780 cells following SAA1 knockdown. Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (D–F) Transwell assay to evaluate the migration and invasion capacity of SKOV3 cells following SAA1 overexpression. Data are presented as mean ± SEM from three independent experiments; unpaired two-tailed Student’s t-test. (G–I) Transwell assay to evaluate the migration and invasion capacity of ID8 cells following SAA1 knockdown. Data are presented as mean ± SEM from three independent experiments; unpaired two-tailed Student’s t-test. (J) Detection of EMT-related proteins (N-Cadherin, E-Cadherin, Vimentin, Snail) in A2780 cells following SAA1 knockdown by Western blotting (top) and densitometric analysis (bottom). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (K) Detection of EMT-related proteins in SKOV3 cells following SAA1 overexpression by Western blotting (top) and densitometric analysis (bottom). Data are presented as mean ± SEM from three independent experiments; unpaired two-tailed Student’s t-tests. (L) Detection of EMT-related proteins in ID8 cells following SAA1 knockdown by Western blotting (top) and densitometric analysis (bottom). Data are presented as mean ± SEM from three independent experiments; unpaired two-tailed Student’s t-tests. (M–N) Luciferase-based in vivo imaging to evaluate the effect of SAA1 knockdown on ovarian cancer progression in tumor-bearing mice. Each dot represents one mouse; data are presented as mean ± SEM; unpaired two-tailed Student’s t-test. (O) Kaplan–Meier analysis to evaluate the effect of SAA1 knockdown on the survival duration of tumor-bearing mice. Log-rank test. (P–Q) Immunohistochemical detection of SAA1, Ki67, E-Cadherin, and N-Cadherin expression in abdominal wall tumors of tumor-bearing mice (P) and quantification of positive staining (Q). Data are presented as mean ± SEM from three mice per group; unpaired two-tailed Student’s t-tests. *Statistical significance: *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001

Fig 3: Increased MDSCs and SAA1 expression associated with ovarian cancer. (A) Flow cytometry analysis of M-MDSCs in PBMCs from patients with benign and malignant ovarian tumors. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (B) Flow cytometry analysis of PMN-MDSCs in PBMCs from patients with benign or malignant ovarian tumors. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (C) Flow cytometry analysis of M-MDSCs in tumor-infiltrating immune cells from patients with benign or malignant ovarian tumors. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (D) Flow cytometry analysis of PMN-MDSCs in tumor-infiltrating immune cells from patients with benign or malignant ovarian tumors. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (E) Flow cytometry analysis of Treg cells in PBMCs from patients with benign or malignant ovarian tumors. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (F) qRT-PCR analysis of the mRNA expression levels of iNOS, IDO and Arg-1 in tumor tissues from patients with benign or malignant ovarian tumors. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (G) Heatmap showing the top upregulated differentially expressed genes following co-culture of SKOV3 cells and MDSCs based on analysis of the GSE145374 dataset. (H) SAA1 expression in ovarian cancer tissues based on the GEPIA database. Each dot represents one sample; data are presented as mean ± SEM; unpaired two-tailed Student’s t-test. (I) Kaplan–Meier analysis of overall survival in ovarian cancer patients stratified by SAA1 expression using GEPIA. Log-rank test. (J) Detection of SAA1 mRNA expression in benign and malignant ovarian tumor tissues by qRT-PCR. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (K) Detection of SAA1 protein levels in serum from patients with benign and malignant ovarian tumors by ELISA. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (L) Association between SAA1 expression and FIGO stage in ovarian cancer based on the public GSE51088 dataset. Each dot represents one sample; Mann–Whitney U test. (M) Association between SAA1 expression and histological grade in ovarian cancer based on the public GSE63885 dataset. Each dot represents one sample; Kruskal–Wallis test. (N) SAA1 protein expression in paired tumor (T) and adjacent paracancerous (P) tissues assessed by Western blotting (left) and densitometric analysis (right). Each dot represents an individual patient; paired two-tailed Student’s t-test. (O) Localization and expression of SAA1 in tumor and paracancerous tissues detected by immunohistochemistry (n = 3). Representative images are presented at 100× and 400× magnification. (P) SAA1 mRNA expression in normal ovarian epithelial cells and ovarian cancer cell lines assessed by qRT-PCR. Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (Q) Detection of SAA1 protein expression in normal ovarian epithelial cells and ovarian cancer cell lines by Western blotting (left) and densitometric analysis (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. *Statistical significance: *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001

Fig 4: SAA1 released by ovarian cancer cells recruits MDSCs and promotes MDSCs differentiation via TLR2/4. (A–B) Flow cytometry analysis of TLR2 expression on the surface of PMN-MDSCs (A) and M-MDSCs (B) from the peripheral blood of patients with benign or malignant ovarian tumors. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (C–D) Flow cytometry analysis of TLR4 expression on the surface of PMN-MDSCs (C) and M-MDSCs (D) from the peripheral blood of patients with benign or malignant ovarian tumors. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (E–F) Correlation analysis between SAA1 expression and TLR2 (E) or TLR4 (F) expression in ovarian cancer using the TIMER database; Spearman’s rank correlation test. (G–H) Rescue experiments assessing the effect of adding recombinant SAA1 protein or the TLR2/4 inhibitor SsnB on the ability of A2780 cell supernatants to recruit MDSCs. Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Tukey’s multiple comparisons test. (I–J) Rescue experiments assessing the effect of recombinant SAA1 protein or SsnB on the ability of ID8 cell supernatants to recruit MDSCs. Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Tukey’s multiple comparisons test. (K–L) Rescue experiments assessing the effect of adding recombinant SAA1 protein or SsnB on the ability of A2780 cell supernatants to induce GMP differentiation into MDSCs. Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Tukey’s multiple comparisons test. *Statistical significance: *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, not significant

Fig 5: SAA1–mediated IL-1β release from MDSCs enhances SAA1 expression in ovarian cancer via the IL-1β/IL-1R/NF-κB axis. (A) Cytokine expression profile in MDSCs following SAA1 stimulation, detected by qRT-PCR. Data are presented as mean ± SEM from five independent experiments; unpaired two-tailed Student’s t-test. (B) Correlation analysis between SAA1 and IL1B expression in ovarian cancer using the TIMER database; Spearman’s rank correlation test. (C) IL-1β expression and secretion in MDSCs after SAA1 treatment across time points (0, 6, 12, 24 h), detected by qRT-PCR (left) and ELISA (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (D) IL-1β expression and secretion in MDSCs after adding the TLR2/4 inhibitor SsnB to SAA1 treatment, detected by qRT-PCR (left) and ELISA (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (E) SAA1 expression and secretion in SKOV3 cells after IL-1β treatment across time points (0, 6, 12, 24 h), detected by qRT-PCR (left) and ELISA (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (F) Detection of SAA1 protein expression in SKOV3 cells after IL-1β treatment across time points (0, 6, 12, 24 h) by Western blotting (left) and densitometric analysis (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (G) Prediction of transcription factors binding to the SAA1 promoter using TFDB, GTRD, and JASPAR databases. (H) Expression of predicted transcription factors in SKOV3 cells with or without IL-1β treatment, detected by qRT-PCR. Data are presented as mean ± SEM from three independent experiments; unpaired two-tailed Student’s t-tests. (I) SAA1 expression and secretion in SKOV3 cells after adding the NF-κB inhibitor APDC to IL-1β treatment, detected by qRT-PCR (left) and ELISA (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (J) SAA1 expression and secretion in SKOV3 cells after adding the IL-1 receptor antagonist (IL-1RA) to IL-1β treatment, detected by qRT-PCR (left) and ELISA (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (K) Detection of NF-κB signaling molecules in SKOV3 cells after adding APDC to IL-1β treatment by Western blotting (left) and densitometric analysis (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (L) Detection of NF-κB signaling molecules in SKOV3 cells after adding IL-1RA to IL-1β treatment by Western blotting (left) and densitometric analysis (right). Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (M) ChIP assay to detect the binding of transcription factor P65 to the SAA1 promoter in SKOV3 cells. Data are presented as mean ± SEM from three independent experiments; one-way ANOVA followed by Dunnett’s multiple comparisons test. (N) IL1B expression in ovarian cancer tissues based on the GEPIA database. Each dot represents one sample; data are presented as mean ± SEM; unpaired two-tailed Student’s t-test. (O) Association between IL1B expression and FIGO stage in ovarian cancer based on the GSE51088 dataset. Each dot represents one sample; Mann–Whitney U test. (P) Association between IL1B expression and histological grade in ovarian cancer based on the GSE63885 dataset. Each dot represents one sample; Kruskal–Wallis test. (Q) Detection of IL1B mRNA expression in benign and malignant ovarian tumor tissues by qRT-PCR. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. (R) Detection of IL-1β protein levels in serum from patients with benign and malignant ovarian tumors by ELISA. Each dot represents an individual patient; data are presented as mean ± SEM; Mann–Whitney U test. *Statistical significance: *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, not significant