Platelet-derived extracellular vesicles inhibit ferroptosis and promote distant metastasis of nasopharyngeal carcinoma by upregulating ITGB3

Abstract

Nasopharyngeal carcinoma (NPC) is a malignancy with high metastatic and invasive nature. Distant metastasis contributes substantially to treatment failure and mortality in NPC. Platelets are versatile blood cells and the number of platelets is positively associated with the distant metastasis of tumor cells. However, the role and underlying mechanism of platelets responsible for the metastasis of NPC cells remain unclear. Here we found that the distant metastasis of NPC patients was positively correlated with the expression levels of integrin β3 (ITGB3) in platelet-derived extracellular vesicles (EVs) from NPC patients (P-EVs). We further revealed that EVs transfer occurred from platelets to NPC cells, mediating cell-cell communication and inducing the metastasis of NPC cells by upregulating ITGB3 expression. Mechanistically, P-EVs-upregulated ITGB3 increased SLC7A11 expression by enhancing protein stability and activating the MAPK/ERK/ATF4/Nrf2 axis, which suppressed ferroptosis, thereby facilitating the metastasis of NPC cells. NPC xenografts in mouse models further confirmed that P-EVs inhibited the ferroptosis of circulating NPC cells and promoted the distant metastasis of NPC cells. Thus, these findings elucidate a novel role of platelet-derived EVs in NPC metastasis, which not only improves our understanding of platelet-mediated tumor distant metastasis, but also has important implications in diagnosis and treatment of NPC.

Keywords: ITGB3; extracellular vesicles; ferroptosis; nasopharyngeal carcinoma; platelets.

Figure 1

Platelet-derived EVs of NPC patients promote metastasis of NPC cells. (A) ELISA of ITGB3 on EVs from platelet-rich plasma and platelets of healthy volunteers (n = 6), NPC patients without distant metastasis (n = 10), and NPC patients with distant metastasis (n = 6). (B) Transmission electron microscopy analysis of EVs from platelets of healthy volunteers, NPC patients without distant metastasis, and NPC patients with distant metastasis (Scale bar, 100 nm). (C) Western blot of ITGB3 and EVs markers (ALIX, TSG101, CD9, and CD63) in platelets and platelet-derived EVs of healthy volunteers and NPC patients. PLTs: Platelets. (D) Nanoparticle tracking analysis to determine size distribution and total number of EVs isolated from the same platelet count of healthy volunteers and NPC patients. (E) Confocal microscopy image showing the internalization of PKH26-labeled H-EVs and P-EVs (red) by 6-10B and 5-8F cells. Hoechst 33342 was used to stain nuclei (blue). Differential interference contrast (DIC) was used to observe the stereoscopic structure of cells (Scale bar, 10 µm). (F) Wound healing assay showing effects of H-EVs and P-EVs treatment on 6-10B and 5-8F cells migration. (G) Matrigel invasion assay was performed to measure the effects of H-EVs and P-EVs on 6-10B and 5-8F cells invasion. (H) Western blot analysis for EMT markers E-cadherin and Vimentin in 6-10B and 5-8F cells after treatment with H-EVs and P-EVs. (I) Adhesion assay showing effects of H-EVs and P-EVs internalization on 6-10B and 5-8F cells adhesion. (J) Clonogenic assays were performed to reveal the effects of H-EVs and P-EVs on clone formation capacity of 6-10B and 5-8F cells. (K) 3D spheroid forming assay determined the effect of H-EVs and P-EVs on 6-10B and 5-8F tumorsphere propagation (Scale bar, 50 µm). Experiments were performed in triplicate.

Figure 2

P-EVs promote NPC metastasis by upregulating ITGB3. (A) Heatmap displaying the hierarchical clustering of genes in 6-10B and 5-8F cells in response to P-EVs treatment. Gene expressions are indicated by the color intensities of red (upregulated) or green (downregulated). (B) Expression of potential genes upregulated by P-EVs treatment in 6-10B and 5-8F cells; numbers indicate quantity of genes in each differentially expressed genes (DEGs) subset. (C) Column diagram represents expression of integrin genes following P-EVs treatment in 6-10B and 5-8F cells. (D) Flow cytometry of cell surface expression of ITGB3 on 6-10B and 5-8F cells treatment with H-EVs or P-EVs. Mouse IgG was used as an isotype control. (E) Western blot of ITGB3 expression in 6-10B and 5-8F cells treatment with H-EVs or P-EVs. (F-G) Wound healing assay (F) and matrigel invasion assay (G) showing cell migration and invasion of P-EVs-treated, P-EVs and cilengitide co-treated, ITGB3 overexpressed, and ITGB3 knockout 6-10B and 5-8F cells (Scale bar, 200 µm). KO: Knockout; OE: Overexpression. (H-I) Clone formation capacity (H) and single-cell-derived clone growth (Scale bar, 100 µm) (I) of P-EVs-treated, P-EVs and cilengitide-co-treated, ITGB3-overexpressed, and ITGB3 knockout 6-10B and 5-8F cells using clonogenic assays. (J) Cell adhesion of P-EVs-treated, P-EVs and cilengitide-co-treated, and ITGB3-overexpressed 6-10B and 5-8F cells using a parallel plate flow adhesion assay (Shear stress, 5 dynes/cm²; Scale bar, 36.71 µm).

Figure 3

P-EVs upregulate ITGB3 expression to inhibit ferroptosis in NPC cells. (A-B) Intracellular ROS levels (A) and cellular GSH/GSSG ratio (B) in P-EVs-treated and ITGB3-overexpressed 6-10B and 5-8F cells by fluorescence microplate reader. (C) Cell viability of P-EVs-treated, P-EVs and cilengitide-co-treated, ITGB3-overexpressed, and ITGB3 knockout 6-10B and 5-8F cells using CCK-8 assay. (D) Cell viability of RSL3-treated, RSL3 and P-EVs-co-treated, ITGB3 overexpression combined with RSL3 treated 6-10B and 5-8F cells. (E) Confocal imaging showed the lipid peroxidation activity (Green) using C11 BODIPY 581/591 in P-EVs-treated, ITGB3-overexpressed, ITGB3 knockout, RSL3-treated, RSL3 and P-EVs-co-treated, and ITGB3 overexpression combined with RSL3 treated 6-10B and 5-8F cells. C11^Non-ox = non-oxidized C11 (Red), C11^Ox = oxidized C11 (Green) (Scale bar, 5 µm). (F) Mitochondrial morphology of P-EVs-treated, ITGB3-overexpressed, ITGB3 knockout, RSL3-treated, RSL3 and P-EVs-co-treated, and ITGB3 overexpression combined with RSL3 treated 6-10B and 5-8F cells were observed by transmission electron microscopy (Scale bar, 500 nm). (G) Flow cytometry analysis of mitochondrial membrane potential in P-EVs-treated, ITGB3-overexpressed, ITGB3 knockout, RSL3 (10 µM)-treated, RSL3 and P-EVs-co-treated, and ITGB3 overexpression combined with RSL3 treated 6-10B and 5-8F cells using JC-10. (H) Flow cytometry analysis of intracellular free iron (Fe²⁺) levels in P-EVs-treated, ITGB3-overexpressed, ITGB3 knockout, RSL3-treated, RSL3 and P-EVs-co-treated, and ITGB3 overexpression combined with RSL3 treated 6-10B and 5-8F cells using the fluorescent indicator Phen Green SK. (I) Select enrichment plots highlighting the enriched iron ion binding-related gene identified using GSEA in P-EVs-treated NPC cells. (J) Western blot analysis of GPX4 expression in P-EVs-treated, ITGB3-overexpressed, ITGB3 knockout, RSL3 (10µM)-treated, RSL3 and P-EVs-co-treated, and ITGB3 overexpression combined with RSL3 treated 6-10B and 5-8F cells. Experiments were repeated at least thrice. Data represent the mean ± SD (^*p<.05; ^**p<.01; ^***p<.001).

Figure 4

P-EVs-upregulated ITGB3 promotes SLC7A11 expression by enhancing protein stability and activating MAPK/ERK pathway. (A) Representative immunofluorescence of ITGB3 and SLC7A11 in 6-10B and 5-8F spheroids released from 3D spheroid culture. Images are maximum intensity z-projections from a confocal z-stack. Hoechst 33342 was used to stain the nuclei of cells (blue) (Scale bar, 20 µm). (B) Flow cytometric analysis of cell surface expression of ITGB3 and SLC7A11 on 6-10B and 5-8F cells treatment with or without P-EVs. Mouse IgG and Rabbit IgG were used as isotype control, respectively. (C) Immunofluorescence microscopy of ITGB3 and SLC7A11 in 6-10B and 5-8F pretreated with P-EVs. Hoechst 33342 was used to stain the nuclei of cells (blue) (Scale bar, 5 µm). (D) 6-10B and 5-8F cells were lysed with RIPA buffer. The interaction between ITGB3 and SLC7A11 was detected by IP using ITGB3 antibody, followed by detection using ITGB3 and SLC7A11 western blotting antibodies. Rabbit IgG was used as isotype control. (E) Western blot analysis of ITGB3, SLC7A11, and ubiquitinylated SLC7A11 in P-EVs-treated and ITGB3-overexpressed 6-10B and 5-8F cells. Abundance of ubiquitinylated SLC7A11 was determined via co-immunoprecipitation with an anti- SLC7A11 antibody as the immunoprecipitant. (F-G) Western blot analysis of phospho-Akt, Akt, phospho-Stat3, Stat3, phospho-Erk1/2, Erk1/2 (F), Nrf2, and ATF4 (G) in P-EVs-treated, P-EVs and cilengitide-co-treated, ITGB3-overexpressed, and ITGB3 knockout 6-10B and 5-8F cells. (H) Western blot analysis of Nrf2 and ATF4 in P-EVs-treated 6-10B and 5-8F cells pretreated with DMSO control, AZD5363, SH-4-54, and SCH772984. (I) ChIP-qPCR analysis of Nrf2 and ATF4 binding to the SLC7A11 promoter region in 6-10B and 5-8F cells. ChIP assay was performed using anti-Nrf2 or anti-ATF4 antibody. Rabbit IgG was used as isotype control. ChIP-qPCR enrichment was shown as the percentage of input DNA. Experiments were repeated at least thrice. Data represent the mean ± SD (^***p<.001).

Figure 5

P-EVs inhibit ferroptosis and promote metastasis of NPC cells through ITGB3-upregulated SLC7A11. (A-B) Intracellular ROS levels (A) and cell viability (B) of P-EVs-treated, ITGB3-overexpressed, SLC7A11 knockout, SLC7A11 knockout combined with P-EVs treated, and SLC7A11 knockout combined with ITGB3 overexpressed 6-10B and 5-8F cells by fluorescence microplate reader and CCK-8 assay. (C) Confocal imaging showed the lipid peroxidation activity (Green) using C11 BODIPY 581/591 in P-EVs-treated, ITGB3-overexpressed, SLC7A11 knockout, SLC7A11 knockout combined with P-EVs treated, and SLC7A11 knockout combined with ITGB3 overexpressed 6-10B and 5-8F cells. C11^Non-ox = non-oxidized C11 (Red), C11^Ox = oxidized C11 (Green) (Scale bar, 5 µm). (D) Flow cytometry analysis of intracellular free iron (Fe²⁺) levels in P-EVs-treated, ITGB3-overexpressed, SLC7A11 knockout, SLC7A11 knockout combined with P-EVs treated, and SLC7A11 knockout combined with ITGB3 overexpressed 6-10B and 5-8F cells using the fluorescent indicator Phen Green SK. (E) Western blot analysis of GPX4 expression in P-EVs-treated, ITGB3-overexpressed, SLC7A11 knockout, SLC7A11 knockout combined with P-EVs treated, and SLC7A11 knockout combined with ITGB3 overexpressed 6-10B and 5-8F cells. (F-I) Cell invasion (F), cell adhesion (G), single-cell-derived clone growth (H), and tumorsphere propagation (I) of P-EVs-treated, ITGB3-overexpressed, SLC7A11 knockout, SLC7A11 knockout combined with P-EVs treated, SLC7A11 knockout combined with ITGB3 overexpressed, RSL3-treated, P-EVs and RSL3-co-treated, and ITGB3 overexpression combined with RSL3 treated 6-10B and 5-8F cells using matrigel invasion assay, adhesion assay, clonogenic assay (Scale bar, 200 µm), and 3D spheroid forming assay (Scale bar, 200 µm), respectively. Experiments were repeated at least thrice. Data represent the mean ± SD (^*p<.05; ^***p<.001).

Figure 6

P-EVs enhance distant metastasis of NPC cells and cause shortened survival of NPC-bearing mice. (A) Experimental design of cell transplantation with 6-10B and 5-8F cells by intraperitoneal (IP) injection and intravenous (IV) injection, respectively, and subsequent in vivo studies. (B) Western blot analysis of ITGB3, SLC7A11, E-cadherin, Vimentin, Nrf2, ATF4, and GPX4 in 6-10B and 5-8F ascitic cells developed from intraperitoneally injected cells following treatment with or without P-EVs. (C) Flow cytometric analysis of cell surface expression of ITGB3 and SLC7A11 on 6-10B and 5-8F ascitic cells developed from intraperitoneally injected cells following treatment with or without P-EVs. Mouse IgG and Rabbit IgG were used as isotype control, respectively. (D-E) Intracellular ROS levels (D) and cellular GSH/GSSG ratio (E) in vehicle- and P-EVs-treated circulating NPC cells by fluorescence microplate reader. (F) Histological analysis of the lung and liver tissues obtained from 6-10B- and 5-8F-transplanted mice following treatment with or without P-EVs (Scale bar, 50 µm). (G) Number of tumor nodules in the lung and liver metastases of 6-10B and 5-8F hematogenous metastatic mouse model following treatment with or without P-EVs. (H) Wet lung weight of 6-10B- and 5-8F-transplanted mice following treatment with or without P-EVs. (I) Kaplan-Meier survival curves for mice intravenously transplanted with 6-10B and 5-8F cells following treatment with or without P-EVs. P values were determined by the log-rank (Mantel-Cox) test. Experiments were repeated at least thrice. Data represent the mean ± SD (^**p<.01; ^***p<.001).