Endocytic pathway inhibition attenuates extracellular vesicle-induced reduction of chemosensitivity to bortezomib in multiple myeloma cells

Abstract

Extracellular vesicles (EVs), including exosomes and microvesicles, derived from bone marrow stromal cells (BMSCs) have been demonstrated as key factors in the progression and drug resistance of multiple myeloma (MM). EV uptake involves a variety of mechanisms which largely depend on the vesicle origin and recipient cell type. The aim of the present study was to identify the mechanisms involved in the uptake of BMSC-derived small EVs (sEVs) by MM cells, and to evaluate the anti-MM effect of targeting this process. Methods: Human BMSC-derived sEVs were identified by transmission electron microscopy, nanoparticle tracking analysis, and western blot. The effects of chemical inhibitors and shRNA-mediated knockdown of endocytosis-associated genes on sEV uptake and cell apoptosis were analyzed by flow cytometry. The anti-MM effect of blocking sEV uptake was evaluated in vitro and in a xenograft MM mouse model. Results: sEVs derived from BMSC were taken up by MM cells in a time- and dose-dependent manner, and subsequently promoted MM cell cycling and reduced their chemosensitivity to bortezomib. Chemical endocytosis inhibitors targeting heparin sulphate proteoglycans, actin, tyrosine kinase, dynamin-2, sodium/proton exchangers, or phosphoinositide 3-kinases significantly reduced MM cell internalization of BMSC-derived sEVs. Moreover, shRNA-mediated knockdown of endocytosis-associated proteins, including caveolin-1, flotillin-1, clathrin heavy chain, and dynamin-2 in MM cells suppressed sEV uptake. Furthermore, an endocytosis inhibitor targeting dynamin-2 preferentially suppressed the uptake of sEV by primary MM cells ex vivo and enhanced the anti-MM effects of bortezomib in vitro and in a mouse model. Conclusion: Clathrin- and caveolin-dependent endocytosis and macropinocytosis are the predominant routes of sEV-mediated communication between BMSCs and MM cells, and inhibiting endocytosis attenuates sEV-induced reduction of chemosensitivity to bortezomib, and thus enhances its anti-MM properties.

Keywords: chemosensitivity; endocytosis; exosomes; extracellular vesicles; multiple myeloma.

Figure 1

Human BMSC-derived sEVs can be taken by MM cells. (A) Transmission electron microscopy images of human BMSC-derived sEVs. (B) Size distribution of human BMSC-derived sEVs was determined using the nanoparticle tracking analysis. (C) Exosomal positive markers, including CD63, CD9, flotillin-1, as well as a negative marker calreticulin, in BMSC and BMSC-derived sEV lysate were measured using western blot. (D) 50 μg/mL DID-labeled BMSC-derived sEVs or DID control solution was added to four MM cell lines, including RPMI 8226, MM1S, U266, and H929. After 24 h of culture, the fluorescence signal of DID in these cells was examined using flow cytometry. PBS was added to cells and included as a control. (E) 50 μg/mL DID-labeled BMSC-derived sEVs were added to four MM cell lines and the mean and median fluorescence intensity of DID in these cells was determined using flow cytometry after the culture for indicated times. n=3. Error bar, mean ± SD.

Figure 2

Human BMSC-derived sEVs promote MM cell cycle. (A) MM1S cells were cultured with or without 50 or 100 μg/mL BMSC-derived sEVs for 48 h and the cell cycle was determined using PI staining and flow cytometry. Representative flow cytometry plots are shown in the left panel. The proportions of MM1S cells in different cell cycle stages were measured using flow cytometry and displayed using histogram (right panel). (B) RPMI 8226, (C) U266, or (D) H929 cells were cultured with or without 50 or 100 μg/mL BMSC-derived sEVs for 48 h and the proportions of cells in different cell cycle phases were determined using PI staining and flow cytometry. One representative result in triplicate of three experiments was presented by histograms. Similar results were obtained in three independent experiments. (E) MM1S, RPMI 8226, U266, or H929 cells were cultured with or without 50 or 100 μg/mL DID-labeled BMSC-derived sEVs for 48 h and representative flow cytometry plots are shown. (F) Mean fluorescence intensity of DID in the cells at different cell cycle stages were determined using PI staining and flow cytometry and presented by histograms. One-way ANOVA followed by multiple compressions was used for comparing multiple groups. Error bar, mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 3

Human BMSC-derived sEVs facilitate the drug resistance of MM cells to bortezomib. (A) MM1S, (B) RPMI 8226, (C) U266, or (D) H929 cells were cultured with or without 100 μg/mL BMSC-derived sEVs in the absence or presence of bortezomib (7.5 nM for MM1S, RPMI 8226, and U266 cells, 30 nM for H929 cells) for 48 h and cell viability was measured. One representative result in triplicate of three experiments was presented by histograms. Similar results were obtained in three independent experiments. Student's t-test was used for comparing two groups. (E) MM1S, (F) RPMI 8226, (G) U266, or (H) H929 cells were treated with or without 100 μg/mL BMSC-derived sEVs in the absence or presence of bortezomib for 48 h. Apoptotic cells were determined using 7-AAD and Annexin-V staining and flow cytometry. The proportions of live, early apoptotic or late apoptotic and dead cells were analyzed and presented by histograms. One-way ANOVA followed by multiple compressions was used for comparing multiple groups. Error bar, mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 4

Endocytosis inhibitors decrease the uptake of BMSC-derived sEVs by MM cells. (A) MM1S cells pre-treated with heparin, cytochalasin D (Cyto D), dynasore, genistein, or chlorpromazine (CPZ) at the indicated concentrations for 30 min were cultured with 50 μg/mL DID-labeled BMSC-derived sEVs for another 4 h. The mean fluorescence intensity of DID in these cells was determined using flow cytometry. (B) MM1S cells were treated with heparin, cytochalasin D, dynasore, genistein, or chlorpromazine at the indicated concentrations for 4.5 h in the absence of sEVs and their cell viability was measured. (C) RPMI 8226, U266, or H929 cells were treated with heparin, Cyto D, dynasore, genistein, or CPZ at the indicated concentration for 4.5 h in the absence of sEVs and their cell viability was measured. (D) RPMI 8226, (E) U266, or (F) H929 cells were pre-treated with heparin, Cyto D, dynasore, genistein, or CPZ at the indicated concentration for 30 min and cultured with 50 μg/mL DID-labeled BMSC-derived sEVs for another 4 h. The mean fluorescence intensity of DID in these cells was determined using flow cytometry. One-way ANOVA followed by multiple compressions was used for comparing multiple groups. n=3. Error bar, mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 5

Inhibitors of macropinocytosis impair the uptake of BMSC-derived sEVs by MM cells. (A) MM1S cells pre-treated with EIPA, amiloride, wortmannin, or omeprazole at the indicated concentrations for 30 min were cultured with 50 μg/mL DID-labeled BMSC-derived sEVs for another 4 h. The mean fluorescence intensity of DID in these cells was determined using flow cytometry. (B) MM1S cells were treated with EIPA, amiloride, wortmannin, or omeprazole at the indicated concentrations for 4.5 h in the absence of sEVs and their cell viability was measured. (C) RPMI 8226, U266, or H929 cells were treated with EIPA, amiloride, or wortmannin at the indicated concentration for 4.5 h in the absence of sEVs and their cell viability was measured. (D) RPMI 8226, (E) U266, or (F) H929 cells were pre-treated with EIPA, amiloride, or wortmannin at the indicated concentration for 30 min and cultured with 50 μg/mL DID-labeled BMSC-derived sEVs for another 4 h. The mean fluorescence intensity of DID in these cells was determined using flow cytometry. One-way ANOVA followed by multiple compressions was used for comparing multiple groups. n=3. Error bar, mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 6

Endocytosis-related proteins are involved in the uptake of BMSC-derived sEVs by MM cells. RPMI 8226 and U266 cells were infected with lentivirus expressing shRNA against key endocytic proteins, including (A) CAV-1, (C) CLTC, (E) DNM2, and (G) Flot1 or negative control shRNA (shNC) and the expression of these proteins was determined using western blot. The pixel density of endocytosis-related proteins was quantified and normalized to GAPDH. RPMI 8226 and U266 cells infected with lentivirus expressing shRNA against key endocytic proteins, including (B) CAV-1, (D) CLTC, (F) DNM2, and (H) Flot1, or shNC were cultured with 50 μg/mL DID-labeled sEVs for 4 h and the mean fluorescence intensity of DID in these cells was determined using flow cytometry. Student's t-test was used for comparing two groups. (I-K) RPMI 8226 cells were infected with lentivirus expressing shRNA against DNM2 or CLTC or negative control shRNA (shNC) for 24 h and treated with or without bortezomib in the presence or absence of 100 μg/mL sEVs for another 48 h. (I) Their cell viability was measured using a luminescent cell viability assay. One representative result in triplicate of three experiments was presented by histograms. (J) Apoptotic cells were determined using 7-AAD and Annexin-V staining and flow cytometry. The proportions of live, early apoptotic or late apoptotic and dead cells were analyzed and presented by histograms. (K) Representative flow cytometry plots are shown. One-way ANOVA followed by multiple compressions was used for comparing multiple groups. Error bar, mean ± SD. **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 7

Dynasore inhibits sEV uptake in primary MM cells. (A-F) BMMCs obtained from 5 MM patients were pre-treated with or without dynasore at different final concentrations (12.5, 25, or 50 µM) for 30 min and then incubated with 50 μg/mL DID-labeled sEVs for another 4 h. After staining with anti-mouse CD45-PE, anti-mouse CD38-Pacific Blue, and Annexin V-FITC, cells were acquired using flow cytometry. Live BM cells (Annexin V^-) were gated and analyzed. (A) Gating of MM (CD45^-CD38⁺) and CD45⁺ immune cells from MM patient 1 (left panel). Representative flow cytometry plots showing the DID signal in MM patient 1 MM and immune cells treated with or without dynasore. (B) Median florescence intensity of DID in total, MM, and immune cells was analyzed and presented by histograms. Same analyses were performed using BMMCs isolated from MM patient (C) 2, (D) 3, (E) 4, and (F) 5. One-way ANOVA followed by multiple compressions was used for comparing multiple groups. Error bar, mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 8

Endocytosis inhibitor enhances the anti-MM effect of bortezomib in vitro. (A) RPMI 8226 or (B) U266 cells were pretreated with or without bortezomib for 6 h and then cultured with or without dynasore in the presence or absence of 100 μg/mL sEVs for another 24 h, and the cell viability was measured. One representative result in triplicate of three experiments was presented by histograms. Similar results were obtained in three independent experiments. (C) RPMI 8226 or (D) U266 cells were pretreated with or without bortezomib for 6 h and then cultured with or without dynasore in the presence or absence of 100 μg/mL sEVs for another 24 h, apoptotic cells were determined using 7-AAD and Annexin-V staining and flow cytometry. Representative flow cytometry plots are shown. The proportions of live, early apoptotic or late apoptotic and dead cells were analyzed and presented by histograms. One-way ANOVA followed by multiple compressions was used for comparing multiple groups. Error bar, mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 9

Endocytosis inhibitor enhances the anti-MM effect of bortezomib in vivo and prolongs the survival of MM mice. (A) B-NDG mice inoculated with U266-GFP-Luc cells were treated with bortezomib (Btz) with or without dynasore. At day 2, 3, 5, 6, 7, and 9 after the first treatment, 10 mg/kg dynasore was intraperitoneally injected. 0.3 mg/kg bortezomib were intraperitoneally injected at day 1, 4, 6, 7, 9 after the first treatment. The distribution of MM cells in these mice was measured at the indicated days after the first treatment using a living imaging system. 4 representative mice per group are shown. (B) The total flux in each mouse after treatment for the indicated days was determined and presented (upper panel). Arrows indicate the doses and time points of treatments. The total flux in mice of all groups at day 15 after the first treatment was presented by histogram (lower panel). n (vehicle) =8, n (dynasore) =5, n (btz or btz+dynasore) =6. (C) B-NDG mice inoculated with U266-GFP-Luc cells were treated with bortezomib (Btz) with or without dynasore. At day 1, 2, 4, 5, 7, 8, 9, and 10 after the first treatment, 100 mg/kg dynasore were intraperitoneally injected. 0.3 mg/kg bortezomib were intraperitoneally injected at day 3 and 6 after the first treatment, and at day 8, 0.6 mg/kg bortezomib were intraperitoneally injected. The distribution of U266-GFP-Luc cells in these mice was measured at the indicated days after the first treatment using a living imaging system. (D) The total flux in each mouse after treatment for the indicated days was determined. Arrows indicate the doses and time points of treatments. n (vehicle) =5, n (btz or btz+dynasore) =4, Error bar, mean ± SEM. Student's t-test was used for comparing two groups. (E) Moribund mice with hindlimb paralysis were sacrificed and the Kaplan-Meier curve was established. Arrow indicates the day of the first treatment. *P < 0.05.