Multiple myeloma dell-derived microvesicles are enriched in CD147 expression and enhance tumor cell proliferation

Abstract

Multiple myeloma (MM) is characterized by the clonal expansion of malignant plasma cells within the bone marrow. There is a growing literature that tumor cells release biologically active microvesicles (MVs) that modify both local and distant microenvironments. In this study, our goals were to determine if MM cells release MVs, and if so, begin to characterize their biologic activity. Herein we present clear evidence that not only do both patient MM cells and human MM cell lines (HMCLs) release MVs, but that these MVs stimulate MM cell growth. Of interest, MM-derived MVs were enriched with the biologically active form of CD147, a transmembrane molecule previously shown by us to be crucial for MM cell proliferation. Using MVs isolated from HMCLs stably transfected with a CD147-GFP fusion construct (CD147GFP), we observed binding and internalization of MV-derived CD147 with HMCLs. Cells with greater CD147GFP internalization proliferated at a higher rate than did cells with less CD147GFP association. Lastly, MVs obtained from CD147 downregulated HMCLs were attenuated in their ability to stimulate HMCL proliferation. In summary, this study demonstrates the significance of MV shedding and MV-mediated intercellular communication on malignant plasma cell proliferation, and identifies the role of MV-enriched CD147 in this process.

Figure 1. EM analysis of surface associated MVs

A) MV production as revealed by SEM on 3 HMCLs and 3 primary patient CD138⁺ MM cell samples. Whole cell images are shown in the lower left of each panel and the scales for each image are shown in the bottom right panel (MM pt 3). B) SEM analysis of 2 normal bone marrow PC samples. C) Surface MV comparison by TEM of normal bone marrow PC (left), MM patient PC (middle), and the KP-6 HMCL (right).

Figure 2. MM-derived MV dimensions

A) Histogram showing by flow cytometry the size of ALMC-2 derived annexin-V+ MVs (dark gray histogram) in comparison with 0.2, 0.5, 0.8 and 1.0 μm calibration beads (light gray histograms). B) Nanoparticle tracking analysis of MVs isolated from purified MM patient plasma cells. C, D) TEM of RM43 and ALMC-2 HMCL MVs, respectively.

Figure 3. HMCLs shed MVs are enriched for CD147

A) Flow cytometric analysis of HMCL-derived MVs stained with MV marker annexin-V. CD147 staining identifies the molecule as a component within HMCL MVs (shaded histogram) as compared to the isotype control (open histogram). 1 μM beads used for size verification. B) IEM employing immune-gold particles to view CD147 expressing ALMC-2 MVs (right panel) as compared to the isotype-matched control (left panel). C) Western blotting of HMCL panel for CD147 in 5 μg MV lysates (lower panel) and whole cell lysates (WCL; upper panel). Asterisk (*) denotes ALMC-2 WCLs used as a positive control.

Figure 4. HMCL MV phenotypic analysis

Annexin-V+ HMCL MVs were identified and assessed for CD28, CD147, CD138, CD45 and IGF-IR by flow cytometry (open histogram), revealing CD147 enrichment as compared to other markers in panel. Isotype (shaded histogram) used as a control.

Figure 5. MM platelet free plasma MV characterization

A) Annexin-V+ MVs harvested from MM platelet free BM plasma were assessed for CD147 levels and CD31 and CD42 (open histograms) as compared to an isotype control (shaded histogram). B) Western blotting of platelet free BM plasma MVs harvested from 2 kappa (κ) and 3 lambda (λ) MM patients identifies LCs as a component of the MVs. C) Flow cytometry of DP-6 cell line (λ–restricted) and MM pt (κ–restricted) showing positivity of the appropriate LCs in MVs (open histogram) as compared to isotype control (shaded histogram). D) Platelet free BM plasma MV content of MM patients assessed for expression of a panel of markers. Data represent the percentage of dually stained annexin-V+ and LC-restricted MVs from the patient samples which contain the indicated molecule.

Figure 6. MVs promote HMCL proliferation

A) Confocal microscopy revealing internalization of PKH26 stained ALMC-2 MVs (red) indicated by white arrows in RM43 cells after 4 and 24 hr (50 μg/ml). RM43 cells visualized using cytoplasmic Ig (green) and DAPI (blue) counterstain. B) Flow cytometry revealing uptake of both PKH26 stained KP-6 and ALMC-2 MVs (50 μg/ml) in RM43 cells at 4 and 24 hr time points. C) Western blot showing MV induced phosphorylation of MAPK, mTOR and Akt pathways at 30 and 60 minutes following 50 μg MV stimulation. D) HMCLs KP-6, RM43 and ALMC-2 were used to assess the activity of ALMC-2 MVs. Proliferation was assessed by ³H-TdR incorporation after cells were cultured for 72 hrs under unstimulated (Nil) conditions or with 50 μg/mL MVs. Results are displayed as fold increase over background. E) BrdU incorporation as shown by flow cytometry reveals HMCL LYMM transiting the cell cycle after stimulation of 50 μg/ml ALMC-2 MVs at both 24 and 48 hrs.

Figure 7. Bioactivity of MVs prepared from platelet-free bone marrow patient plasma

A) BM patient platelet free plasma derived MVs from MGUS (left panel) and MM pt samples (right panel) were assessed for CD147 protein levels by ELISA. B) HMCLs RM43 and LYMM were used to assess the activity of various MV samples (50 μg/ml) from MGUS, NTMM and RMM patients. Proliferation was assessed on day 3 by ³H-TdR incorporation and results displayed as fold increase over background. C) HMCL MVs but not normal PBMC-derived MVs induce HMCL proliferation. ALMC-2 (left panel) and KP-6 (right panel) cells were incubated for 72 hrs in the presence of MVs (50 μg/ml) obtained from either HMCL KP-6 cells or normal donor peripheral blood mononuclear cells. IL-6 (1 ng/ml) was used as a positive control and proliferation was assessed by ³H-TdR incorporation.

Figure 8. HMCL uptake of CD147 enriched MVs

A) Incubation of ALMC-2 cells (blue, cIg stain; right panel) at a concentration of 2×10⁶/ml in IMDM+ 0.5% BSA with 50 μg/ml ALMC-2 derived CD147^GFP MVs (green) for 24 hrs revealed CD147 cellular uptake as assessed by immunofluorescence. PI was used as a counterstain. B) ALMC-2 cells were cultured at a concentration of 2×10⁶/ml in IMDM+ 0.5% BSA +/- 50 μg/ml ALMC-2 derived CD147^GFP MVs for 24 and 48 hrs. After incubating for 4 hrs at 37°C in the presence of BrdU, cell cycle analysis was performed. Cells incorporating higher levels of CD147 (GFP^hi; orange dashed box) were shown to be enriched in both S and G2/M phases at both time points (right panels) as compared to cells which incorporated less CD147 (GFP^lo; blue dashed box).

Figure 9. CD147 plays a role in MV bioactivity

A) Western blot verifying CD147 knockdown in ALMC-2 whole cell lysates. Beta actin used as a loading control. B) Western blot confirming downregulation of CD147 in ALMC-2 derived MVs. 72 hrs after transfection of control siRNA or CD147 specific siRNA, MVs were collected. C) Downregulation of CD147 in HMCL-derived MVs as compared to mock transfected MVs decreased the proliferative response observed in ALMC-2 cells measured after 3 days of culture in IMDM+ 0.5 BSA +/− 50 μg/mL MVs. D) MVs were obtained from the KP-6 and ALMC-2 HMCLs 48-72 hours after transfection of either control or CD147 specific siRNA. Untreated KP-6 and ALMC-2 cells were then cultured in the presence of 10 μg of control or CD147 siRNA MVs. Proliferation was assessed by [³H]thymidine incorporation after 72 hours. Proliferation of KP-6 cells with KP-6 MVs (left panel); ALMC-2 wth ALMC-2 MVs (middle panel), and KP-6 cells with ALMC-2 MVs (right panel).

Figure 10. Downregulation of MV CD147 effect on signaling pathways

Western blot assessing the ability of MVs isolated from cells transfected with control siRNA or CD147 siRNA to activate MAPK, mTOR and Akt pathways in ALMC-2 cells. Cells were stimulated for 30 min with 50 μg of MVs.