Abstract
Background/Aim: Hypoxia, a characteristic of the tumor microenvironment, affects tumor cell behavior by altering the secretion pattern of extracellular vesicles (EVs). We investigated the regulatory mechanism and biological significance of EV secretion in multiple myeloma (MM) cells grown under hypoxic culture conditions.
Materials and Methods: Human MM cell lines (RPMI8226 and AMO1) were exposed to long-term hypoxia to generate hypoxia-adapted (HA) cells. EVs released into the culture supernatant were quantified by enzyme-linked immunosorbent assay. The expression of proteins involved in hypoxia-induced signaling was analyzed by western blotting. EV secretion was inhibited using GW4869.
Results: Exposure to hypoxia altered the amount of EVs secreted and the phosphorylation levels of Akt and its target molecule AS160. Pharmacological inhibition of Akt phosphorylation suppressed the increase in EV secretion and phosphorylated AS160 expression in HA RPMI8226 cells. GW4869 induced apoptosis in HA RPMI8226 cells, but not in the parental cell line, in the presence or absence of the secretome including EVs. GW4869 altered the expression of proteins related to glycolysis (HK2) and autophagy (LC3).
Conclusion: Akt signaling modulates EV secretion to maintain homeostasis in RPMI8226 cells exposed to hypoxic conditions.
Introduction
Multiple myeloma (MM) is a neoplasm of plasma cells that expand in the bone marrow. Neoplastic proliferation of plasma cells inhibits hematologic functions and alters bone remodeling, which leads to anemia, dyspnea, bone fractures, and hypercalcemia. Abnormal immunoglobulin (M protein) from MM cells accumulates in the blood and causes progressive kidney damage (1). Despite the availability of various therapeutic agents in clinical practice, most MM patients relapse or become refractory to treatment, as indicated by a global trend toward an increase in the incidence and mortality rates of MM (2).
Hematopoietic stem cells reside in a hypoxic region of the bone marrow (3). In a previous study, we used an MM cell-engrafted mouse model and showed that MM cells are present in the hypoxic endosteal niche during disease progression (4). The effects of hypoxia on the biological behavior and malignant phenotype of MM cells can lead to the acquisition of therapeutic resistance (5). Hypoxia-inducible factor (HIF)-1α, a transcription factor that mediates hypoxic signaling, modulates the expression of genes involved in cell proliferation, apoptosis, and metabolic rewiring. Hypoxia markedly affects the production and downstream function of extracellular vesicles (EVs), which are nano-sized vesicles that transport various molecules, including bioactive cargo (6). Hypoxic MM cells show increased secretion of EVs containing microRNA (miR)-135b, which is transferred to endothelial cells and promotes angiogenesis (7). However, the mechanism regulating the secretion of EVs in hypoxic MM cells remains unknown. In previous work, we showed that chronic exposure to hypoxia promotes the stem cell properties and tumorigenicity of MM cells (5). Here, we examined the relationship between Akt activity and EV secretion in hypoxia-adapted (HA) MM cells. Pharmacological inhibition of EV secretion indicated that EVs play an important role in the survival of MM cells under hypoxic conditions.
Materials and Methods
Cell lines. The human MM cell lines RPMI8226 and AMO1 were purchased from Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (Braunschweig, Germany). Both cell lines were cultured in Roswell Park Memorial Institute 1640 medium (RPIM1640; FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) supplemented with 20% fetal bovine serum and 1% penicillin/streptomycin at 37°C in a humidified atmosphere of 5% CO2 and 95% air. For hypoxic adaptation, parental cells were maintained in a specialized incubator (MODEL 9000EX; WakenBteck, Kyoto, Japan) in which the levels of O2 and CO2 were 1.0% and 5%, respectively. HA MM cells were maintained under hypoxic culture conditions for >1 month and then used for experiments. All cell lines were regularly tested for mycoplasma infection using the Venor®GeM Classic Mycoplasma Detection Kit (Minerva Biolabs, Berlin, Germany).
Treatments. Cobalt chloride (Nacalai Tesque, Kyoto, Japan) was dissolved in Ca2+- and Mg2+-free phosphate buffered saline. Dimethyl sulfoxide was used to dissolve pictilisib (MedChemExpress, Monmouth Junction, NJ, USA) and GW4869 (Sigma-Aldrich, St. Louis, MO, USA). Stock solutions were diluted to the desired concentrations using growth medium or MM cell-conditioned medium (CM). CM was prepared by centrifugation (1,500 rpm, 5 min) of culture medium collected 72 h after cell seeding. The vehicle group was treated with equal amounts of solvent for each drug concentration.
Quantification of EVs secreted from MM cells. MM cell-derived CM was collected at an appropriate time after cell seeding with or without drug treatment and subjected to step-wise (ultra)centrifugation (2,000 × g, 20 min; 10,000 × g, 30 min; 100,000 × g, 2 h). The final pellet of the EV-enriched fraction was processed using a CD63/CD63 Exosome ELISA Kit (Hakarel, Ibaraki, Japan) according to the manufacturer’s protocol. EV amounts were normalized to viable cell numbers in the collected CM. The numbers of viable and dead cells were determined using Trypan blue staining.
Western blot analysis. MM cells were lysed in 0.2% sodium dodecyl sulfate (SDS) buffer containing 62.5 mM Tris-HCl, 5% glycerol, a protease inhibitor cocktail (Sigma), 40 mM sodium fluoride, and 2 mM sodium orthovanadate. Protein concentration was determined using the BCA Protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA). Total proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes. Membranes were incubated with primary antibodies against the following proteins: HIF-1α (#610958; BD Bioscience, Franklin Lakes, NJ, USA), β-actin (#A5441, Sigma), AKT Serine/Threonine Kinase [Akt; #9272; Cell Signaling Technology (CST), Danvers, MA, USA], phosphorylated-Akt (#9271; CST), TBC1 domain family member 4 (AS160; #2670; CST), phosphorylated-AS160 (#8881; CST), caspase 3 (#9662; CST), cleaved-caspase 3 (#9661; CST), Microtubule Associated Protein 1 Light Chain 3 (LC3; #4108; CST), and hexokinase 2 (HK2; #sc-130358; Santa Cruz Biotechnology, Dallas, TX, USA). Samples were further incubated with horseradish peroxidase-linked antibodies (CST), followed by treatment with ECL™ Prime Western Blotting Detection Reagent (Cytiva, Marlborough, MA, USA). Protein bands were visualized using ImageQuant LAS 500 (Cytiva).
Flow cytometric analysis for apoptosis measurement. Apoptotic cells were detected using eBioscience™ Annexin V Apoptosis Detection Kits (Thermo) and 7-aminoactinomycin D (7-AAD; Wako). After incubation with these reagents, cells were analyzed using the BD LSRFortessa™ X-20 (BD).
Quantification and statistical analysis. Statistical analysis was performed using GraphPad Prism (version 5 and 9; GraphPad Software, San Diego, CA, USA). All graphs show individual values and the mean±standard deviation. Means were compared using the two-tailed unpaired Student’s t-test or one-way ANOVA with Tukey’s test. p<0.05 was considered statistically significant.
Results
Cell context-dependent alterations of EV secretion during hypoxic adaptation. The effect of hypoxic adaptation on EV secretion in MM cells was assessed by quantifying EVs expressing human CD63 in the CM of parental and HA MM cells using the CD63/CD63 Exosome ELISA Kit. EV amounts were significantly higher in HA RPMI8226 cells than in parental cells (Figure 1A), which is consistent with previously published data by Umezu et al. (7). Hypoxic adaptation decreased the EV amounts in AMO1 cells. Low oxygen tension increases HIF-1α expression by inhibiting the function of Egl-9 family hypoxia-inducible factor 1 (EGLN1, PHD2), which triggers the proteasomal degradation of HIF-1α (8). In line with this, HIF-1α was overexpressed in both HA RPMI8226 and AMO1 cells (Figure 1B). Accumulated HIF-1α drives transcriptional changes that mediate adaptation to hypoxic stress (9). We examined the association of HIF-1α up-regulation with EV secretion in HA MM cells by treating parental RPMI8226 and AMO1 cells with the HIF-1α stabilizer cobalt chloride (10). Cobalt chloride up-regulated the expression of HIF-1α, as shown by western blotting (Figure 1C), but had no effect on the amount of EVs in both MM cell lines (Figure 1D). This suggests that hypoxic adaptation results in cell context-dependent alterations of EV secretion in MM cells and that HIF-1α is not involved in the underlying molecular mechanism.
Hypoxic adaptation alters extracellular vesicle (EV) secretion in RPMI8226 and AMO1 cells in a hypoxia-inducible factor (HIF)-1a-independent manner. (A) Cell culture supernatant was collected 24 h after seeding of multiple myeloma (MM) cells, which were cultured under normoxic conditions (parental) or hypoxic conditions (HA). The culture medium from each condition (three samples per group) was sequentially centrifuged at increasing centrifugal forces, generating an EV-enriched 100,000×g pellet. EV quantification was performed using the CD63/CD63 ELISA system. (B) Western blot analysis of the indicated MM cell lysates prepared 24 h post-seeding. (C, D) Parental MM cells were treated with 0.01% PBS (vehicle) or 100 mM CoCl2 for 24 h under normoxic conditions. Protein expression was analyzed by western blotting (C) and ELISA was used for quantification of EVs in the conditioned media (D; three samples per group). The representative images in (B) and (C) were obtained from each analysis using two samples per group. The data shown in graphs include all points, mean, and standard deviation. p-Values were calculated using the two-tailed unpaired Student’s t-test; **p<0.01 and ***p<0.001.
Akt regulates EV secretion in RPMI8226 cells during hypoxic adaptation. Considering that hypoxic stress influences Akt activity in MM cells (5), we measured the phosphorylation levels of Akt in HA MM cells. In the RPMI8226 cell line, the expression of phosphorylated Akt was higher in HA cells than in parental cells (Figure 2A), whereas in the AMO1 cell line, hypoxic adaptation decreased phospho-Akt levels. This pattern is consistent with the hypoxic adaptation-associated alterations in EV secretion described in Figure 1A. We therefore investigated whether Akt activity is involved in the alterations in the amount of EVs secreted during hypoxic adaptation in MM cells. Treatment of HA RPMI8226 cells with the inhibitor of phosphoinositide 3-kinase pictilisib (11) for 24 h significantly decreased EV secretion (Figure 2B). The down-regulation of phosphorylated Akt by pictilisib was confirmed by western blotting (Figure 2C). These results indicate that Akt activity plays a role in the hypoxic adaptation-associated increase of EV secretion in RPMI8226 cells. To elucidate the mechanism by which Akt regulates EV secretion, we focused on the Akt substrate AS160, which regulates intracellular vesicle trafficking (12). Consistent with the enhanced Akt activity, western blot analysis showed that phosphorylated AS160 was up-regulated in RPMI8226 cells undergoing hypoxic adaptation (Figure 2D). The expression of phospho-AS160 was remarkably decreased in HA AMO1 cells, which could be due to the down-regulation of total AS160 rather than a decrease in Akt activity. Treatment of HA RPMI8226 cells with pictilisib decreased AS160 phosphorylation (Figure 2E). Taken together, these results suggest that Akt activation promotes EV secretion in HA RPMI8226 cells, and AS160 may be a key downstream molecule of Akt signaling regulating EV secretion.
AKT serine/threonine kinase signaling mediates extracellular vesicle (EV) secretion in multiple myeloma (MM) cells with hypoxic adaptation. (A, D) Parental or hypoxia-adapted (HA) MM cells were lysed 24 h post-seeding, and the expression of the indicated proteins was analyzed by western blotting. (B, C, E) HA RPMI8226 cells were treated with 0.2% dimethylsulfoxide (DMSO; vehicle) or 0.5 mM pictilisib under hypoxic conditions. Cell culture supernatant was collected 24 h post-seeding and then subjected to the CD63/CD63 ELISA system (B; three samples per group). MM cell lysates were prepared at 3 h after the indicated treatment and analyzed by western blotting (C, E). The representative images in (A) and (D) were obtained using two samples per group. The data shown in (C) and (E) are representative of two blotting analyses using one sample per group. The data shown in graphs include all points, mean, and standard deviation. p-Values were calculated using the two-tailed unpaired Student’s t-test; **p<0.01.
Inhibiting EV secretion induces apoptotic cell death in HA RPMI8226 cells. We found that EV secretion is increased in HA RPMI8226 cells. Next, we inhibited EV secretion using GW4869, a known inhibitor of EV production (13, 14) and examined the effects on MM cell proliferation and survival. The results of ELISA confirmed that GW4869 suppressed EV secretion in parental and HA RPMI8226 cells (Figure 3A). GW4869 increased the number of dead cells in HA, but not in parental RPMI8226 cells (Figure 3B). To characterize the GW4869-induced cell death, we measured the rate of apoptosis by staining cells with fluorescein-conjugated Annexin V and 7-AAD. Flow cytometry analysis showed that GW4869 significantly increased the rate of apoptosis in HA RPMI8226 cells compared with that in the vehicle control (Figure 3C). Western blot analysis showed that GW4869 up-regulated cleaved-caspase 3, a marker of apoptosis, in HA RPMI8226 cells (Figure 3D). Autocrine activity in tumor cell EVs is involved in tumor development and progression (15, 16). We hypothesized that decreased EV secretion suppresses autocrine signaling leading to cell death. To prove this hypothesis, HA RPMI8226 cells were exposed to CM containing GW4869. At 24 h post-treatment, the number of dead cells did not differ significantly between groups treated with GW4869 in fresh medium and in CM (Figure 3E). These results suggest that EV secretion is a critical regulator of RPMI8226 cell homeostasis for survival under hypoxic conditions.
GW4869 induces apoptosis in hypoxia-adapted (HA) RPMI8226 cells. (A) Parental and HA RPMI8226 cells were cultured with 0.4% dimethylsulfoxide (DMSO; vehicle) or 10 mM GW4869 for 24 h and 3 h, respectively. The resulting cell culture media were subjected to the CD63/CD63 ELISA system for EV quantification (three samples per group). (B) Numbers of dead cells at 72 h after the indicated treatments are shown (three samples per group). (C) Vehicle- or GW4869-treated RPMI8226 cells with hypoxic adaptation were stained with fluorescein-Annexin V and 7-aminoactinomycin D and analyzed by flow cytometry. Representative dot plots and the rates of Annexin V positive cells at 72 h post-treatment are shown (three samples per group). (D) HA RPMI8226 cells were lysed at 24 h after the indicated treatment, and the expression of cleaved- and total caspase 3 was analyzed. The representative image was obtained from the analysis of two samples per group. (E) HA RPMI8226 cells were cultured with vehicle, GW4869, and GW4869 diluted in cell-conditioned medium of HA RPMI8226 cells (GW4869 + CM) for 72 h, and the number of dead cells was determined by Trypan blue staining. The two measurements at the top of the indicated treatments (vehicle and GW4869) were performed in fresh medium. The data shown in the graphs include all points, mean, and standard deviation. p-Values were calculated using the two-tailed unpaired Student’s t-test (A-C) and one-way ANOVA with Tukey’s test (E); *p<0.05 and **p<0.01.
Effect of EV secretion failure on HK2 protein expression and autophagy. A metabolic shift to glycolysis is a common cellular adaptation to hypoxia, and HIF-1α–induced HK2 acts as a rate-limiting enzyme that is indispensable for metabolic reprogramming (17). In MM cells, HK2 promotes anti-apoptotic functions by activating autophagy in response to hypoxic stress (18). We thus performed western blot analysis of HK2 and autophagy-related molecules in RPMI8226 cells. HK2 was up-regulated in HA RPMI8226 cells compared with the parental line, and GW4869 treatment suppressed the up-regulation of HK2 (Figure 4A). Upon autophagic signaling, the cytosolic form of LC3 (LC3-I) is conjugated to phosphatidylethanolamine, and the lipidated form of LC3, known as LC3-II, localizes to the autophagosomal membrane (19). Autophagosomes then fuse with lysosomes, which leads to the degradation of LC3-II. Hypoxic adaptation increased the expression of LC3-II relative to that of its precursor, as observed by comparing untreated HA RPMI8226 and parental cell lines (Figure 4B). GW4869 treatment down-regulated LC3-II in HA RPMI8226 cells, suggesting that inhibiting EV secretion affects the autophagosome quantity. Considering the dynamics of LC3-II expression during autophagy, it remains unclear whether the GW4869-induced changes in LC3-II are due to the suppression of autophagosome formation or their degradation. Taken together, the results suggest that GW4869 modulates the autophagic process by down-regulating HK2 in HA RPMI8226 cells.
GW4869 affects protein expression of hexokinase 2 (HK2) and the autophagic process in hypoxia-adapted (HA) RPMI8226 cells. (A, B) Parental and HA RPMI8226 cells were cultured with 0.4% dimethylsulfoxide (DMSO; vehicle) or 10 mM GW4869 for 24 h. Whole cells were lysed and analyzed by western blotting. The representative images were obtained from each analysis using two samples per group.
Discussion
Several exogenous stimulants influence EV secretion patterns (20, 21). The effect of low oxygen tension on EV secretion (22, 23) and Akt activity (5) depends on cell type and the experimental setting. In this study, we showed that hypoxic adaptation increases EV secretion through the activation of Akt in RPMI8226 cells. Akt, activated by extracellular matrix stiffening, promotes guanosine triphosphate (GTP) loading of Rab8 by the guanine nucleotide exchange factor Rabin8, which leads to EV secretion (24). The Rab GTPase AS160 negatively regulates Rab8-mediated vesicle trafficking by converting GTP into guanosine diphosphate (GTP hydrolysis) on the target protein (25, 26). Activated Akt phosphorylates AS160 (27), thereby disrupting the interaction with Rab8. These findings indicate that Akt positively controls Rab8-mediated EV secretion via the post-transcriptional modification of these two effectors (Rabin8 and AS160). Consistent with these data, we showed that hypoxic adaptation promoted AS160 phosphorylation by activated Akt in RPMI8226 cells; however, whether Rab8 is associated with the increase in EV secretion remains to be determined. The present data obtained using RPMI8226 cells are consistent with previous reports describing hypoxia-enhanced EV secretion as a general phenomenon in three human MM cell lines (RPMI8226, KMS-11, and U266; 7). We found that EV secretion was decreased in AMO1 cells under long-term hypoxic exposure. This can be explained by the suppression of Akt/AS160 axis activity by hypoxic adaptation, as described above.
The molecular components of EVs are affected by the culture conditions of the cells from which the EVs are derived. In response to hypoxia, tumor cells overexpress oncogenic miRNAs and proteins, which are transferred to surrounding cells via EVs to promote tumor development (28, 29). In this study, we investigated whether hypoxia-induced changes in EV secretion affect MM cell survival by inhibiting EV secretion using GW4869. A subset of EVs is generated by membrane invagination in endosomes. GW4869 arrests ceramide-driven membrane invagination by inhibiting neutral sphingomyelinase, which converts sphingomyelin to ceramide (30, 31). Inhibition of EV secretion by GW4869 induced cell death in HA RPMI8226 cells, and simultaneous treatment with CM containing EVs did not restore cell survival. Under normoxic culture conditions, GW4869 had no effect on cell death in RPMI8226 cells. These results suggest that EV secretion failure acts as an endogenous stress leading to cell death in MM cells under hypoxic conditions. EV secretion prevents the aberrant activation of the DNA damage response in non-senescent and senescent human cells (32), and hypoxia causes genetic instability leading to DNA damage (33). Therefore, it would be useful to investigate whether DNA damage mediates GW4869-induced apoptosis in RPMI8226 cells under hypoxic conditions. Alternatively, the present findings indicate that suppression of HK2-mediated autophagy activation may be responsible for cell death; however, mechanistic links between HK2 expression and EV secretion remain elusive. HK2 transcription is positively regulated by HIF-1α. EVs convey hypoxia-inducible miRNAs which target HIF-1α (34-37). Alterations in EV secretion could cause cytosolic accumulation of these miRNAs, which may suppress HIF-1α-mediated HK2 induction.
Conclusion
The findings of this study indicate that adaptation to hypoxia promotes EV secretion via Akt signaling activation in RPMI8226 MM cells, and the enhanced EV secretion is critical for cell survival under hypoxic stress (Figure 5). These results suggest that EV secretion is a promising therapeutic target for therapy-refractory MM cells in the hypoxic microenvironment.
Model depicting the suggested molecular mechanism underlying hypoxic adaptation-associated extracellular vesicle (EV) secretion in RPMI8226 cells and its impact on their survival. AKT serine/threonine kinase (Akt) signaling is activated in hypoxia-adapted RPMI8226 cells and promotes their EV secretion. The Akt substrate TBC1 domain family member 4 (AS160) potentially involves the Akt-regulated EV secretion. Interference with the Akt activation by pictilisib inhibits the EV secretion. In addition, EV secretion maintains homeostasis in RPMI8226 cells to ensure survival under hypoxia. Inhibition of the EV secretion by GW4869 causes apoptosis under hypoxic conditions.
Acknowledgements
The authors thank Dr. Kazuyuki Takata (Joint Research Laboratory, Division of Integrated Pharmaceutical Sciences, Kyoto Pharmaceutical University) for technical assistance.
Footnotes
Authors’ Contributions
YT: Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Supervision, Validation, Writing – original draft, Writing – review & editing. SN, YY, and AF: Data curation, Formal analysis, Investigation, Validation, Visualization, Writing – review & editing. SH and EA: Writing – review & editing.
Conflicts of Interest
The Authors have no conflicts of interest to disclose in relation to this study.
Funding
This work was supported in part by Japan Society for the Promotion of Science grant numbers 40779724 (to Y.T.) and by Ministry of Education, Culture, Sports, Science and Technology-Supported Program for the Strategic Research Foundation at Private Universities 2015-2019 (S1511024L).
- Received February 3, 2025.
- Revision received February 19, 2025.
- Accepted February 20, 2025.
- Copyright © 2025 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).