SNAP23-Mediated Perturbation of Cholesterol-Enriched Membrane Microdomain Promotes Extracellular Vesicle Production in Src-Activated Cancer Cells

Abstract

Extracellular vesicles (EVs) originating from intraluminal vesicles (ILVs) formed within multivesicular bodies (MVBs), often referred to as small EV (sEV) or exosomes, are aberrantly produced by cancer cells and regulate the tumor microenvironment. The tyrosine kinase c-Src is upregulated in a wide variety of human cancers and is involved in promoting sEV secretion, suggesting its role in malignant progression. In this study, we found that activated Src liberated synaptosomal-associated protein 23 (SNAP23), a SNARE molecule, from lipid rafts to non-rafts on cellular membrane. We also demonstrated that SNAP23 localized in non-rafts induced cholesterol downregulation and ILV formation, resulting in the upregulation of sEV production in c-Src-transformed cells. Furthermore, the contribution of the SNAP23-cholesterol axis on sEV upregulation was confirmed in pancreatic cancer cells. High SNAP23 expression is associated with poor prognosis in patients with pancreatic cancer. These findings suggest a unique mechanism for the upregulation of sEV production via SNAP23-mediated cholesterol downregulation in Src-activated cancer cells.

INTRODUCTION

Cancer cells secrete large amounts of extracellular vesicles (EVs) containing characteristic cargo molecules, including lipids, proteins, and nucleic acids. These EVs are taken up by distal and proximal cells, inducing phenotypic changes in the recipient cells; and contribute to cancer progression by forming the tumor microenvironment.^1,2) Thus, the mechanisms that regulate EV biogenesis in cancer cells have attracted considerable interest.³⁾ Small EVs (sEV), known as exosomes, originate from intraluminal vesicles (ILVs) within multivesicular bodies (MVB), which are released upon fusion with the plasma membrane.⁴⁾ sEV secretion involves multiple steps and numerous effectors. However, the detailed molecular mechanisms underlying their regulation remain elusive. Previously, we demonstrated that Src interacts with Alix, an endosomal sorting complex required for transport (ESCRT)-associated protein, to promote ILV formation and in Src-overexpressing cancer cells.⁵⁾

c-Src is a tyrosine kinase that plays critical roles in cell proliferation, survival, adhesion, and migration. Whereas the activity of Src is strictly regulated by multiple mechanisms (e.g., phosphorylation and microRNA-mediated expression) in normal cells, their dysregulations lead to cancer progression.^6,7) Src is overexpressed and activated in various types of cancers, suggesting its relevance in cancer malignancies.^8,9) We previously reported a mechanism by which the oncogenic potential of Src is controlled via its sequestration in lipid rafts.^10,11) Lipid rafts are defined as nanoscale sphingolipids and cholesterol-rich membrane microdomains, thereby forming a liquid-ordered phase with low membrane fluidity.^12,13) This membrane microdomain compartmentalizes cellular processes by organizing membrane receptors and signal transduction proteins, and plays crucial roles in membrane dynamics, including endocytosis and vesicle secretion.¹⁴⁾

We hypothesized that Src activation in cancer cells might also affect EV secretion through perturbation of lipid rafts since we previously observed that Src induces a decrease in cholesterol and an increase in ceramide in the cellular membrane.^10,15) Indeed, Alix knockdown reduced the Src-dependent EV increase by only half of the sEV secretion in a previous study.⁵⁾ This indicated the existence of other mechanisms underlying the promotion of sEV secretion in Src-activated cancer cells.

The EV secretion step is regulated by the soluble N-ethylmaleimide sensitive factor attachment protein receptor (SNARE) complex, with target membrane SNARE (t-SNARE) on the plasma membrane and v-SNARE in the secreted vesicles, which facilitate vesicle fusion with the plasma membrane, leading to sEV secretion. In this study, we demonstrated that synaptosome-associated protein 23 (SNAP23), a t-SNARE protein of the SNAP25 family,¹⁶⁾ promotes sEV production via downregulation of cholesterol in Src-transformed cells. SNAP23 reportedly localizes to the plasma membrane and assembles into the SNARE complex in lipid rafts by palmitoylation and phosphorylation.^17–19) We revealed that Src activation induces SNAP23 liberation from lipid rafts to non-rafts and that this dislocation of SNAP23 is responsible for the cholesterol downregulation. This phenomenon was also confirmed in Src-upregulated human cancer cells, supporting the correlation between SNAP23 expression and malignancy. Our findings provide a novel mechanism for the upregulation of sEVs in cancer cells and new insights into the role of SNAP23 in cancer progression.

MATERIALS AND METHODS

Cells and Reagents

Csk-deficient (Csk^−/−) mouse embryonic fibroblasts (MEFs) were cultured as previously described.⁹⁾ Human pancreatic cancer cell line, PANC-1 was obtained from the American Type Culture Collection (ATCC). All cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich, St. Louis, MO, U.S.A.) supplemented with 10% fetal bovine serum (FBS; Biosera, Kansas, MO, U.S.A.). Cells were cultured at 37 °C in a humidified chamber with 5% CO₂. Water-soluble cholesterol was obtained from Sigma-Aldrich.

Immunochemical Analysis

Cells and sEV were lysed in 2×sodium dodecyl sulfate (SDS) sample buffer (1.0 M Tris–HCl pH 6.8, 10% SDS, 50% Glycerol), and immunoblotting was performed as described previously.⁹⁾ The following antibodies were used: anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (sc-32233, Santa Cruz Biotechnology, Santa Cruz, CA, U.S.A.), anti-SNAP23 (ab3340, Abcam, Cambridge, U.K.), anti-Src (ab-1, Merck, Darmstadt, Germany), anti-Src pY416 (D49G4, Cell Signaling technology, Danvers, MA, U.S.A.), anti-Src pY529 (44-662G, Thermo Fisher Scientific, Waltham, MA, U.S.A.), anti-Csk (sc-166560, Santa Cruz Biotechnology). Immunocytochemistry was performed as previously described.¹⁰⁾ Fluorescence was observed using a ZEISS LSM 800 confocal microscope with an Airyscan (Carl Zeiss, Oberkochen, Germany).

Quantification of sEVs

For EV preparation, cells were plated onto a 150-mm culture dish at a density of 1.5 × 10⁶ cells and cultured for 48 h. Cell supernatants were filtered through a 0.22-µm filter (MilliporeSigma, Burlington, MA, U.S.A.) and ultracentrifuged at 110000 × g for 70 min at 4 °C (SW41Ti rotor; Beckman Coulter, Brea, CA, U.S.A.). The concentration and size distribution of sEV were determined by nanoparticle tracking analysis (NTA) using a NanoSight LM10 (Malvern Panalytical, Malvern, U.K.) as described previously.⁵⁾

Gene Expression, Short Hairpin RNA (shRNA) and Small Interfering RNA (siRNA)

All gene transfer experiments were performed using the pCX4 retroviral vector.²⁰⁾ The human Rab5 (Q79L) mutant conjugated with DsRed was provided by Dr. Hiroshi Hanafusa (Nagoya University, Japan). c-Src, c-Src conjugated with enhanced green fluorescent protein (EGFP), Csk, and SNAP23 were PCR-amplified and subcloned into the pCX4 vector. A mutant form of SNAP23 (SNAP23-ΔQP) was generated using PCR. Lentiviral SNAP23 shRNA vectors were obtained from Sigma-Aldrich. Both siRNA against the control (ID: NA), mixed mouse SNAP23 (ID: 20619) and human SNAP23 (ID: 8773) were purchased from siTOOLs Biotech (Planegg, Germany).

Live-Cell Imaging

For enlarged endosome imaging, cells were plated on a glass-bottom dish (Matsunami glass) at a density of 2 × 10⁵ cells. After 24 h of incubation, cells were transfected with Rab5 (Q79L)-DsRed and incubated for 24 h. Enlarged endosomes were imaged using a ZEISS LSM 800 confocal microscope with an Airyscan.

Cholesterol Assay

To measure the cholesterol concentration, cells were lysed in ODG buffer¹⁰⁾ and analyzed with the Amplex Red cholesterol assay kit (Thermo Fisher Scientific). The protein concentration was determined using a BCA protein assay kit (Thermo Fisher Scientific).

Subcellular Fractionations

Fractionation of the membrane compartments was performed as previously described.¹⁰⁾ Briefly, cells were lysed with a buffer containing 0.25% Triton X-100 and separated on a discontinuous sucrose gradient (5/35/42.5%) by ultracentrifugation at 40000 rpm for 16 h at 4 °C. The separation of detergent-resistant membrane fractions (DRMs, defined as lipid raft) and non-DRMs (non-raft) was confirmed by immunoblotting with transferrin receptor and caveolin-1 as markers of non-DRMs and DRMs, respectively.

Lipid Analysis

Supercritical fluid chromatography coupled with triple quadrupole mass spectrometry (SFC-QqQMS) was used for lipid analyses. We prepared the extraction solution, supplemented with 10 µL of the Mouse SPLASH^® LIPIDOMIX^® Mass Spec Standard (Avanti Polar Lipids, Alabaster, AL, U.S.A.) and 10 µL of internal standard mix [Ceramide d18:1–17 : 0 and hexosylceramide d18:1 (d5)–18 : 1, 10 µmol/L; monoacylglycerol (MAG) 17 : 1, 100 µmol/L; cholesterol (d7), 300 µmol/L; and free fatty acid 17 : 0, 500 µmol/L] to 1 mL methanol. To extract the lipids, cells were washed twice with 3 mL of pre-warmed (37 °C) D-PBS. The subsequent extraction process was performed as described previously.²¹⁾

Lipids were measured using SFC–QqQMS according previously described with some modifications.²²⁾ Briefly, SFC–QqQMS analysis was performed using an Agilent 1260 InfinityII SFC system equipped with an Agilent 6470 A triple quadrupole LC/MS system with an Agilent jet stream (AJS) ESI interface (Agilent Technologies, Santa Clara, CA, U.S.A.). Cholesterol, phosphatidylcholine (PC), alkyl–acyl phosphatidylcholine/ alkenyl–acyl phosphatidylcholine, sphingomyelin (SM), lysophosphatidylcholine, lysophosphatidylethanolamine, phosphatidylglycerol, phosphatidylinositol (PI), phosphatidylserine (PS), and phosphatidic acid were measured as previously described.²¹⁾ Free fatty acid, MAG, diacylglycerol, triacylglycerol, ceramide, hexosylceramide, phosphatidylethanolamine (PE), and alkenyl–acyl phosphatidylethanolamine were separated on an Waters Viridis HSS C18 SB Column (3.0 × 100 mm, 1.8 µm; Waters, Milford, MA, U.S.A.) maintained at 60 °C. The mobile phase consisted of supercritical carbon dioxide as solution A and 95% (v/v) methanol/water containing 0.1% (w/v) ammonium acetate as solution B. The flow rate was 1 mL/min, and the injection volume was 2 µL. The gradient of solution B was as follows: 1% at 0 min, 50% at 25 min and held until 28 min, 0% at 28.1 min and maintained until 30 min. The MS conditions were identical to those described previously.²¹⁾ The data acquired using SFC–QqQMS were analyzed using MassHunter software (version 10.0, Agilent Technologies). To calculate the amounts of lipids in cells, the mol of individual metabolite species detected in the technical blank were subtracted from cell samples. The negative values were transferred to zero.

Gene Expression Analysis of Pancreatic Cancer Samples

Data analysis was performed using UCSC Xena (https://xenabrowser.net/), which provides access to data stored in The Cancer Genome Atlas (TCGA) database.

Statistical Analysis

Data are presented as mean ± standard deviation. Statistical significance was calculated using the Student’s t-test or one-way ANOVA with Dunnett’s post-hoc analysis using XLSTAT for Microsoft Excel. The test results were reported as two-tailed p-values, where p < 0.05 was considered statistically significant.

Principal least squares-discriminant analysis (PLS-DA) for the composition of lipids were performed using MetaboAnalyst.²³⁾

RESULTS

SNAP23 Promotes Src-Dependent sEV Production by Upregulating ILV Formation

We previously reported that Src-transformed Csk^−/− cells exhibited increased production of sEV, in which activated Src was encapsulated.⁵⁾ Using these cells, we observed that pooled siRNA-mediated SNAP23 knockdown suppressed sEV (Figs. 1a, b). SNAP23 knockdown in these cells did not appreciably affect Src expression or activity (Fig. 1a). Rescue experiments using siRNA-resistant human SNAP23 expression in SNAP23-knockdown cells restored sEV production in Src-transformed cells (Figs. 1a, b). However, SNAP23-dependent sEV production were not observed in the non-transformed Csk^−/− cells (Figs. 1c, d). These observations indicated that SNAP23 is crucial for sEV upregulation in Src-transformed cells.

Fig. 1. SNAP23 Promotes sEV Production by Enhancing ILV Formation in Multivesicular Bodies in Src-Induced Transformation

(a) Total cell lysate from c-Src transformed Csk^−/− cells expressing control (−) or SNAP23 pooled siRNA (siSNAP23) with or without si-resistant human SNAP23 (hSNAP23) were analyzed by immunoblotting with the indicated antibodies. (b) NTA analysis of isolated sEV particles from cells indicated in panel (a). (c) Total cell lysates from non-transformed Csk^−/− cells expressing control (−) or SNAP23 pooled siRNA (siSNAP23) were immunoblotted with the indicated antibodies. (d) NTA analysis of isolated sEV particles from cells indicated in panel (c). (e) Csk^−/− cells expressing Src-EGFP were transfected with Rab5 (Q79L)-DsRed and imaged by confocal microscopy. Scale bar = 10 µm. white arrowhead: filled endosome, red arrowhead: empty endosome. (f) Quantification of Src-filled endosomes in (e). The relative percentage of Rab5 (Q79L)-DsRed endosomes filled with Src-EGFP in cells indicated in (e) are shown. ** p < 0.01 and *** p < 0.001, n.s., not significant.

We subsequently examined the role of SNAP23 in Src-dependent ILV formation in the multivesicular body (MVB).⁵⁾ To observe ILV formation by confocal fluorescence microscopy, endosomes were enlarged by transfection with a constitutively active Rab5 Q79L mutant. We have already confirmed that Src itself is loaded into the inner vesicles of endosomes and is useful as an ILV marker in Csk^−/− cells expressing Src-EGFP (Csk^−/−/Src-EGFP/Rab5(Q79L)).⁵⁾ In siControl-transfected cells, more than half of endosomes contained Src-EGFP-encapsulated vesicles, whereas this ratio is decreased to 25% by SNAP23 knockdown (Figs. 1e, f). Restoration of SNAP23 expression using the human gene fully restored the ratio of the endosomes filled by ILVs. These results suggested that SNAP23 is crucial for ILV formation and promotes sEV formation in Src-transformed cells.

SNAP23 Regulates Cholesterol Content in Src-Transformed Cells

Since the lipid composition of membrane microdomains affects ILV formation in endosomes and EV production,^24–26) we examined the effects of SNAP23 on lipid metabolism. Supercritical fluid chromatography-triple quadrupole mass spectrometry (SFC-QqQMS) analysis was used to analyze the lipid content in Src-transformed cells. Partial least squares-discriminant analysis (PLS-DA) was performed to identify the critical molecules affected by SNAP23 knockdown and clarified that cholesterol levels prominently increased among the lipids detected (Figs. 2a–c). Indeed, cholesterol assay showed that SNAP23 knockdown increased cholesterol level (Fig. 2d). These observations demonstrated that SNAP23 preferentially regulates cholesterol homeostasis in Src-transformed cells.

Fig. 2. SNAP23 Regulates Cholesterol Content in Src-Transformed Cells

(a–c) Csk^−/−/Src cells expressing control (siCont) or SNAP23 pooled siRNA (siSNAP23) were subjected to lipid analysis. Lipids were measured by SFC-QqQMS. (a) PLS-DA score plots for siCont (red) and siSNAP23 (green). The contribution ratios were 75.4% and 10.6% for Component1 and Component2, respectively. (b) PLS-DA loading plots. Red circle shows cholesterol. (c) Top 15 lipids of variable importance in projection (VIP) scores. The heat map on the right of the figure shows which is more common in each group. (d) Csk^−/−/Src cells expressing control (siCont) or SNAP23 pooled siRNA (siSNAP23) cells were subjected to cholesterol assay. * p < 0.05.

To examine the effect of increased cholesterol on sEV production, cholesterol was added to the Src-transformed cells. Cholesterol treatment significantly suppressed sEV production in a dosage-dependent manner (Figs. 3a, b). Under the same conditions, cholesterol treatment significantly decreased Src localization to the inner vesicles of endosomes (Figs. 3c, d). Collectively, these results suggest that SNAP23-mediated downregulation of cholesterol is crucial for ILV formation and promotes sEV production in Src-transformed cells.

Fig. 3. Cholesterol Suppresses sEV Production by Inhibiting ILV Formation

(a) Csk^−/−/Src cells treated with cholesterol at the indicated concentrations examined by NTA analysis of isolated sEV particles. (b) Quantification of the sEV particles in (a). (c) Csk^−/− cells expressing Src-EGFP transfected with Rab5 (Q79L)-DsRed, treated with or without 50 µM cholesterol, and imaged by confocal microscopy. (d) Quantification of Src-filled endosomes. The relative percentage of Rab5 (Q79L)-DsRed endosomes filled with Src-EGFP in cells indicated in (c) are shown. ** p < 0.01, n.s., not significant.

Src-Induced Sequestration of SNAP23 from Lipid Rafts Promotes Cholesterol Downregulation

Since SNAP23 assembles the SNARE complex with other SNARE proteins, including syntaxin4 and VAMP2, in lipid rafts,¹⁸⁾ we examined the effect of Src activation on the membrane distribution of SNAP23 (Fig. 4a). In Csk^−/− cells, SNAP23 was widely distributed in the DRMs (lipid rafts) and non-DRMs (non-rafts) fractions; however, SNAP23 was localized exclusively to non-DRMs fractions in Src-transformed cells (Fig. 4b). When Csk, a negative regulator of Src, was reintroduced into Csk^−/−/Src cells, SNAP23 localization was recovered, as observed in non-transformed Csk^−/− cells, suggesting that Src activation induced the liberation of SNAP23 from lipid rafts (Figs. 4a, b). To investigate the effect of SNAP23 localized in non-rafts, we introduced a SNAP23 mutant (SNAP23-ΔQP; QPSRI^118–122 to AAAAA), which abolishes binding to palmitoyl transferase and causes SNAP23 to be distributed to non-DRMs fractions in Src-transformed cells (Fig. 4c). As expected, the expression of SNAP23-ΔQP decreased cholesterol levels compared to the wild-type SNAP23 (Fig. 4d). Accompanied by cholesterol downregulation, sEV production was increased by the expression of SNAP23-ΔQP (Fig. 4e). These findings suggest that SNAP23 is excluded from non-rafts by Src activation and that SNAP23-mediated cholesterol downregulation promotes sEV production.

Fig. 4. Localization of SNAP23 to Non-raft Compartments Is Associated with Cholesterol Regulation

(a) Total cell lysate from Csk^−/− cells expressing vector (mock), Src (Src), or Src and Csk (Src/Csk) were immunoblotted with the indicated antibodies. (b) DRMs and non-DRMs of cells in (a) were separated on sucrose density gradients. Aliquots of the fractions were immunoblotted with SNAP23 antibody. (c) DRM and non-DRM of Csk^−/−/Src cells expressing wild-type (WT) or ΔQP SNAP23 were separated on sucrose density gradients. (d) Cells in (c) were subjected to cholesterol assay. (e) NTA analysis of isolated sEV particles from cells indicated in (c). ** p < 0.01.

Role of SNAP23 on sEV Production in Human Cancer Cells

To confirm SNAP23-mediated sEV upregulation in cancer cells, SNAP23 was knocked down in human pancreatic cancer cells with Src activation (Fig. 5a). sEV production was suppressed by SNAP23 knockdown in the PANC-1 cells (Fig. 5b). We also observed that exogenously added cholesterol suppressed sEV production in these cells (Fig. 5c). The knockdown of SNAP23 significantly suppressed colony-forming activity, suggesting that SNAP23-mediated sEV secretion is required for the maintenance of the growth of cancer cells (Fig. 5d). The effect of SNAP23 expression on the overall survival rate of patients with pancreatic cancer was investigated to determine the functional relevance of SNAP23 (Fig. 5e). A cohort of 89 patients from the TCGA database was classified into two groups based on SNAP23 gene expression and revealed that patients with high expression of SNAP23 had a significantly poor prognosis in pancreatic cancer. These observations suggest that SNAP23-mediated downregulation of cholesterol may promote sEV production-associated cancer malignancies.

Fig. 5. Role of SNAP23 in Human Cancer Cells

(a) Total cell lysate from PANC-1 cells transfected with control siRNA (siCont), SNAP23 pooled siRNA (siSNAP23), control shRNA (shCont) or SNAP23 shRNA (shSNAP23) and were immunoblotted with the indicated antibodies. (b) NTA analysis of isolated sEV particles from cells indicated in (a). (c) PANC-1 cells treated with 50 µM cholesterol for 48 h were subjected to NTA analysis for the quantitative measurement of isolated sEVs. (d) Soft-agar colony-formation assays of PANC-1 cells indicated in (a). Representative dishes from three independent experiments (left panels) and colony numbers (right graph) are shown. (e) Kaplan–Meier plots of SNAP23 in pancreatic cancer (TCGA data set). (f) Schematic model of SNAP23 role in regulating sEV production. When Src is activated, SNAP23 is redistributed to non-rafts, resulting in cholesterol downregulation, which induces membrane fluidity and promotes Src-induced ILV formation, resulting in sEV production. ** p < 0.01 and *** p < 0.001.

DISCUSSION

This study revealed that SNAP23 is a crucial mediator of Src-induced upregulation of sEVs. A schematic model for SNAP23-mediated regulation of sEV is depicted in Fig. 5f. When Src is activated, as observed in multiple cancers, SNAP23 is redistributed to non-rafts, resulting in cholesterol downregulation, which induces membrane fluidity and promotes Src-induced ILV formation, resulting in sEV production.

We revealed that activated Src translocates SNAP23 from lipid rafts to non-rafts. By contrast, Src inactivated by Csk returns SNAP23 to lipid rafts (Figs. 4a, b). In Csk^−/− cells, EGFP-conjugated SNAP23 was observed in the plasma membrane and endosomal membrane (data not shown). These findings suggest that Src regulates the localization of SNAP23 in the plasma membrane and endosomal membrane, derived from the plasma membrane. Phosphorylated SNAP23 translocates to the plasma membrane for SNARE complex assembly, and palmitoylated SNAP23 is localized to lipid rafts.¹⁴⁾ However, we did not detect the alteration of posttranslational modifications of SNAP23, including phosphorylation and palmitoylation, in Src-transformed cells (data not shown). Our findings suggest that the localization of SNAP23 is regulated by changes in the content or amount of lipid rafts under Src activation. Src perturbs lipid rafts via phosphorylation of Caveolin-1 in MDA-MB-231 cells.²⁷⁾ Furthermore, EGFR/Src/extracellular signal-regulated kinase (ERK) signaling phosphorylates YTH domain-containing family protein 2 (YTHDF2) and inhibits LXRα-dependent cholesterol homeostasis in glioblastoma cells.²⁸⁾ The accumulation of cholesterol in lipid rafts during the acquisition of EGFR-inhibitor resistance promotes the EGFR/Src/ERK signaling pathway for cell proliferation in non-small cell lung cancer (NSCLC).²⁹⁾ We also have demonstrated that Src activation changes the amount of lipid rafts by decreasing cholesterol and increasing ceramide.^10,15) These studies support that Src signaling plays a role in the alteration of cholesterol-enriched lipid raft, which may induce the liberation of SNAP23 from lipid rafts.

Notably, the dislocation of SNAP23 from lipid rafts to non-rafts also caused a reduction in cholesterol (Fig. 4d). Therefore, we speculated that Src activation triggers positive feedback of SNAP23-mediated cholesterol downregulation. Since cholesterol stiffens lipid membranes,³⁰⁾ the decrease of cholesterol can accelerate ILV formation by increasing membrane fluidity.³¹⁾ Enlarged EV size after treatment with cholesterol (Fig. 3a) supports the fact that cholesterol-induced membrane rigidity suppresses inward budding for ILV formation. Although SNAP23 is known to assemble the SNARE complex in lipid rafts, the role of SNAP23 in non-rafts is largely unknown.¹⁸⁾ Our findings on SNAP23-mediated regulation of cholesterol homeostasis provide new insights into the role of SNAP23 in non-rafts. Cholesterol-induced downregulation of ILV formation supports a previous report that cholesterol suppresses EV secretion in vivo.³²⁾

We observed that SNAP23 expression level in pancreatic cancer is inversely correlated with that of sterol regulatory element-binding protein 1 (SREBP1), which promotes cholesterol synthesis³³⁾ in the TCGA dataset (data not shown). The expression of SNAP23 reportedly decreased cholesterol and increased membrane fluidity in rat vascular smooth muscle cells (VSMCs).³⁴⁾ Although extensive analysis is required to elucidate the detailed mechanism underlying SNAP23-mediated cholesterol downregulation, SNAP23 might possibly affect cholesterol metabolism via SREBP1-mediated transcriptional control.

We also observed that the expression level of SNAP23 in pancreatic cancer correlated with prognosis (Fig. 5e). Malignant alterations in pancreatic cancer reportedly increased EV production.³⁵⁾ Although further extensive studies using human cancer tissues and EVs are required, our results support the possibility of SNAP23 as a diagnostic biomarker of cancer progression.

In conclusion, we demonstrated the mechanism underlying that SNAP23 in non-rafts decreases cholesterol levels and promotes ILV formation, resulting in the upregulation of sEV production in Src-transformed cells. These findings suggest that lipid raft components, including cholesterol and its regulators, may offer new opportunities for anti-cancer interventions targeting EV production in some types of cancers.

Acknowledgments

We thank Ms. S. Ikeda and M. Hasebe for their technical assistance at Keio University. This work was supported by JST, CREST (JPMJCR17H4 to C.O.), and a Grant-in-Aid for Scientific Research (B) (20H03456 to C.O.), Advanced Genome Research and Bioinformatics Study to Facilitate Medical Innovation (GRIFIN) from the AMED (JP18km0405209 to A.H.).

Author Contributions

Conceived and designed the experiments: C.O., F.M.

Performed the experiments: F.M., R.H., A.H.

Analyzed the data: F.M., R.H., A.H., C.O.

Wrote and revised the paper: F.M., R.H., A.H., C.O.

Conflict of Interest

The authors declare no conflict of interests.

REFERENCES

• 1) Kalluri R. The biology and function of exosomes in cancer. J. Clin. Invest., 126, 1208–1215 (2016).
• 2) Costa-Silva B, Aiello NM, Ocean AJ, et al. Pancreatic cancer exosomes initiate pre-metastatic niche formation in the liver. Nat. Cell Biol., 17, 816–826 (2015).
• 3) EL Andaloussi S, Mäger I, Breakefield XO, Wood MJ. Extracellular vesicles: biology and emerging therapeutic opportunities. Nat. Rev. Drug Discov., 12, 347–357 (2013).
• 4) Stuffers S, Sem Wegner C, Stenmark H, Brech A. Multivesicular endosome biogenesis in the absence of ESCRTs. Traffic, 10, 925–937 (2009).
• 5) Hikita T, Kuwahara A, Watanabe R, Miyata M, Oneyama C. Src in endosomal membranes promotes exosome secretion and tumor progression. Sci. Rep., 9, 3265 (2019).
• 6) Sirvent A, Mevizou R, Naim D, Lafitte M, Roche S. Src family tyrosine kinases in intestinal homeostasis, regeneration and tumorigenesis. Cancers, 12, 2014 (2020).
• 7) Okuzaki D, Yamauchi T, Mitani F, Miyata M, Ninomiya Y, Watanabe R, Akamatsu H, Oneyama C. c-Src promotes tumor progression through downregulation of microRNA-129-1-3p. Cancer Sci., 111, 418–428 (2020).
• 8) Tai YL, Chu PY, Lai IR, Wang MY, Tseng HY, Guan JL, Liou JY, Shen TL. An EGFR/Src-dependent beta4 integrin/FAK complex contributes to malignancy of breast cancer. Sci. Rep., 5, 16408 (2015).
• 9) Oneyama C, Hikita T, Nada S, Okada M. Functional dissection of transformation by c-Src and v-Src. Genes Cells, 13, 1–12 (2008).
• 10) Oneyama C, Hikita T, Enya K, Dobenecker MW, Saito K, Nada S, Tarakhovsky A, Okada M. The lipid raft-anchored adaptor protein Cbp controls the oncogenic potential of c-Src. Mol. Cell, 30, 426–436 (2008).
• 11) Oneyama C, Iino T, Saito K, Suzuki K, Ogawa A, Okada M. Transforming potential of Src family kinases is limited by the cholesterol-enriched membrane microdomain. Mol. Cell. Biol., 29, 6462–6472 (2009).
• 12) Marsh D. Cholesterol-induced fluid membrane domains: a compendium of lipid-raft ternary phase diagrams. Biochim. Biophys. Acta, 1788, 2114–2123 (2009).
• 13) Schade DS, Shey L, Eaton RP. Cholesterol review: a metabolically important molecule. Endocr. Pract., 26, 1514–1523 (2020).
• 14) Salaün C, Gould GW, Chamberlain LH. Lipid raft association of SNARE proteins regulates exocytosis in PC12 cells. J. Biol. Chem., 280, 19449–19453 (2005).
• 15) Kajiwara K, Yamada T, Bamba T, Fukusaki E, Imamoto F, Okada M, Oneyama C. c-Src-induced activation of ceramide metabolism impairs membrane microdomains and promotes malignant progression by facilitating the translocation of c-Src to focal adhesions. Biochem. J., 458, 81–93 (2014).
• 16) Kádková A, Radecke J, Sorensen JB. The SNAP-25 protein family. Neuroscience, 420, 50–71 (2019).
• 17) Foster LJ, Yeung B, Mohtashami M, Ross K, Trimble WS, Klip A. Binary interactions of the SNARE proteins syntaxin-4, SNAP23, and VAMP-2 and their regulation by phosphorylation. Biochemistry, 37, 11089–11096 (1998).
• 18) Puri N, Roche PA. Ternary SNARE complexes are enriched in lipid rafts during mast cell exocytosis. Traffic, 7, 1482–1494 (2006).
• 19) Greaves J, Gorleku OA, Salaun C, Chamberlain LH. Palmitoylation of the SNAP25 protein family: specificity and regulation by DHHC palmitoyl transferases. J. Biol. Chem., 285, 24629–24638 (2010).
• 20) Akagi T, Sasai K, Hanafusa H. Refractory nature of normal human diploid fibroblasts with respect to oncogene-mediated transformation. Proc. Natl. Acad. Sci. U.S.A., 100, 13567–13572 (2003).
• 21) Hayasaka R, Tabata S, Hasebe M, Ikeda S, Ohnuma S, Mori M, Soga T, Tomita M, Hirayama A. Metabolomic analysis of small extracellular vesicles derived from pancreatic cancer cells cultured under normoxia and hypoxia. Metabolites, 11, 215 (2021).
• 22) Takeda H, Izumi Y, Takahashi M, Paxton T, Tamura S, Koike T, Yu Y, Kato N, Nagase K, Shiomi M, Bamba T. Widely-targeted quantitative lipidomics method by supercritical fluid chromatography triple quadrupole mass spectrometry. J. Lipid Res., 59, 1283–1293 (2018).
• 23) Pang Z, Chong J, Zhou G, de Lima Morais DA, Chang L, Barrette M, Gauthier C, Jacques PE, Li S, Xia J. MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res., 49 (W1), W388–W396 (2021).
• 24) Trajkovic K, Hsu C, Chiantia S, Rajendran L, Wenzel D, Wieland F, Schwille P, Brugger B, Simons M. Ceramide triggers budding of exosome vesicles into multivesicular endosomes. Science, 319, 1244–1247 (2008).
• 25) Skotland T, Hessvik NP, Sandvig K, Llorente A. Exosomal lipid composition and the role of ether lipids and phosphoinositides in exosome biology. J. Lipid Res., 60, 9–18 (2019).
• 26) Skryabin GO, Komelkov AV, Savelyeva EE, Tchevkina EM. Lipid rafts in exosome biogenesis. Biochemistry (Mosc.), 85, 177–191 (2020).
• 27) Yoon HJ, Kim DH, Kim SJ, Jang JH, Surh YJ. Src-mediated phosphorylation, ubiquitination and degradation of Caveolin-1 promotes breast cancer cell stemness. Cancer Lett., 449, 8–19 (2019).
• 28) Fang R, Chen X, Zhang S, Shi H, Ye Y, Shi H, Zou Z, Li P, Guo Q, Ma L, He C, Huang S. EGFR/SRC/ERK-stabilized YTHDF2 promotes cholesterol dysregulation and invasive growth of glioblastoma. Nat. Commun., 12, 177 (2021).
• 29) Pan Z, Wang K, Wang X, Jia Z, Yang Y, Duan Y, Huang L, Wu ZX, Zhang JY, Ding X. Cholesterol promotes EGFR-TKIs resistance in NSCLC by inducing EGFR/Src/Erk/SP1 signaling-mediated ERRalpha re-expression. Mol. Cancer, 21, 77 (2022).
• 30) Chakraborty S, Doktorova M, Molugu TR, Heberle FA, Scott HL, Dzikovski B, Nagao M, Stingaciu LR, Standaert RF, Barrera FN, Katsaras J, Khelashvili G, Brown MF, Ashkar R. How cholesterol stiffens unsaturated lipid membranes. Proc. Natl. Acad. Sci. U.S.A., 117, 21896–21905 (2020).
• 31) Wang S, Allen N, Liang XH, Crooke ST. Membrane destabilization induced by lipid species increases activity of phosphorothioate-antisense oligonucleotides. Mol. Ther. Nucleic Acids, 13, 686–698 (2018).
• 32) Abdullah M, Nakamura T, Ferdous T, Gao Y, Chen Y, Zou K, Michikawa M. Cholesterol regulates exosome release in cultured astrocytes. Front. Immunol., 12, 722581 (2021).
• 33) Yokoyama C, Wang X, Briggs MR, Admon A, Wu J, Hua X, Goldstein JL, Brown MS. SREBP-1, a basic-helix-loop-helix-leucine zipper protein that controls transcription of the low density lipoprotein receptor gene. Cell, 75, 187–197 (1993).
• 34) Yoon MS, Won KJ, Kim DY, Hwang DI, Yoon SW, Jung SH, Lee KP, Jung D, Choi WS, Kim B, Lee HM. Diminished lipid raft SNAP23 increases blood pressure by inhibiting the membrane fluidity of vascular smooth-muscle cells. J. Vasc. Res., 52, 321–333 (2015).
• 35) Romano R, Picca A, Eusebi LHU, Marzetti E, Calvani R, Moro L, Bucci C, Guerra F. Extracellular vesicles and pancreatic cancer: insights on the roles of miRNA, lncRNA, and protein cargos in cancer progression. Cells, 10, 1361 (2021).