KRAS-Dependent Cancer Cells Promote Survival by Producing Exosomes Enriched in Survivin

Abstract

Mutations in KRAS frequently occur in human cancer and are especially prevalent in pancreatic ductal adenocarcinoma (PDAC), where they have been shown to promote aggressive phenotypes. However, targeting this onco-protein has proven to be challenging, highlighting the need to further identify the various mechanisms used by KRAS to drive cancer progression. Here, we considered the role played by exosomes, a specific class of extracellular vesicles (EVs) derived from the endocytic cellular trafficking machinery, in mediating the ability of KRAS to promote cell survival. We found that exosomes isolated from the serum of PDAC patients, as well as from KRAS-transformed fibroblasts and pancreatic cancer cells, were all highly enriched in the cell survival protein Survivin. Exosomes containing Survivin, upon engaging serum-starved cells, strongly enhanced their survival. Moreover, they significantly compromised the effectiveness of the conventional chemotherapy drug paclitaxel, as well as a novel therapy that combines an ERK inhibitor with chloroquine, which is currently in clinical trials for PDAC. The survival benefits provided by oncogenic KRAS-derived exosomes were markedly reduced when depleted of Survivin using siRNA or upon treatment with the Survivin inhibitor YM155. Taken together, these findings demonstrate how KRAS mutations give rise to exosomes that provide a unique form of intercellular communication to promote cancer cell survival and therapy resistance, as well as raise interesting possibilities regarding their potential for serving as therapeutic targets and diagnostic markers for KRAS-dependent cancers.

1. Introduction

The RAS family of small guanosine triphosphate (GTP)-binding proteins, which consists of NRAS, HRAS, and KRAS, are widely expressed and have the ability to activate various cytoplasmic and nuclear proteins (e.g. transcription factors) that promote the expression of genes involved in cell growth, migration, and survival^1,2. Importantly, ~20% of all human cancer patients, and as many as 95% of PDAC patients, have a mutated form of RAS that renders the protein constitutively active^1–4. Among the most common of these activating KRAS mutations in pancreatic cancer represents substitutions of aspartic acid, valine, or arginine for a glycine residue at position 12. The expression of these KRAS point mutants in PDAC tumors and cell lines has been shown to increase the activation of extracellular signal-regulated kinase (ERK), phosphoinositide 3-kinase (PI3K), and their downstream signaling responses, which strongly promote aggressive phenotypes, including invasiveness, metastasis, and therapy resistance^2,5,6. This has made PDAC notoriously difficult to treat and highly lethal, with a very poor (~10%) 5-year survival rate⁷.

Despite extensive efforts by researchers, as well as the pharmaceutical industry, to develop strategies to target aberrant RAS activity for more than 30 years, as yet no approved therapy has emerged. This had fueled the perception that RAS is “undruggable”⁸. However, more recent findings have begun to challenge this notion. Small molecule drug candidates that specifically bind and inhibit the activity of the KRAS^G12C mutant protein have recently been identified and shown promising activities in clinical trials^9–12. However, since KRAS^G12C mutations comprise only 2% of KRAS mutations in PDAC, strategies that target key effectors of oncogenic KRAS and/or specific cellular processes that they promote may offer the most promising therapeutic approaches for the majority of PDAC. Along these lines, Bryant et al. recently showed the combined pharmacological inhibition of KRAS-stimulated ERK activation and autophagy, a degradative process in cells that involves the formation of autophagosomes, potently blocked the growth of PDAC cell lines and patient derived xenografts (PDXs) in mice¹³. The effectiveness of this new strategy as a treatment for PDAC is currently being evaluated in clinical trials (NCT04132505, NCT04386057, and NCT03825289). However, while they hold promise, there is still a good deal more to learn regarding the different mechanisms used by RAS to drive oncogenic transformation, with the hope that this information can be used as the basis for the development of new approaches for therapeutic intervention against RAS-dependent cancers.

EVs are lipid-enclosed packages of cargo that are released by cells, and have been heavily implicated in intercellular communication in a wide variety of biological and disease contexts. There are two broad classes of EVs, microvesicles (MVs) and exosomes, which differ in size and the mechanisms underlying their biogenesis. MVs, which range from ~200–1000 nm in diameter, are formed by the direct outward budding and fission of the plasma membrane¹⁴. Exosomes, which are smaller than MVs (i.e., ~50–150 nm in diameter), are derived from multi-vesicular bodies (MVBs) containing intraluminal vesicles that form within the endo-lysosomal trafficking pathway. Typically, the majority of MVBs within cells fuse with lysosomes where their contents are degraded; however, some MVBs escape this fate and instead fuse with the plasma membrane, where their intraluminal vesicle contents (i.e., exosomes) are released into the extracellular space. Both MVs and exosomes produced by cancer cells can be transferred to surrounding cancer cells, as well as to non-cancerous (‘normal’) cells, where they have been shown to influence the tumor microenvironment, and to promote cell growth, survival, invasion, and metastatic spread¹⁴.

EVs contain many different types of cargo including proteins, RNA transcripts, micro-RNAs, metabolites, and even fragments of genomic DNA¹⁵. Interestingly, cancer cells often produce more MVs and exosomes, compared to their normal cellular counterparts. They are also highly enriched with certain cargo that, when transferred to cells targeted by EVs, can significantly influence their behavior¹⁴. For example, we have shown that depleting MDA-MB-231 breast cancer cells of the deacetylase sirtuin 1 by shRNA caused these cells to generate large numbers of exosomes containing the inhibitor of apoptosis (IAP) family member, Survivin^16–19. The exosomes derived from these cells strongly promoted the survival and invasive activity of those cancer cells they engaged, and these effects were blocked when exosomes were depleted of Survivin. Interestingly, PDAC cells have been shown to express high levels of Survivin, and this protein has also been detected in serum from PDAC patients, suggesting that Survivin could potentially be used as a marker for the disease^20,21. While the effects of oncogenic RAS mutants on several different cellular processes have been extensively studied, how they influence EV formation has not been determined. Here, we show that the serum collected from PDAC patients contain exosomes that are highly enriched in Survivin, whereas exosomes from non-PDAC patients contain far less of this protein. Exosomes generated by KRAS^G12D transformed fibroblasts, as well as those derived from mouse and human PDAC cells lines, were found to be similarly enriched in Survivin and capable of strongly enhancing cell survival and drug resistance, effects that were dependent on their Survivin cargo. Considering the challenges encountered with treating certain types of KRAS-driven cancers, such as PDAC, these findings suggest that exosomes produced in response to oncogenic KRAS expression could potentially serve as therapeutic targets, while their contents might provide valuable diagnostic information.

2. Methods and materials

2.1. Cell Culture, transfections, and treatments

All cells were cultured in an incubator at 37°C and 5% CO₂. MEFs expressing a Tet-Off inducible form of HA-tagged KRAS^G12D were generated as described previously²², and cultured in Dulbecco’s Modified Eagle Medium (DMEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco). The expression of HA-tagged KRAS^G12D in these cells was suppressed by the addition of 1.0 μg/mL doxycycline (Millipore) to the growth medium, while its removal induced HA-tagged KRAS^G12D expression. AK192 cells expressing a Tet-On inducible form of KRAS^G12D were generated as described previously²³, and cultured in Roswell Park Memorial Institute 1640 medium (RPMI-1640, Gibco) containing 10% FBS. The expression of KRAS^G12D in AK192 cells was induced by the addition of 1.0 μg/mL doxycycline to the growth medium, while its removal inhibited the expression of KRAS^G12D. Primary MEFs (pMEFs) were obtained from The Cornell Stem Cell and Transgenic Core, and were maintained in DMEM containing 10% FBS. The human pancreatic cancer cell lines, PANC-1, MIA PaCa-2, and BxPC-3, were obtained from the ATCC. The PANC-1 and MIA PaCa-2 cells were maintained in DMEM containing 10% FBS, while BxPC-3 cells were grown in RPMI-1640 containing 10% FBS. To knockdown Survivin, silencer select siRNA targeting Survivin (Thermo Fisher, #S62463), or negative control (NC) siRNA (Thermo Fisher, #4390844), was introduced into cells using Lipofectamine RNAiMAX (Thermo Fisher). To ectopically express Survivin in cells, the cDNA encoding Survivin was isolated from the 83 glioblastoma stem cell line²⁴ and cloned into the pcDNA3.1/V5-HIS-TOPO vector (Invitrogen). This plasmid, along with the empty vector, were introduced into cells using polyethyleneimine (PEI, Sigma). As indicated, cells were treated with various combinations of 1.0 μM SCH772984 (Selleck), 6.3 μM chloroquine (Sigma), and 0.25 μM YM155 (Selleck).

2.2. Soft agar assay

MEFs cultured in DMEM containing 10% FBS, 0.3% agarose (Sigma), and without or with 1.0 μg/mL doxycycline, were plated on top of a base layer of DMEM containing 10% FBS and 0.6% agarose. The cells were re-fed every 4 days by the addition of 1.0 mL DMEM containing 10% FBS, 0.3% agarose, and supplemented with nothing, 1.0 μg/mL doxycycline, and/or exosomes isolated from the indicated cells. Two weeks later, the percentage of colonies that formed for each condition were determined with at least 200 cells being counted per sample.

2.3. Cell survival assay using trypan blue

Cells grown in six-well plates to 50–70% confluency were washed 5 times with 2 mL of sterile PBS before being maintained in serum-free medium supplemented with nothing, or the indicated combinations of 5% or 0.5% FBS, 1.0 μM SCH772984, 6.3 μM chloroquine, 0.30 μM paclitaxel, and 5 μg (normalized based on protein mass) of exosomes isolated from various cell types that were maintained under different culturing conditions. After 36 hours, the attached and floating cells were collected, treated with a 1:1 ratio of 0.04% trypan blue (Gibco), and counted using a TC-20 automated cell counter (Bio-Rad) to determine the percentage of viable cells. In many cases, cell survival was plotted as the percent increase in viability, compared to the percent of cell death determined for control cells without any treatment.

2.4. Cell viability assay using CCK-8

Cells (3000 total) were plated in each well of a 96-well plate and treated with medium containing 0.5% FBS, supplemented without or with 0.10 μM YM155. Forty-eight hours later, the cells were assayed according to the manufacturer’s instructions (Dojindo). Briefly, the cells were treated with 10 μL of the CCK-8 reagent and incubated at 37°C for 1 hour. The 96-well plate was then analyzed using the SPARK^® Multimode Microplate Reader (Tecan) at an absorbance of 450 nm.

2.5. Isolating exosomes from serum

The human serum samples were obtained from the Tissue Procurement Facility at the University of North Carolina. One mL aliquots of serum obtained from non-PDAC, and PDAC, patients were centrifuged at 300 × g for 10 minutes to remove cells and debris. The supernatant was moved to a new tube and centrifuged again at 120,000 × g for 4 hours to pellet exosomes. The pellet was resuspended in 3 mL of phosphate buffered saline (PBS) and centrifuged at 120,000 × g for another 4 hours. The supernatant was removed, and the pellets were lysed using lysis buffer (25 mM Tris, 100 mM NaCl, 1 mM EDTA, 1 mM DTT, 1 mM NaVO₄, 1 mM β-glycerol phosphate, 1% Triton X-100, 1 μg/mL aprotinin, and 1 μg/mL leupeptin).

2.6. Isolating exosomes from conditioned medium

Cells grown in 15 cm plates to a confluency of 80% were treated as indicated. The cells were then washed 2 times with 15 mL of PBS before being maintained in serum-free medium for 14 hours. The conditioned medium was collected, centrifuged at 300 × g to remove cells and debris, and subjected to filtration using a PVDF filter with 0.22 μm pore size (Millipore). The filtrate was ultracentrifuged at 120,000 × g for 4 hours, and the resulting exosome pellet was resuspended using lysis buffer or sterile PBS.

2.7. Immunofluorescence

Cells grown on glass coverslips and treated as indicated were fixed using 3.7% formaldehyde (Sigma), permeabilized using PBS containing 0.1% Triton X-100 (Sigma), and blocked using PBS containing 10% bovine serum albumin (BSA). The cells were incubated with a HA antibody (1:400, Cell Signaling Technology, #3724) for 90 minutes, followed by incubation with Alexa-568-conjugated anti-rabbit IgG (1:400, Thermo Fisher, #A-11011) for 90 minutes. The coverslips were washed several times with PBS and mounted on a slide with VECTASHIELD fluorescence mounting medium containing DAPI (Vector Laboratories) to label the nuclei. Images of the cells were acquired using a Zeiss Axioskop 40 microscope and a 63x objective, and processed using ImageJ.

2.8. Fluorescence microscopy of cells treated with exosomes labeled with FM 1-43FX

Cells grown in 15 cm plates to a confluency of 80% were washed 2 times with 15 mL of PBS and maintained in serum-free medium for 14 hours. The conditioned medium was collected, centrifuged at 300 × g to remove cells and debris, and treated with 5 μg/mL FM 1-43FX (Invitrogen) for 10 minutes. The exosomes were then isolated as described above, resuspended in sterile PBS, and the different samples were normalized based on protein mass. Primary MEFs (pMEFs) grown on coverslips were treated without or with the FM 1-43FX-labeled exosomes for 30 minutes at 37°C and 5% CO₂. The cells were washed 3 times with PBS, fixed with 3.7% formaldehyde, and mounted on a glass slide using VECTASHIELD fluorescence mounting medium containing DAPI. Images of the cells were acquired using a Zeiss Axioskop 40 microscope and a 63x objective, and processed using ImageJ.

2.9. Western blot analysis

The protein concentration of each lysate was determined using the Bradford Assay (Bio-Rad). The lysates were normalized based on protein concentration, resolved on 4–20% gradient SDS-PAGE gels (Invitrogen), and the proteins were transferred to PVDF membranes (Thermo Fisher). The membranes were blocked using 5% BSA dissolved in TBST (19 mM Tris, 2.7 mM KCl, 137 mM NaCl, and 0.5 % Tween-20), and then were incubated overnight at 4°C with one of the following antibodies (all from Cell Signaling Technology) diluted 1:1000 in TBST; HA (#3724), β-actin (#3700), Flotillin-2 (#3436), HSP90 (#4877), IκBα (#4812), CD81 (#10037), Survivin (#2080), GAPDH (#5174), RAS (#3965), V5 (#13202), Vinculin (#13901), Caspase 3 (#9662), AKT (#9272), phospho AKT (T308, #9275), ERK 1/2 (#4695), phospho ERK 1/2 (T202/Y204, #9101), STAT3 (#9136), phospho STAT3 (S727, #9134), phospho STAT3 (Y705, #9145), NFκB (#8242), phospho NFκB (S536, #3033). The next day, the membranes were incubated with anti-rabbit IgG, or anti-mouse IgG, HRP-linked antibodies (#7074 and #7076) for one hour, washed with TBST, and exposed to ECL (Perkin Elmer) reagent. The membrane was exposed to HyBlot CL^® Autoradiography Film (Thomas Scientific), and the film was developed using a Konica Minolta SRX-101A. Quantification of Western blots was performed using ImageJ. Regions of interest (ROI) of identical size were placed around each band of interest, as well as its corresponding loading control, and quantified. The band intensity for a given protein in the controls was normalized to the loading control and set at 1.0. The band intensity for the same protein in the experimental conditions was similar determined and presented as a ratio relative to the control.

2.10. Nanoparticle tracking analysis (NTA)

Conditioned medium collected from serum starved cells were centrifuged at 300 × g to remove cells and debris. The size and number of MVs and exosomes present in the partially clarified medium was determined by Nanosight NS300 (Malvern). For each sample, five 30-seconds videos were recorded, and the diffusion of the particles were analyzed to determine their size distribution and concentration.

2.11. Transmission electron microscopy

Copper grids coated with formvar/carbon were glow discharged in air. Conditioned medium (10 μL) was placed on the grid for 2 minutes, and the excess media was removed using filter paper. The grid was then stained twice, each time by applying 10 μL of 2 % uranyl acetate for 30 seconds and removing the excess solution with filter paper. After air-drying for 5 minutes, the sample was visualized using a Philips Morgagni 268 Transmission Electron Microscope.

2.12. Quantitative PCR

RNA isolated from cells maintained under the indicated conditions using the RNeasy Mini Kit (Qiagen) was reverse-transcribed to complementary DNA (cDNA) using Superscript III Reverse Transcriptase (Invitrogen) and oligo dT. Quantitative PCR (qPCR) was performed on the cDNA using iTaq SYBR Green Supermix (Bio-Rad) and the 7500 Real-Time PCR System (Applied Biosystem). The primers used to amplify mouse Survivin (AAGGAATTGGAAGGCTGGG, TTCTTGACAGTGAGGAAGGC) and actin (GTGACGTTGACATCCGTAAAGA, GCCGGACTCATCGTACTCC) were from Integrated DNA Technology.

2.13. Data analysis

All experiments were performed at least three independent times. Quantitative data are presented as means ± standard error. Statistical significance of the experiments was determined using Student’s t-tests; ****; p < 0.0001, ***; p < 0.001, **; p < 0.01, *; p < 0.05. Statistical analyses were performed using OriginPro.

3. Results

3.1. Exosomes isolated from the serum of PDAC patients contain Survivin

To begin to examine how oncogenic forms of RAS influence the production of EVs, we took advantage of the fact that a vast majority (~95%) of PDAC patients have a mutated form of KRAS. Thus, we examined whether PDAC patients released exosomes containing cargo that promotes cancer progression. One ml aliquots of serum obtained from five non-PDAC patients, and thirteen PDAC patients, were subjected to differential centrifugation to isolate exosomes (Fig. 1A). The exosome preparations were then lysed and analyzed by Western blot for the general EV marker Flotillin-2, which was detected in each of the samples (Fig. 1B, bottom panels). Survivin expression was increased in the exosomes from 8 out of 13 PDAC patients, whereas its level was significantly lower in the non-PDAC patients (Figs. 1B, top panels, and 1C).

Figure 1. Serum exosomes isolated from PDAC patients contain Survivin.

(A) The procedure used to isolate exosomes from serum samples.

(B) Western blots of exosomes isolated from serum samples taken from 5 non-PDAC patients and 13 PDAC patients. The blots were probed for Survivin (top panels) and Flotillin-2 (bottom panels).

(C) Quantification of Survivin expression levels detected in the samples shown in (B). The level of Survivin in each sample was normalized to Flotillin-2 expression. The data shown in (C) represents the mean ± standard error. Statistical significance was determined using a Student’s t-test; * p < 0.05.

3.2. Characterizing exosomes generated by MEFs expressing KRAS^G12D

To investigate the relationship between KRAS and the production of exosomes enriched in Survivin, mouse embryonic fibroblasts (MEFs) stably expressing an inducible, HA-tagged form of the KRAS^G12D mutant protein (HA-KRAS^G12D) were generated. These cells were initially maintained in medium containing doxycycline to suppress the expression of HA-KRAS^G12D. Upon removing doxycycline for 48 hours, HA-KRAS^G12D expression was then markedly increased, either when readout by Western blot analysis (Fig. 2A) or by immunofluorescence (Fig. 2B) using an HA antibody. Both control cells (plus doxycycline) and induced cells (upon removal of doxycycline) were subjected to anchorage-independent growth assays. Figs. 2C and 2D show that control MEFs lacking KRAS^G12D expression (−G12D) were unable to form colonies in soft agar, while ~40 % of MEFs induced to express KRAS^G12D (+G12D) were highly effective at promoting colony formation. Cell death assays were performed using trypan blue and showed that control MEFs (−G12D) were significantly more sensitive to serum-deprivation induced cell death, compared to cells expressing the oncogenic KRAS^G12D mutant (+G12D, Fig. 2E). This difference was not observed when the cells were grown in medium containing 10% serum (Supplementary Fig. S1).

Figure 2. MEFs expressing HA-KRAS^G12D have transformed characteristics.

(A) Western blot analysis using an HA antibody was performed on MEFs expressing an inducible form of HA-KRAS^G12D (whole cell lysates; WCL) that is under the control of doxycycline (Dox). Actin was used as the loading control.

(B) Immunofluorescence using an HA antibody was performed on the cells described in (A). The control MEFs (−G12D), and MEFs expressing KRAS^G12D (+G12D), were also stained with DAPI to label nuclei. The scale bars are 25 μm.

(C) Images of the soft agar assays performed on control MEFs (−G12D), and MEFs expressing HA-KRAS^G12D (+G12D). The scale bars are 50 μm.

(D) Quantification of the soft agar assays shown in (C).

(E) Cell survival assays using trypan blue were performed on serum starved control MEFs (−G12D), and MEFs expressing HA-KRAS^G12D (+G12D).

The data shown in (D) and (E) represent the mean ± standard error. All experiments were performed a minimum of three independent times, and statistical significance was determined using Student’s t-tests; *** p < 0.001 and **** p < 0.0001.

We next examined how oncogenic KRAS affected the formation and shedding of EVs. The conditioned medium collected from cultures of serum starved MEFs, that either lacked or inducibly expressed HA-KRAS^G12D, were subjected to a low speed centrifugation step to remove intact cells and debris, before being examined by nanoparticle tracking analysis (NTA) to determine the size and number of EVs present in each sample. This analysis indicated that control and KRAS^G12D expressing MEFs produced similar numbers and sizes of EVs, with the majority being less than 200 nm in diameter, suggesting they are exosomes (Fig. 3A, and Supplementary Fig. S2). The same samples were also visualized by electron microscopy and shown to contain many exosome-sized EVs exhibiting the expected cup-shaped morphology (Fig. 3B).

Figure 3. MEFs expressing activated KRAS generate exosomes enriched in Survivin.

(A) NTA was performed on the conditioned medium collected from an equivalent number of control MEFs (−G12D, red line), and MEFs expressing HA-KRAS^G12D (+G12D, black line), to determine the number and sizes of EVs in each sample. The solid lines represent means, and the shaded areas denote standard errors.

(B) Electron microscopy (EM) images of the conditioned medium samples described in (A). The scale bars are 50 nm.

(C) The procedure used to isolate exosomes from conditioned medium.

(D) Western blot analysis using antibodies that recognize the general EV markers Flotillin-2 and HSP90, the exosome marker CD81, and the cell-specific marker IκBα, was performed on the KRAS^G12D expressing MEFs (WCL) and the exosomes (EXO) and MVs these cells produce.

(E) Western blot analysis using HA and Survivin antibodies was performed on control MEFs (−G12D), and MEFs expressing HA-KRAS^G12D (+G12D). GAPDH was used as the loading control. The relative expression levels of Survivin were quantified and included below the blot.

(F) Western blot analysis using a Survivin antibody was performed on the exosomes isolated from the cells described in (E). HSP90 was used as the loading control. The relative expression levels of Survivin were quantified and included below the blot.

(G) Western blot analysis using V5 and HA antibodies were performed on control MEFs (−G12D), and MEFs expressing KRAS^G12D (+G12D), that had been transfected with V5-Survivin. GAPDH was used as the loading control.

(H) Western blot analysis using a V5 antibody was performed on the exosomes isolated from the cells described in (G). HSP90 was used as the loading control.

We then wanted to see whether the content of exosomes produced by MEFs changed upon the induction of KRAS^G12D expression, and, specifically, whether they became enriched in Survivin. To resolve exosome-sized vesicles from MVs (i.e., vesicles larger than 200 nm in diameter), the partially clarified conditioned medium from control MEFs, and MEFs inducibly expressing KRAS^G12D, were filtered using a 0.22 μm pore size PVDF membrane to retain MVs, while the flow-through or filtrate was subjected to centrifugation at 120,000 × g to pellet exosomes (Fig. 3C). The pellets isolated from both MEFs expressing KRAS^G12D and control cells were lysed and analyzed by Western blot (Fig. 3D). The cytosolic signaling protein IκBα was clearly detectable in the whole cell lysates (WCL), but was absent from both the MV and exosome fractions. The blot was also probed for the general EV markers Flotillin-2 and heat shock protein 90 (HSP90), as well as the exosome-specific marker CD81. Flotillin-2 and HSP90 were present in the both the MV and exosome preparations, while CD81 was highly enriched in the exosome fraction.

Control MEFs, MEFs inducibly expressing KRAS^G12D, as well as the exosomes produced by each of these cell types, were collected and evaluated for Survivin expression. While the control MEFs had barely detectable levels of Survivin, its expression was strongly up-regulated in MEFs following the induction of HA-KRAS^G12D, as determined both by Western blot analysis and quantitative PCR (qPCR) (Fig. 3E, middle panel, and Supplementary Fig. S3). Likewise, exosomes derived from MEFs expressing KRAS^G12D were enriched in Survivin (Fig. 3F). To better understand how KRAS results in the generation of exosomes containing Survivin, we ectopically expressed a V5-tagged form of the protein (i.e., V5-Survivin) in both control MEFs, and MEFs expressing KRAS^G12D (Fig. 3G). Essentially equivalent amounts of V5-Survivin were detected in the exosome preparations isolated from both control and KRAS^G12D expressing MEFs (Fig. 3H), suggesting that the enrichment of Survivin in exosomes derived from cells expressing KRAS^G12D is likely a direct consequence of the high expression of Survivin in KRAS-transformed cells.

3.3. Exosomes from MEFs expressing KRAS^G12D promote cell survival and transformation

We next wanted to determine whether exosomes isolated from control and KRAS^G12D-expressing MEFs can be transferred to target cells. Two complementary approaches were used to address this question. The first involved collecting the conditioned medium from control and KRAS^G12D-expressing cells, followed by incubating the medium with the fluorescent membrane dye FM 1-43FX. The medium was filtered to remove the larger MVs, and the flow-through was subjected to high speed centrifugation to pellet exosomes and remove unincorporated dye. The labelled exosomes were then resuspended in fresh cell culture medium and used to treat naïve primary MEFs (pMEFs) for 30 minutes, at which point the cells were washed extensively with PBS, fixed, and visualized by fluorescence microscopy. Very little fluorescence was detected in untreated pMEFs. However, cells treated with the labelled exosomes isolated from both control MEFs and MEFs expressing KRAS^G12Dexhibited a strong fluorescence signal (Fig. 4A), indicating that exosomes isolated from each of the cell types were similarly capable of being transferred to cells.

Figure 4. The exosomes from MEFs expressing activated KRAS can be transferred to other cells.

(A) Fluorescence images of pMEFs that had been left untreated (Control), or were treated with FM 1-43FX labeled exosomes isolated from control MEFs (−G12D Exosomes), and MEFs expressing HA-KRAS^G12D (+G12D Exosomes). The cells were also stained with DAPI to label nuclei. The scale bars are 25 μm.

(B) Western blot analysis using a V5 antibody was performed on MEFs expressing KRAS^G12D that had been mock transfected or transfected with V5-Survivin (top panels labelled WCLs of Donor +G12D MEFs). Two batches of exosomes were collected from these cells. Western blot analysis using a V5 antibody was performed on one batch of exosomes that had been lysed (middle panels labelled Exosomes of Donor +G12D MEFs), while the second batch was resuspended in PBS and then used to treat pMEFs for two hours, at which point the cells were lysed and blotted with a V5 antibody. Untreated pMEFs were used as a control (bottom panels labelled WCLs of recipient cells). GAPDH and HSP90 were used as loading controls. The experimental procedure is depicted to the right.

For the second approach, MEFs expressing KRAS^G12D were either mock transfected, or transfected with V5-Survivin (Fig. 4B, top panels). Two batches of exosomes were collected from the cells; one batch was analyzed by Western blot to further demonstrate the presence of V5-Survivin in the exosomal preparations from cells transfected with the epitope-tagged protein, while being absent in exosomes isolated from the mock transfectants (Fig. 4B, middle panels). The second batch of exosomes was resuspended in serum-free medium and used to treat pMEFs for 2 hours, at which point the cells were washed with PBS, lysed, and Western blotted with V5 antibody. Fig. 4B (bottom panels) shows that V5-Survivin expression was only detected in pMEFs treated with exosomes that contained the ectopically expressed, epitope-tagged protein, thus further demonstrating that exosomes and their associated cargo are capable of being transferred to cells.

Survivin is a member of the IAP family and has been shown to promote cell survival by inhibiting the activation of caspases 3 and 7, two important components of the apoptotic machinery¹⁹. Since exosomes derived from MEFs expressing an oncogenic KRAS mutant were highly enriched in Survivin, we examined whether these exosomes were capable of promoting cell survival. Depriving pMEFs of serum led to the cleavage and activation of caspase 3 (Fig. 5A, compare lanes 1 and 2), resulting in potent cell death (Fig. 5B). Treating serum starved pMEFs with exosomes isolated from control MEFs did not alter the levels of cleaved caspase 3 detected (Fig. 5A, compare lanes 2 and 3), and consequently, this treatment only modestly enhanced cell survival, relative to serum starved cells (Fig. 5C). However, pMEFs treated with the exosomes from MEFs expressing KRAS^G12D had a lower amount of cleaved caspase 3 (Fig. 5A, compare lanes 3 and 4); moreover, the extent of serum starvation-induced death in these cells was also strongly inhibited (Fig. 5C).

Figure 5. The exosomes produced by MEFs expressing HA-KRAS^G12D promote cell survival.

(A) Western blot analysis using a caspase 3 antibody was performed on pMEFs that had been maintained in serum-free medium supplemented with nothing, exosomes isolated from either control MEFs (−G12D) or MEFs expressing KRAS^G12D (+G12D), or 5% fetal bovine serum (FBS) for 24 hours. Vinculin was used as the loading control. The relative expression levels of cleaved caspase 3 were quantified and included below the blot.

(B) Cell survival assays using trypan blue were performed on pMEFs that had been maintained for 36 hours in medium containing 5% FBS, or lacking serum (Serum starved).

(C) Cell survival assays using trypan blue were performed on serum starved pMEFs that had been treated with exosomes isolated from control MEFs (−G12D), or MEFs expressing KRAS^G12D (+G12D). The findings shown represent the increases in cell survival determined for each condition, compared to untreated serum starved pMEFs.

(D) Western blot analysis using a Survivin antibody was performed on MEFs expressing KRAS^G12D that had been treated with negative control siRNA (NC) or Survivin siRNA. GAPDH was used as the loading control. The relative expression levels of Survivin were quantified and included below the blot.

(E) NTA analysis was performed on the conditioned medium collected from an equivalent number of the cells described in (D). The solid lines represent means, and the shaded areas denote standard errors.

(F) Western blot analysis using a Survivin antibody was performed on the exosomes derived from the cells described in (D). HSP90 was used as the loading control. The relative expression levels of Survivin were quantified and included below the blot.

(G) Cell survival assays using trypan blue were performed on serum starved pMEFs that had been treated with the exosomes isolated from HA-KRAS^G12D expressing MEFs transfected with negative control siRNA (NC siRNA) or Survivin siRNA. The findings shown represent the increases in cell survival determined for each condition, compared to untreated serum starved pMEFs.

The data shown in (B), (C), (E), and (G) represent the mean ± standard error. All experiments were performed a minimum of three independent times, and statistical significance was determined using Student’s t-tests; * p < 0.05, ** p < 0.01, and **** p < 0.0001.

The exosomes derived from MEFs expressing activated KRAS^G12D were also assayed for their ability to promote cellular transformation, as read-out using colony formation in soft agar. Treatment of MEFs with exosomes collected from control cells failed to increase colony formation, whereas treatment with exosomes derived from KRAS^G12D expressing MEFs was quite effective in promoting growth in soft agar (Supplementary Figure S4A and S4B).

3.4. Exosomes depleted of Survivin lose their ability to promote cell survival

To determine whether Survivin played an important role in the ability of exosomes shed by KRAS^G12D-expressing MEFs to mediate their effects on survival, we knocked-down its expression using siRNA (Fig. 5D, middle panel). While knocking-down Survivin caused only a slight change in the number of exosomes shed by the MEFs expressing KRAS^G12D (Fig. 5E, and Supplementary Fig. S5), there was a significant reduction in the amount of Survivin present within these exosomes (Fig. 5F, top panel). After normalizing the amount of exosomes collected for each condition based on protein mass, we assayed their ability to promote the survival of serum starved pMEFs. Unlike exosomes isolated from control MEFs expressing KRAS^G12D (i.e., treated with negative control siRNA; NC siRNA), which promoted cell survival, treating pMEFs with the exosomes isolated from cells lacking Survivin was ineffective (Fig. 5G).

We next determined whether PDAC cells expressing KRAS^G12D are also enriched in Survivin. We examined mouse AK192 PDAC cells that were engineered to express an inducible form of KRAS^G12D when cultured in medium containing doxycycline (Fig. 6A)^13,23. NTA performed on the conditioned medium collected from an equal number of AK192 cells, lacking or expressing KRAS^G12D, showed that the cells generated comparable numbers of exosome sized EVs (Fig. 6B, and Supplementary Fig. S6). However, the levels of Survivin detected in AK192 cells, as well as in the exosomes generated by the cells, were directly correlated with the expression of KRAS^G12D. Specifically, Survivin was clearly detectable in AK192 cells expressing KRAS^G12D (Fig. 6C, middle panel, and Supplementary Fig. S7), and in exosomes isolated from these cells (Fig. 6D, top panel), whereas, when KRAS^G12D expression was suppressed by the removal of doxycycline, the levels of Survivin detected in the whole cell lysates were greatly reduced (Fig. 6C, middle panel). Similarly, exosomes produced by these cells also contained reduced levels of Survivin (Fig. 6D, top panel).

Figure 6. KRAS-dependent PDAC cells also produce exosomes enriched in Survivin.

(A) Western blot analysis using a RAS antibody was performed on AK192 cells engineered to express an inducible form of KRAS^G12D that is under the control of doxycycline (Dox). GAPDH was used as the loading control.

(B) NTA was performed on the conditioned medium collected from an equivalent number of AK192 cells not expressing (−G12D, red line), or expressing (+G12D, black line), KRAS^G12D. The solid lines represent means, and the shaded areas denote standard errors.

(C) Western blot analysis using RAS and Survivin antibodies was performed on AK192 cells (WCL) not expressing (−G12D), or expressing (+G12D), KRAS^G12D. GAPDH was used as the loading control. The relative expression levels of Survivin were quantified and included below the blot.

(D) Western blot analysis using a Survivin antibody was performed on the exosomes isolated from the cells described in (C). HSP90 was used as the loading control. The relative expression levels of Survivin were quantified and included below the blot.

(E) Cell survival assays using trypan blue were performed on serum starved pMEFs that had been treated with exosomes isolated from AK192 cells not expressing (−G12D), or expressing (+G12D), KRAS^G12D. The findings shown represent the increases in cell survival determined for each condition, compared to untreated serum starved pMEFs.

(F) Western blot analysis using RAS and Survivin antibodies were performed on AK192 cells expressing KRAS^G12D that had been treated with negative control siRNA (NC) or Survivin siRNA. GAPDH was used as the loading control. The relative expression levels of Survivin were quantified and included below the blot.

(G) Western blot analysis using a Survivin antibody was performed on the exosomes isolated from the cells described in (F). HSP90 was used as the loading control. The relative expression levels of Survivin were quantified and included below the blot.

(H) NTA was performed on the conditioned medium collected from an equivalent number of AK192 cells expressing KRAS^G12D that had been treated with negative control siRNA (NC siRNA) or Survivin siRNA. The solid lines represent means, and the shaded areas denote standard errors.

(I) Cell survival assays using trypan blue were performed on serum starved pMEFs treated with the exosomes derived from KRAS^G12D expressing AK192 cells that had been treated with negative control siRNA (NC siRNA) or Survivin siRNA. The findings shown represent the increases in cell survival determined for each condition, compared to untreated serum starved pMEFs.

The data shown in (B), (E), (H), and (I) represent the mean ± standard error. All experiments were performed a minimum of three independent times, and statistical significance was determined using Student’s t-tests; * p < 0.05.

We then examined the ability of exosomes produced by AK192 cells expressing KRAS^G12D to promote cell survival. Serum-starved pMEFs were left untreated, or were treated with exosomes isolated from control cells and cells induced to express KRAS^G12D for 36 hours, at which point their viability was determined. There was nearly a 40% increase in the survival of serum starved pMEFs when treated with exosomes generated by KRAS^G12D expressing AK192 cells, compared to the untreated control cells (Fig. 6E, bar labelled +G12D). However, exosomes derived from AK192 cells that were not induced to express KRAS^G12D only marginally increased cell survival (bar labelled -G12D).

To determine whether the Survivin associated with exosomes from AK192 cells expressing KRAS^G12D was responsible for mediating a survival benefit, the protein was knocked-down using siRNA. This effectively depleted both the cellular and exosomal expression levels of Survivin by more than 85% (Figs. 6F and 6G). The exosomes were then assayed for their cell survival-promoting capabilities. Knocking-down Survivin in AK192 cells expressing KRAS^G12D did not significantly change the number of exosome sized vesicles released by the cells (Fig. 6H, and Supplementary Fig. S8). After isolating the exosomes produced by the cells, and normalizing them based on protein mass, they were used to treat serum starved pMEFs. Here again, we found that exosomes derived from control AK192 cells expressing KRAS^G12D strongly promoted cell survival (Fig. 6I, bar labelled NC siRNA). However, when exosomes depleted of Survivin were examined, their ability to promote survival under serum-deprivation conditions was reduced to background levels, i.e., the amount of cell death observed when pMEFs were serum starved (Fig. 6I, bar labelled Survivin siRNA).

3.5. Exosomes enriched in Survivin promote drug resistance in human PDAC cells.

Human PDAC patients have an extremely poor prognosis, in part, because they often fail to respond to therapies⁷. Thus, we were interested in seeing whether exosomes produced by PDAC cell lines might play a role in promoting drug resistance. Two widely-studied human KRAS-mutant PDAC cell lines, MIA PaCa-2 and PANC-1, were examined. Both cell types also express Survivin (Figs. 7A and 7B, top panels), and the exosomes that they shed exhibit relatively high levels of this protein (Figs. 7C and 7D, top panels). It is worth noting that exosomes derived from other cancer cell types known to have mutations in KRAS, including MDA-MB-231 breast cancer cells, as well as H23 and A549 lung cancer cells, all contain Survivin^25,26 (Supplementary Fig. S9A and S9B), suggesting that the production of exosomes enriched with this protein may be common to KRAS-driven cancer cells. We then asked whether exosomes shed by these PDAC lines might influence the effectiveness of paclitaxel, or a therapy currently under clinical evaluation in PDAC; specifically, a treatment protocol that consists of a combination of an ERK or MEK inhibitor with hydroxychloroquine. Consistent with previous findings¹³, we found that treating human BxPC-3 PDAC cells (BRAF mutant) with 0.30 μM of paclitaxel, or the combination of 1 μM of the ERK inhibitor SCH772984 (SCH) and 6.3 μM chloroquine (CQ), induced cell death (Figs. 7E and 7F). However, when BxPC-3 cells were treated with exosomes isolated from either MIA PaCa-2 (Figs. 7G and 7H, bars labelled DMSO) or PANC-1 cells (Figs. 7I and 7J, bars labelled DMSO), the effectiveness of paclitaxel, or the drug combination was greatly compromised. We also determined that BxPC-3 cells were capable of generating exosomes containing Survivin, suggesting that they can potentially promote autonomous drug resistance (Supplementary Fig. S10A).

Figure 7. Exosomal Survivin induces drug resistance in human PDAC cells.

(A and B) Western blot analysis using a Survivin antibody was performed on (A) MIA PaCa-2 and (B) PANC-1 cells that had been treated with DMSO, or 0.25 μM YM155. GAPDH was used as the loading control. The relative expression levels of Survivin were quantified and included below this blot.

(C and D) Western blot analysis using a Survivin antibody was performed on the exosomes isolated from the cells described in (A and B). HSP90 was used as the loading control. The relative expression levels of Survivin were quantified and included below this blot.

(E and F) Cell survival assays using trypan blue were performed on BxPC-3 cells that had been treated for 36 hours with (E) DMSO and 0.30 μM paclitaxel (PTX), or (F) DMSO and the combination therapy of 1.0 μM SCH772984 and 6.3 μM chloroquine (SCH+CQ).

(G and H) Cell survival assays using trypan blue were performed on BxPC-3 cells that had been treated for 36 hours with (G) DMSO and 0.30 μM paclitaxel, or (H) DMSO and the combination therapy of 1.0 μM SCH772984 and 6.3 μM chloroquine (SCH+CQ), and exosomes isolated from MIA PaCa-2. The plots represent the increases in cell survival determined for each condition, compared to BxPC-3 treated with only 0.30 μM paclitaxel, or the combination therapy.

(I and J) Cell survival assays using trypan blue were performed on BxPC-3 cells that had been treated for 36 hours with (I) DMSO and 0.30 μM paclitaxel, or (J) DMSO and the combination therapy of 1.0 μM SCH772984 and 6.3 μM chloroquine (SCH+CQ), and exosomes isolated from PANC-1. The plots represent the increases in cell survival determined for each condition, compared to BxPC-3 treated with only 0.30 μM paclitaxel, or the combination therapy.

The data shown in (E)-(J) represent the mean ± standard error. All experiments were performed a minimum of three independent times, and statistical significance was determined using Student’s t-tests; * p<0.05, *** p < 0.001 and **** p < 0.0001.

We then treated MIA PaCa-2 and PANC-1 cells with YM155, a small molecule inhibitor that blocks the transcription of the human BIRC5 gene which encodes Survivin^27,28. This treatment potently reduced the expression of Survivin within 24 hours (Figs. 7A and 7B, lanes labelled YM155), and caused cell death to occur after 48 hours of treatment (Supplementary Fig. S10B). Exosomes produced by PDAC cells treated with YM155 also showed corresponding reductions in Survivin levels (Figs. 7C and 7D, lanes labelled YM155). When exosomes isolated from MIA PaCa-2 and PANC-1 cells treated with YM155 were added to BxPC-3 cells, they were unable to promote resistance to the either drug treatment (Figs. 7 G–J, bars labelled YM155).

We further examined how treating PANC-1 and MIA PaCa-2 cells with the combination of the ERK inhibitor SCH772984 and chloroquine would affect their ability to generate exosomes containing Survivin. Extended treatment of these cells with this drug combination tended to cause a reduction in Survivin expression in the cells, as well as in the exosomes that these cells generated (Supplementary Figs. S11A–D). This effect appears to be rather specific, as the drug combination did not alter the activation of several other proteins known to promote cell survival (supplementary Fig. S11E), and may explain why it is being considered as a treatment for PDAC¹³.

4. Discussion

The findings described in this study show that KRAS-transformed MEFs, as well as PDAC cells that express oncogenic KRAS mutants, generate exosomes containing the protein Survivin, which are capable of conferring neighboring cells with a significant survival benefit and resistance to a promising drug therapy in clinical trials (Fig. 8). They also suggest that the ability of exosomes to mediate intercellular communication within a developing tumor could have a significant impact on disease progression. This may be especially the case, given that primary tumors are subjected to a number of stresses including nutrient deprivation and hypoxia, which have the potential to cause their cells to undergo apoptosis and for the tissue to become necrotic^29–32. KRAS mutations are known to occur at the initial stages of tumorigenesis in pancreatic cancer^1,7,33,34. Therefore, the ability of pancreatic tumor cells to shed exosomes that contain Survivin and provide a significant survival advantage may represent a key step in enabling the cancer to progress to more advanced and aggressive stages. Our findings also raise two important questions. First, do PDAC cells that express different mutant forms of KRAS produce exosomes enriched with Survivin, and second, do other forms of cancer driven by KRAS use a similar mechanism of drug resistance? Interestingly, breast and lung cancer cell lines whose transformed phenotypes are dependent on KRas also generate exosomes containing Survivin (Supplementary Fig. S9).

Figure 8. Diagram showing how KRAS-dependent PDAC cells produce exosomes that promote survival.

The expression of mutant forms of KRAS (KRAS^mt) in PDAC cells strongly increases the production of exosomes enriched in Survivin. These exosomes, and the Survivin they contain, can be detected in the serum isolated from PDAC patients (left), transferred to fibroblasts (middle) that reside within the tumor microenvironment, as well as other PDAC cells (right) and promote their survival and drug resistance.

Exosomes that contain Survivin can provide an additional advantage to tumor progression through their ability to confer drug resistance²⁴. Recently, a promising therapeutic strategy was suggested based on studies showing that by combining inhibitors that target protein kinases activated downstream of KRAS, namely MEK or ERK, with compounds such as chloroquine which block autophagy, it was possible to significantly inhibit the growth of pancreatic cancer cells both in cell culture and in mice^13,35. However, we found that exosomes containing Survivin were able to significantly blunt the effectiveness of this combination treatment. These findings then raised the question whether blocking Survivin expression in KRAS-dependent cancer cells might offer a novel strategy for inhibiting their growth and survival. Indeed, we found that treating pancreatic cancer cells with the compound, YM155, which inhibits Survivin expression, was able to effectively diminish the ability of their shed exosomes to promote drug resistance. Although YM155 has not been shown to provide significant clinical benefits in some phase 2 or 3 trials in lung and prostate cancers, nor as yet in melanoma, this compound has passed phase 1 trials with acceptable safety and patient tolerance^28,36–39. Thus, it would be interesting to see if combining YM155 with ERK/MEK inhibitors and chloroquine might provide a safe option and a potentially new therapeutic strategy for pancreatic cancer.

Our findings raise yet an additional approach worth considering; specifically, blocking the biogenesis of exosomes in aggressive cancer cells, or targeting them in a manner that inhibits their ability to transfer Survivin to surrounding cells and the survival advantages that it confers. Several compounds, including GW4869, PitStop 2, as well as Dynasore, have been shown to inhibit exosome biogenesis or uptake^40–43. Moreover, a recent report suggests that the biogenesis of a unique class of exosomes produced specifically by cancer cells bearing an oncogenic KRAS mutation is dependent on the activity of the small GTPase RAB13⁴⁴. This then raises the possibility that the use of small molecules which target RAB13 and block exosome biogenesis in pancreatic cancer cells could provide a synergistic benefit when combined with YM155 to reduce Survivin expression, or with the combination of MEK/ERK and autophagy inhibitors.

Finally, the presence of Survivin in exosomes shed by pancreatic cancer cells makes it potentially attractive as a diagnostic indicator. Thus far, a number of proteins have been examined as possible clinical markers of pancreatic cancer, including cancer antigen (CA) 19-9, CA 125, and cancer embryonic antigen (CEA)^7,45. However, while CA 19-9 level can be used to monitor disease progression in symptomatic patients, none of these proteins has yet to be shown to be a reliable diagnostic marker due to issues both with specificity and sensitivity^7,46,47. Our findings that Survivin can be clearly detected in exosomes isolated from pancreatic patient serum samples is interesting because a number of studies have shown increasing levels of this protein in the blood of pancreatic cancer patients^20,21,48. Moreover, because Survivin is not expressed in normal adult tissues, it is an especially attractive candidate for a diagnostic indicator of cancer progression^19,48,49. However, a good deal of work still needs to be done in this regard, as Survivin was detected in the exosomes derived from 8 out of 13 PDAC patients, which is fewer than would be expected given the prevalence of oncogenic KRAS as a driver of this disease. Therefore, future studies will be aimed at assessing the ability to reliably detect exosomal Survivin in the plasma obtained from mouse models for pancreatic cancer, as well as from patient samples, in order to determine whether the sensitivity of detection is sufficient to consider using exosomes containing this protein as a new type of diagnostic marker.

Supplementary Material

1

Supplementary Figure S1 Cell survival assays using trypan blue were performed on control MEFs (−G12D), and MEFs expressing KRAS^G12D (+G12D) grown in medium containing 10% serum. The data shown represents the mean ± standard error. The experiment was performed a minimum of three independent times, and statistical significance was determined using a Student’s t-test; n.s. not significant.

Supplementary Figure S2 Quantification of the exosomes (i.e., the EVs smaller than 200 nm) produced by the MEFs shown in Fig. 3A. The data shown represents the mean ± standard error. The experiment was performed a minimum of three independent times, and statistical significance was determined using a Student’s t-test; n.s. not significant.

Supplementary Figure S3 qPCR was performed to determine the levels of Survivin transcript in control MEFs (−G12D), or MEFs expressing KRAS^G12D (+G12D). The data shown represents the mean ± standard error. The experiment was performed three independent times, and statistical significance was determined using a Student’s t-test; **** p < 0.0001.

Supplementary Figure S4 (A) Images of the soft agar assays performed on control MEF treated with exosomes isolated from either control MEFs (−G12D Exo), or MEFs expressing KRAS^G12D (+G12D Exo). The scale bars are 50 μm.

(B) Quantification of the soft agar assays shown in (A). Untreated MEFs (No Exo) was used as the control.

The data shown in (B) represents the mean ± standard error. The experiment was performed a minimum of three independent times, and statistical significance was determined using Student’s t-tests; n.s., not significant, ** p < 0.01, **** p < 0.0001.

Supplementary Figure S5 Quantification of the exosomes (i.e., EV smaller than 200 nm) produced by MEFs induced to express KRAS^G12D transfected with negative control siRNA (NC siRNA) or siRNA targeting Survivin (Survivin siRNA) shown in Fig. 5E. The data shown represents the mean ± standard error, and the experiment was performed a minimum of three independent times, and statistical significance was determined using a Student’s t-test; n.s. not significant.

Supplementary Figure S6 Quantification of the exosomes (i.e., EVs smaller than 200 nm) produced by the AK192 cells shown in Fig. 6B. The data shown represents the mean ± standard error. The experiment was performed a minimum of three independent times, and statistical significance was determined using a Student’s t-test; n.s. not significant.

Supplementary Figure S7 qPCR was performed to determine the levels of Survivin transcript in AK192 cells expressing (+G12D), or not expressing (−G12D), KRAS^G12D. The data shown represents the mean ± standard error. The experiment was performed three independent times, and statistical significance was determined using a Student’s t-test; *** p < 0.001.

Supplementary Figure S8 Quantification of the exosomes (i.e., EVs smaller than 200 nm) produced by the AK192 cells induced to express KRAS^G12D, and transfected with negative control siRNA (NC siRNA) or siRNA targeting Survivin (Survivin siRNA) shown in Fig. 6H. The data represents the mean ± standard error. The experiment was performed a minimum of three independent times, and statistical significance was determined using a Student’s t-test; n.s. not significant.

Supplementary Figure S9 (A) Western blot analysis using a Survivin antibody was performed on MDA-MB-231 (231), A549, and H23 cells. GAPDH was used as the loading control.

(B) Western blot analysis using a Survivin antibody was performed on the exosomes generated by the cells described in (A). HSP90 was used as the loading control.

Supplementary Figure S10 (A) Western blot analysis using a Survivin antibody was performed on exosomes generated by PANC-1, MIA PaCa-2, and BxPC-3 cells. HSP90 was used as the loading control.

(B) Cell viability assays using CCK-8 was performed on PANC-1, MIA PaCa-2, and BxPC-3 treated with medium containing 0.5% FBS and supplemented with either DMSO, or 0.10 μ M YM155 for 48 hours.

The data shown in (B) represents the mean ± standard error. The experiment was performed a minimum of three independent times, and statistical significance was determined using Student’s t-tests; **** <0.0001

Supplementary Figure S11 (A and B) Western blot analysis using a Survivin antibody was performed on (A) MIA PaCa-2 cells and (B) PANC-1 cells that had been treated with DMSO, or the combination of 1.0 μM SCH772984 and 6.3 μM chloroquine (SCH+CQ) for 60 hours. GAPDH was used as the loading control. The relative expression levels of Survivin were quantified and included below this blot.

(C and D) Western blot analysis using a Survivin antibody was performed on the exosomes produced by (C) MIA PaCa-2 cells and (D) PANC-1 cells that had been treated with DMSO, or the combination of 1.0 μM SCH772984 and 6.3 μM chloroquine (SCH+CQ) for 60 hours. HSP90 was used as the loading control. The relative expression levels of Survivin were quantified and included below this blot.

(E) Western blot analysis using ERK 1/2, phopho ERK1/2 (T202/Y204), AKT, phospho AKT (T308), NFκB, phospho NFκB (S536), STAT3, phopho STAT3 (S727 or Y705), and Survivin antibodies was performed on MIA PaCa-2 and PANC-1 cells treated with DMSO, or the combination of 1.0 μM SCH772984 and 6.3 μM chloroquine (SCH+CQ). GAPDH was used as the loading control.

Highlight Statement.

• KRAS is mutated in 95% of human PDAC patients.

• Survivin can be detected in the exosomes isolated from PDAC patients.

• KRAS dependent cancer cells generate exosomes enriched in Survivin.

• Exosomes containing Survivin promote cell survival and drug resistance.

Abbreviations

EV

extracellular vesicle

MVB

multi-vesicular body

MV

microvesicle

PDAC

pancreatic ductal adenocarcinoma

PDX

patient derived xenograft

IAP

inhibitor of apoptosis protein

MEF

mouse embryonic fibroblast

NTA

nanoparticle tracking analysis

WCL

whole cell lysate